High Court quashes aquaculture and foreshore licences for Minister's failure to consider ALAB's 2021 decision.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

THE HIGH COURT PLANNING & ENVIRONMENT JUDICIAL REVIEW IN THE MATTER OF SECTION 73 OF THE FISHERIES (AMENDMENT) ACT 1997 (AS AMENDED)

2021/823JR BETWEEN SALMON WATCH IRELAND CLG APPLICANT AND THE AQUACULTURE LICENCES APPEALS BOARD AND THE MINISTER FOR AGRICULTURE, FOOD AND THE MARINE, THE MINISTER FOR ENVIRONMENT, CLIMATE AND COMMUNICATIONS, IRELAND AND THE ATTORNEY GENERAL RESPONDENTS AND BRADÁN FANAD TEO TRADING AS MARINE HARVEST IRELAND COMHLUCHT IASCAIREACHTA FANAD TEORANTA TRADING AS MOWI IRELAND SAVE BANTRY BAY, CARE OF ALEC O’DONOVAN, BREDA O’SULLIVAN, JOHN BRENDAN O’KEEFFE DENIS O’SHEA, KIERAN O’SHEA AND JASON O’SHEA BANTRY SALMON AND TROUT ANGLERS’ ASSOCIATION CHRIS HARRINGTON, VINCENT O’SULLIVAN, PETER MURPHY AND CHRIS FORKER GALWAY BAY AGAINST SALMON CAGES JOHN HUNT FRIENDS OF THE IRISH ENVIRONMENT INLAND FISHERIES IRELAND FEDERATION OF IRISH SALMON AND SEA TROUT ANGLERS NOTICE PARTIES AND 2021/828JR BETWEEN:

INLAND FISHERIES IRELAND APPLICANT AND AQUACULTURE LICENCES APPEALS BOARD THE MINISTER FOR AGRICULTURE, FOOD AND THE MARINE RESPONDENTS AND BRADÁN FANAD TEORANTA TRADING AS MARINE HARVEST IRELAND, SAVE BANTRY BAY, SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

THE RESIDENTS OF ROOSK, ADRIGOLE, JOHN BRENDAN O’KEEFFE, DENIS O’SHEA, KIERAN O’SHEA AND JASON O’SHEA, BANTRY SALMON AND TROUT ANGLERS’ ASSOCIATION,

C. HARRINGTON, V. O’SULLIVAN, P. MURPHY, C. FORKER, COOMHOLA SALMON & TROUT ANGLERS’ ASSOCIATION, GALWAY BAY AGAINST SALMON CAGES, SALMON WATCH IRELAND, JOHN HUNT, FRIENDS OF THE IRISH ENVIRONMENT, FEDERATION OF IRISH SALMON AND SEA TROUT ANGLERS NOTICE PARTIES AND 2021/831 JR BETWEEN:

PETER SWEETMAN FEDERATION OF IRISH SALMON AND SEA TROUT ANGLERS JOHN BRENDAN O'KEEFFE APPLICANTS AND AQUACULTURE LICENCE APPEALS BOARD MINISTER FOR AGRICULTURE FOOD AND THE MARINE IRELAND AND THE ATTORNEY GENERAL RESPONDENTS AND BRADAN FANAD TEORANTA T/A MARINE HARVEST IRELAND COMHLUCHT IASCAIREACHTA FANAD TEORANTA T/A MOWI IRELAND INLAND FISHERIES IRELAND NOTICE PARTIES

JUDGMENT OF MR JUSTICE HOLLAND DELIVERED 12 JULY 2024

Contents

JUDGMENT OF MR JUSTICE HOLLAND DELIVERED 12 JULY 2024 ........................................................................2

PART 1 – INTRODUCTORY AND GENERAL MATTERS AND CORE GROUNDS ........................................................9

INTRODUCTION & BRIEF CHRONOLOGY ..............................................................................................................9

The Site, the Salmon Farm, the Parties, the Impugned Aquaculture Licence & Foreshore Licence ................9

Judicial Review is not an Appeal on the Merits ............................................................................................. 12

ALAB............................................................................................................................................................... 14

The Geography & WFD Waterbodies ............................................................................................................ 15

The Proceedings & Structure of Judgment – a brief note ............................................................................. 16

Brief Chronology ............................................................................................................................................ 17

Effects on Wild Salmonids and FwPM – EIA Only & Licensing Despite Significant Effects............................ 25

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Some Practical Notes ..................................................................................................................................... 28

LEGISLATION ...................................................................................................................................................... 28

Fisheries (Amendment) Act 1997 .................................................................................................................. 29

Foreshore Act 1933 ....................................................................................................................................... 35

EIA Legislation – General, Aquaculture & Foreshore Licensing..................................................................... 36

EIA & Foreshore Licences .......................................................................................................................... 38

EIA & Aquaculture Licences – 2014 Directive – Transitional Provisions & their Transposition ................ 40

Inland Fisheries Act 2010............................................................................................................................... 41

Habitats Regulations 2011............................................................................................................................. 42

SWI PROCEEDINGS – OUTLINE OF PLEADINGS. ................................................................................................. 45

SWI – Reliefs Sought ...................................................................................................................................... 45

SWI Grounds – Domestic ............................................................................................................................... 46

SWI Grounds – EU Law .................................................................................................................................. 47

SWI Grounds – Validity .................................................................................................................................. 49

IFI PROCEEDINGS – OUTLINE OF PLEADINGS. ................................................................................................... 50

IFI – Reliefs Sought ........................................................................................................................................ 50

IFI Grounds – Domestic ................................................................................................................................. 50

IFI Core Grounds – EU Law ............................................................................................................................ 52

SWEETMAN PROCEEDINGS – OUTLINE OF PLEADINGS..................................................................................... 52

PS – Reliefs Sought ........................................................................................................................................ 52

PS Core Grounds – Domestic ......................................................................................................................... 53

PS Core Grounds – EU Law ............................................................................................................................ 54

PS Core Grounds 10 – 13 – Habitats Directive .............................................................................................. 57

PS Core Ground – WFD .................................................................................................................................. 57

PS Core Grounds – EIA Directive ................................................................................................................... 58

PS Core Ground – Transposition / Invalidity.................................................................................................. 60

PLEADINGS IN JUDICIAL REVIEW ....................................................................................................................... 61

CONDITIONS IN LICENCES – VALIDITY & INTERPRETATION – GENERAL NOTE.................................................. 62

Conditions – Certainty & Enforceability ........................................................................................................ 62

Conditions – Incorporation by Reference of Standards and the Like............................................................ 64

Conditions – Severability ............................................................................................................................... 65

PART 2 – SPECIFIC GROUNDS OF CHALLENGE ................................................................................................... 66

AQUACULTURE LICENCE APPEAL – DE NOVO ................................................................................................... 66

BIAS, FAIR PROCEDURES, FURTHER INFORMATION PROCESSES DELAY & FACILITATION OF MOWI –

INTRODUCTORY NOTE ....................................................................................................................................... 72

BIAS – OBJECTIVE & STRUCTURAL ..................................................................................................................... 73

Bias – Introduction & the Law of Objective Bias ........................................................................................... 73

Bias – External to the Impugned Process ...................................................................................................... 79

Bias – Prejudgment – Internal to the Impugned Process? ............................................................................ 80

Prejudgment Generally.............................................................................................................................. 80

Prejudgment – Internal to the Impugned Process? – Pattern of Erroneous Decisions............................. 83

Bias – Advisors To Decision-Makers .............................................................................................................. 86

Bias – Allegations ........................................................................................................................................... 87

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Bias – Some pleading issues ...................................................................................................................... 89

Bias – Structural............................................................................................................................................. 91

Structural Bias – the Statutory Architecture ............................................................................................. 91

Structural Bias – SWI’s Case ...................................................................................................................... 92

Structural Bias – the Sweetman Applicants’ Case ..................................................................................... 94

Structural Bias – Opposition ...................................................................................................................... 95

Structural Bias – Discussion & Decision ..................................................................................................... 96

Bias – ALAB’s relationship with the Marine Institute, the Aquaculture Appeals Office and the DAFM ..... 116

Bias – Specific Allegations............................................................................................................................ 120

Bias – RPS & ALAB meeting 24 November 2015 – Decision .................................................................... 120

Bias – Dr Saunders’ other Role & Declaration of Interests...................................................................... 123

Bias – ALAB/Saunders Interaction as to Saunders Report 2016/2017 .................................................... 128

Bias – Chair of Oral Hearing – 2017 ......................................................................................................... 141

Bias & Unfairness – Interactions as to AA/NIS ........................................................................................ 147

Bias – ALAB/Saunders Interactions as to Saunders Final Report 2020 ................................................... 168

BIAS & DELAY – FURTHER INFORMATION PROCESSES & NON-CIRCULATION OF REPORTS ........................... 168

Further Information – Introduction............................................................................................................. 168

Further Information – the Law .................................................................................................................... 169

Further Information – Including After the Oral Hearing – Decision. ........................................................... 173

Bending Over Backwards To Grant The Licence – Conclusion .................................................................... 176

ALAB’S ACCEPTANCE OF THE RPS REPORT ...................................................................................................... 176

DELAY – TIME EXTENSION/BREACH OF TIMELIMITS – S.56 of the 1997 Act................................................... 177

S.56 of the 1997 Act – Expedition – Duty and Objective of ALAB ............................................................... 177

The s.56 Notices and Notes Thereon .......................................................................................................... 178

Delay – Pleaded Bases of Challenge ............................................................................................................ 183

Delay – Opposition ...................................................................................................................................... 184

General Principles & Cases as to Timeliness ............................................................................................... 186

Holland 2023, Lafarge Redland 2000, Diesel 2020, Wexele 2010, Behan 2020. ..................................... 187

Browne 2011, Dolan 1975 ....................................................................................................................... 189

Cork County Council 2010, Celtic Roads 2013, Irish Refining 1990, Saleem 2011. ................................. 190

Dietacaron 2004 ...................................................................................................................................... 193

The S.56 Notices – Delay in Serving the First and Lack of Reasons ............................................................. 195

Delay Generally & the Proper Remedy........................................................................................................ 198

Unreasonable delay – Decision ................................................................................................................... 203

BIAS – GENERAL, PREJUDGMENT & DELAY ..................................................................................................... 209

EIA – 2011 OR 2014 DIRECTIVE?...................................................................................................................... 213

EIA – PROCEDURAL PURPOSE, PRECAUTIONARY PRINCIPLE AND SIGNIFICANCE OF UNCERTAINTY ............. 217

EIA – AQUACULTURE LICENCE – WAS IT DONE? ............................................................................................. 228

EIA – Introduction ........................................................................................................................................ 228

EIA – Screening ............................................................................................................................................ 229

Did ALAB do EIA? ......................................................................................................................................... 230

EIA – Notice to IFI ........................................................................................................................................ 232

EIA & HUMAN HEALTH – CONSUMPTION OF FARMED SALMON ................................................................... 234

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

EIA – DESCRIPTION AND ASSESSMENT OF PROJECT, NUMBER OF CAGES ..................................................... 238

EIA – AQUACULTURE LICENCE – CUMULATIVE IMPACT.................................................................................. 242

EIA – Cumulative Impact – Dr Saunders’ Final Report 8 December 2021 ................................................... 242

EIA – Cumulative Impact – ALAB’s Determination ...................................................................................... 244

EIA – Cumulative Impact – General Conclusion .......................................................................................... 245

LICE – INTRODUCTION & RELATION TO WILD SALMON & FwPM ................................................................... 245

IFI CG4 & PARTICULARS THEREOF ................................................................................................................... 248

LICE – SBM, CLAMS – SITE ALTERNATION, SYNCHRONOUS STOCKING AND HYDROLOGICAL ISOLATION ..... 249

The Licensed Production Cycle, SBM & CLAMS ........................................................................................... 249

Fallowing & the Fallowing Protocol 2000 ................................................................................................ 250

Meaning of Synchronicity ........................................................................................................................ 251

SBM & SBMPs .......................................................................................................................................... 251

Single Bay Site Alternation & Synchronous Stocking .................................................................................. 255

Tidal Excursion, Hydrological Isolation and Adjacency of Salmon Farms.................................................... 256

Irrational Failure to Require Single or Synchronous Bay Management – Decision..................................... 258

LICE – DR SAUNDERS’ FINAL REPORT 8 DECEMBER 2020 – INCLUDING AS TO SBM SYNCHRONICITY ........... 258

LICE – PROTOCOLS & SBM – LICENCE CONDITIONS VALIDITY ........................................................................ 262

Lice – Duration of Fallowing – Interpretation of the Aquaculture Licence – Decision................................ 262

Lice – Aquaculture Licence Conditions, SWI CG8 & Content of Sea Lice Protocol #3 ................................. 263

Validity of Condition 7.3 & the SBMP Condition – Enforceability – Initial Observations ........................ 264

Lice – Is there an SBMP for Bantry Bay?...................................................................................................... 265

Lice – RPS Report 2015 – Dispersion modelling .......................................................................................... 266

Lice – MOWI’s IPM/SBMP – October 2016 & s.47 Reply 2 November 2016............................................... 268

Lice Treatment Trigger Levels – Voluntary? ............................................................................................ 270

ALAB Conclusion on the MOWI IPM/SBMP 2016 and is MOWI Bound to Implement it? ...................... 271

Lice – Saunders Interim report 31 December 2016 on Lice and FwPM ...................................................... 272

Lice – Marine Institute/Jackson Response 9 February 2017 ....................................................................... 272

Lice – Oral Hearing 2017 & Report Thereon ............................................................................................... 273

Lice – sEIS 2018 & Responses thereto ......................................................................................................... 276

Publication of sEIS & Responses by Objectors, NPWS, Marine Institute, IFI ........................................... 278

MERC AA Report, September 2020 ......................................................................................................... 280

Lice Larva Buoyancy in EIA........................................................................................................................... 280

Lice Larva Buoyancy – Conclusion ........................................................................................................... 283

Lice Dispersal Modelling Generally ............................................................................................................. 284

Lice & Risk to Salmonids – ALAB’s Determination, 29 June 2021. .............................................................. 284

Lice & Risk to Salmonids – Selective Evidence – Decision ........................................................................... 285

LICE & SBM – IMPOSSIBILITY OF COMPLIANCE WITH LICENCE AS TO SBM .................................................... 287

SWI CG8 ....................................................................................................................................................... 287

SWI Submissions .......................................................................................................................................... 288

ALAB & MOWI Opposition ........................................................................................................................... 290

ALAB & MOWI Submissions......................................................................................................................... 292

Lice & SBM – Impossibility of Compliance with Licence as to SBM – Discussion & Decision...................... 294

Summary.................................................................................................................................................. 298

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

LICE & FWPM – EVIDENCE OF DR GOOD, O’TOOLE FWPM REPORT & ALAB CONCLUSION ........................... 298

LICE & FwPM – O’TOOLE REPORT – NON-CIRCULATION................................................................................. 301

SALMON ESCAPES – EIA & AA, the 2014 ESCAPE, CAGES & SEALS ................................................................. 306

Escapes – Introduction ................................................................................................................................ 306

Escapes – ALAB’s Determination ................................................................................................................. 307

Escapes – Pleadings ..................................................................................................................................... 308

Timing & Scope of EIA & AA – Basic Principles ............................................................................................ 312

Escapes – the Matrix ................................................................................................................................... 312

The Risk of and posed by Escapes – Basic Information and its Statutory Recognition ........................... 312

Escapes – The Aquaculture Licence as to Cage Specification .................................................................. 317

Escapes – History, the 2014 Escape, the Minister’s 2015 EIA as to the 2014 Escape ............................. 318

Escapes – MOWI’s s.47 Reply of 2 November 2016 as to Cage Specifications & ALAB’s Response ....... 320

Dr Saunders’ Interim Report of December 2016 ..................................................................................... 321

Withdrawn s.47 request as to 2014 Escape & Events at the Oral Hearing in February 2017 (and Bias) 323

The CEI Ruling of 16 January 2017........................................................................................................... 325

ALAB’s s.47 Notice to DAFM & Withdrawal thereof ............................................................................... 327

The Resumed Oral Hearing – the 2014 Escape & the Oral Hearing Report ............................................ 331

Escapes – DAFM Reports on 2014 Escape – Adequacy of EIA – Decision ................................................... 335

Escapes – EIA – CAGES, STORMS & SEALS ................................................................................................... 339

Mitigation of Escape once it has Occurred .............................................................................................. 339

Prevention of escape – DAFM EIA 2015 .................................................................................................. 339

Cage Specifications, Site Conditions & Seals – Saunders’ Final Report – unable to evaluate risk of escapes

................................................................................................................................................................. 340

Escapes – ALAB’s Determination & the Aquaculture Licence ................................................................. 343

EIA – Cages & Seal Scarers- Reliance on Later Regulatory or other Decisions & Boland Conditions ...... 346

Escapes – DAFM Structural Design Protocol, April 2016 ......................................................................... 353

Escapes – Cage Information and Drawings before ALAB & its Adequacy ............................................... 358

Escapes – ALAB argument - The Minister’s EIA – Cage Design & Specification .......................................... 362

Escapes – EIA – Information on the 2014 Escape and as to Cage Design & Specification – Decision......... 363

SEALS – Escape of Salmon and Acoustic Disturbance ..................................................................................... 364

Seals – AA & EIA Aspects & Introduction .................................................................................................... 364

Seals – Pleadings.......................................................................................................................................... 365

The Sweetman Applicants’ Grounds ....................................................................................................... 366

The Pleaded Opposition .......................................................................................................................... 368

ALAB’s Determination – as to AA ............................................................................................................ 370

Seals – the Evidence & Analysis Thereof ..................................................................................................... 371

EIS 2011, EIA 2015 & Oral Hearing Report 2017 ..................................................................................... 371

Coram Report February 2018 & Observations Thereon.......................................................................... 373

Coram – Comment & the Standard applicable in AA Screening ............................................................. 376

Coram as an AA Screening - Conclusion .................................................................................................. 381

After the Coram Seal Report, ALAB’s AA Screening out of Risk to Seals, February 2018, ALAB Decision to do AA April/June 2019, MOWI’s NIS July 2020 & Consideration of Seals thereafter.............................. 381

Position in February 2018 as to AA Screening of Risk to Seals – Decision .............................................. 385

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Seals – Dr Saunders’ Final Report December 2020 & ALAB’s Reaction................................................... 386

Seals – MERC AA Report September 2020, Marine Institute Screening Matrix, MOWI’s s.47 Reply March

2021 & MERC’s briefing note May 2021. ................................................................................................ 388

The Plea and Evidence as to Aquaculture Stewardship Council Certification ......................................... 389

Seals – ALAB’s AA Conclusion Statement May 2021 & ALAB’s Determination June 2021.......................... 391

EIA of risk of Farmed Salmon escape by Seal Predation ......................................................................... 391

ALAB’s AA Conclusion Statement May 2021 ........................................................................................... 391

ALAB’s Determination – as to AA ........................................................................................................... 393

The Aquaculture Licence ......................................................................................................................... 394

Seals – Scarers not proposed – FitzPatrick v An Bord Pleanála & Apple ..................................................... 395

Proffering of Mitigation, Aquaculture Licence Condition 4, s.42 Wildlife Act 1976 & the concept of

Disturbance ................................................................................................................................................. 400

Hellfire Massy .............................................................................................................................................. 403

Seals – Conclusion ....................................................................................................................................... 407

OTTERS............................................................................................................................................................. 409

Otters – Introduction ................................................................................................................................... 409

Otters – Pleadings........................................................................................................................................ 411

BIRDS ............................................................................................................................................................... 412

Birds – Pleadings .......................................................................................................................................... 412

Birds – Gittings & Crowe Reports ................................................................................................................ 413

Birds – Decision on the Pleadings as to AA Screening and the Storm Petrel in light of the Foregoing....... 416

Birds – EIA – the DPM Statistical Report May 2021 .................................................................................... 417

Birds – Conclusion ....................................................................................................................................... 421

O’Toole Kelp Report 2021 – Non-Circulation .............................................................................................. 421

WATER FRAMEWORK DIRECTIVE .................................................................................................................... 422

WFD – Introduction ..................................................................................................................................... 422

Dangerous Substances in Aquaculture Regulations 2008 ....................................................................... 423

WFD – Sweetman Applicants’ Pleadings & Conclusion ............................................................................... 424

WFD – Mixing Zones .................................................................................................................................... 425

WFD – The DIN Limit & the “Worst Case Scenario” Approach.................................................................... 425

WFD – SWI Pleadings ................................................................................................................................... 428

WFD – SWI Pleadings – Errors of Law...................................................................................................... 429

WFD – Introduction ..................................................................................................................................... 432

WFD – Complexity ................................................................................................................................... 432

WFD – Introductory Description.............................................................................................................. 434

WFD – Ecological Status, General Definition of Ecological Quality & Ecological Quality Elements ............ 437

WFD – Chemical Status, EQSs and the EQSD............................................................................................... 441

WFD – Monitoring ....................................................................................................................................... 442

WFD – DIN as a Nutrient & Pollutant, “EQS” for DIN and its application in Development Consent Applications

..................................................................................................................................................................... 443

What Constitutes Breach of the <170 µgDIN/l Standard? ...................................................................... 448

WFD – Cross-Examination of Dr Bass and Dr O’Toole - #1 .......................................................................... 449

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

WFD Caselaw – Weser 2016, Protect Natur 2017, Nordrhein-Westfalen 2020, Association France Nature

2022, Sweetman 2021 & Reasons. .............................................................................................................. 452

Weser 2015 – Deterioration .................................................................................................................... 452

Protect Natur 2017 .................................................................................................................................. 454

Sweetman v An Bord Pleanála & Breadán Beo ....................................................................................... 454

Reasons .................................................................................................................................................... 456

Nordrhein-Westfalen 2020 & Association France Nature 2022 – Localised & Temporary Effects ......... 457

WFD – Irish Regulation ................................................................................................................................ 460

WFD – Role of the EPA ................................................................................................................................ 465

WFD – DIN – Approach taken in Licence Application .................................................................................. 469

WFD – Introduction, EIS 2011 & EIA 2015 ............................................................................................... 469

WFD – RPS Report 2015 .......................................................................................................................... 470

WFD – Technical Advisors’ Interim Report (Saunders) 2016................................................................... 478

WFD – Bass/RPS presentation to Oral Hearing – September 2017......................................................... 480

WFD – Oral Hearing & Report thereon.................................................................................................... 481

WFD – sEIS 2018 ...................................................................................................................................... 482

WFD – Technical Advisors’ Final Report (Saunders) 2020 ....................................................................... 486

WFD – ALAB Determination .................................................................................................................... 487

WFD – DIN Issue – Affidavits & Cross-Examination of Dr Bass and Dr O’Toole - #2 ................................... 487

WFD – DIN Issue – Affidavits & Cross-Examination – Introduction & judgment of 16 March 2023 ....... 487 Dr Bass for MOWI – Affidavit 7 July 2022 & Donal Grant for the State – Affidavit 13 July 2022 ............ 489

John Murphy for SWI – Affidavit 21 October 2022 ................................................................................. 490

Dr Bass – Affidavit 19 January 2023 ........................................................................................................ 492

Dr O’Toole for ALAB – Affidavit 23 January 2023 .................................................................................... 494

Admissibility & Weight of Evidence......................................................................................................... 495

Order for Cross-Examination & Cross-Examination of Dr Bass and Dr O’Toole #2 ................................. 498

WFD – DIN Issue – Decision ......................................................................................................................... 503

WFD – EmBZ ................................................................................................................................................ 505

Promulgation and binding force of the EQS for EmBZ ............................................................................ 506

Breach and Adequacy of the EQS for EmBz ............................................................................................. 507

WFD – Benthos ............................................................................................................................................ 508

ALAB – FAIR PROCEDURES............................................................................................................................... 513

DECISION AS TO THE IMPUGNED AQUACULTURE LICENCE ............................................................................ 513

FORESHORE LICENCE ....................................................................................................................................... 515

Foreshore Licence – Introduction & Co-Dependency of Foreshore & Aquaculture Licences .................... 515

Foreshore Licence – Regard to Aquaculture Licence Decision .................................................................... 517

Foreshore Licence – Decision ...................................................................................................................... 524

Is EIA of Foreshore Licence applications for Aquaculture Required?.......................................................... 528

Introduction & Purposive Analysis .......................................................................................................... 528

S.13A(6) of the 1933 Act .......................................................................................................................... 530

Principles of Statutory Interpretation ..................................................................................................... 531

Application of Principles of Statutory Interpretation .............................................................................. 539

CONCLUSION ................................................................................................................................................... 542

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

PART 1 – INTRODUCTORY AND GENERAL MATTERS AND CORE GROUNDS1

INTRODUCTION & BRIEF CHRONOLOGY

The Site, the Salmon Farm, the Parties, the Impugned Aquaculture Licence & Foreshore Licence

1. These proceedings concern a proposed salmon farm at a site just south of Shot Head in Outer Bantry Bay, County Cork (the “Salmon Farm”). Bantry Bay is the largest of the rias,2 or drowned river valleys, in the southwest of Ireland. From its eastern head to its entrance to the Atlantic between Sheep's Head and Dursey Island, it is about 39km long, possesses some 200km of coastline and has a nominal sea area of about 230km2 (23,000ha).3

2. For present purposes, Bradan Fanad Teoranta and Comhlucht Iascaireachta Fanad Teoranta4 can be regarded,5 as both having formerly traded as “Marine Harvest Ireland” (“MHI”) and as trading now as MOWI Ireland and I need not distinguish between them.6 I will refer to them collectively as “MOWI”. It is or has been Ireland’s largest producer of farmed salmon.7 Inter alia, since 2008 it operates two licensed salmon growing farms (in substance one) at Roancarraig/Aghabeg in Bantry Bay about 6-8km west of the Shot Head Site. It also owns a small harvesting site further west again at Waterfall. About 5.5 km across the bay, southwest of the Shot Head Site,8 are two licensed sites at Gearhies formerly operated by Murphy’s Irish Seafoods – aka Fastnet Irish Seafoods. While MOWI has asserted that it owned these also,9 it became clear at trial that MOWI intended to purchase them but the sale was not complete.

3. In each of the above-entitled proceedings (collectively “the Proceedings” and respectively “the SWI Proceedings”, “the IFI Proceedings” and “the Sweetman Proceedings”) the respective Applicants seek, primarily, certiorari quashing:

• an Aquaculture Licence (the “Aquaculture Licence”) dated 26 January 202210 issued by the respondent Aquaculture Licence Appeals Board (“ALAB”), on foot of ALAB’s determination dated 29 June 2021, to MOWI, pursuant to the Fisheries (Amendment) Act 1997 (as amended) (“the 1997 Act”) for the cultivation

1 Headings in this judgment are for general guidance and navigation only. In particular, my comments and observations are not confined to sections

headed “Discussion and Decision” or the like.

2 Encyclopaedia Brittanica describes a ria as a “funnel-shaped estuary that occurs at a river mouth and is formed by the submergence of the lower portion

of the river valley. Generally occurring along a rugged coast perpendicular to a mountain chain, …”

EIS for a proposed salmon farm site at Shot Head, Bantry Bay, County Cork, May 2011 Volume 1. Main EIS document. §1.1.

4 Occasionally in the papers, in particular in the NIS of July 2020, referred to as “CIFT”.

5 for all relevant practical and legal purposes, if not technically and precisely accurately.

6 Bradan Fanad Teoranta, is part of the MOWI Ireland Group which is owned by Comhlucht Iascaireachta Fanad Teoranta (“CIFT”).

7 Oral Hearing report 8 November 2017.

8 See below.

9 Affidavit of Catherine McManus sworn 6 July 2022.

10 Bearing identification number “1/2021” – which is presumably erroneous and should read “1/2022” but nothing turns on that.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

of Atlantic Salmon in 18 cages11 on 42.49 hectares12 in Outer Bantry Bay, south of Shot Head, near Adrigole, County Cork (the “Site”) for, in effect, the lesser of 10 years and the continuance in force of the associated Foreshore Licence.

• the associated Foreshore Licence (the “Foreshore Licence”) also dated 26 January 202213 issued by the respondent Minister for Agriculture, Food and the Marine (“the Minister” and “DAFM”) to MOWI in accordance with his determination of 5 September, 2015, and in exercise of the powers conferred by s.3 of the Foreshore Act, 1933 (“the 1933 Act”) for the purpose of the salmon cultivation described in the Aquaculture Licence, on the Site, and to remain in force so long as the Aquaculture Licence remains in force. Foreshore Licence conditions require, inter alia, MOWI to, o use the Site only for the cultivation licensed by the Aquaculture Licence and for no other purpose. o comply fully with all terms and conditions of the Aquaculture Licence.

I will refer to the subject of the licence applications as the “Proposed Development” or the “Salmon Farm” or

the “Shot Head salmon farm”.

4. It is convenient here to record that certiorari of ALAB’s Impugned Decision would not revive the Minister’s decision on MOWI’s aquaculture licence application, the appeals from which were the subject of ALAB’s Impugned Decision - Deerland Construction.14

5. The Aquaculture Licence area and the Foreshore Licence area are the same and comprise the Site. By their respective statutory codes, the effect of,

• the Foreshore Licence is to licence MOWI’s use and occupation of the Site and the placing of the physical salmon farm thereon.15

• the Aquaculture Licence is to licence MOWI to engage in salmon farming on the Site.16

6. Two things are notable as to the foregoing:

• The explicit mutual interdependence and necessary co-existence of the Aquaculture Licence and the Foreshore Licence.

• That the Foreshore Licence was granted in January 2022 on foot of a determination of 5 September, 2015

– well over 6 years earlier.

11 Also called pens.

12 The four corners of the rectangular area are located by grid co-ordinates stipulated in each licence.

Though in fact issued on 29 March 2023.

14 Deerland Construction Limited v The Aquaculture Licences Appeals Board and The Minister for Communications, the Marine and Natural Resources

and Lett & Company Limited [2009] 1 IR 673.

15 Foreshore Act 1933 s.3(1).

16 1997 Act, s.3 “aquaculture” means the culture or farming of any species of fish, ……… “aquaculture licence” means a licence granted under section 14

to engage in aquaculture ………

The marine salmon farm production cycle lasts a nominal two years from the transfer of smolt to sea cages to harvest in the second year of the cycle and

thereafter the fallowing of the farm site until the commencement of the next production cycle, at the same time in Year 3.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

7. I was not addressed on the rationale for requiring two licences, not one, for aquaculture projects. No doubt it is good, if not readily apparent. And such an architecture is generally decided upon on a balance of advantage against disadvantage. But this case amply demonstrates the disadvantages of such a dual process

– arguable discoordination between the two statutory regimes – which the Applicants have sought to exploit.

8. Simplifying very considerably, MOWI’s proposed salmon farm will consist primarily of “cages” or “pens” located in Bantry Bay south of Shot Head and in which the farmed salmon are held for about 2 years while growing to a saleable size. These cages consist of weighted nets with closed bottoms, hung from floating circular collars which are anchored to the seabed in a particular location.17 The salmon will be fed via a feed distribution pipe from a barge.

9. It is appropriate to take judicial notice that, as indeed is also readily to be inferred from the dispute in these proceedings and the list of parties to the proceedings, salmon farming has for decades aroused passionate, controversial and genuine disagreement between its advocates and its opponents – as to its alleged economic and environmental merits and demerits. Amongst the opponents, representative of salmon angling18 are prominent. They allege that, in various ways – not least by lice infestation, salmon farming damages wild salmon stocks.

10. Unusually, one applicant for judicial review in the proceedings is a statutory body, Inland Fisheries Ireland (IFI). Its principal function is “the protection, management and conservation of the inland fisheries resource” of the State. IFI may be broadly identified with the conservation and exploitation of wild, as opposed to farmed, salmon. It is a prescribed consultee in aquaculture licence applications. Notably, IFI alleges objective bias against ALAB.

11. SWI and the Sweetman Applicants can fairly be described as environmental activists. The Applicants share broadly similar fears of what they consider to be the risks of environmental impact posed by the Salmon Farm. Inter alia, they fear adverse effect on the conservation status of wild salmon. They fear such adverse effect by way of increased sea lice infestation, and resultant mortality, of wild salmon and by way of risk of genetic damage to wild salmon stocks by interbreeding with escaped farmed salmon, thereby reducing their ability to survive in the wild. They also fear that elevated nutrient levels deriving from sea lice treatment residues, waste feed and faecal matter from the operation of the salmon farm will lower Bantry Bay water quality.

17 Generally illustrated in MOWI’s 2011 EIS Non-Technical Summary p19.

18 That is to say, sports angling for wild salmon by rod and line.

19 Established under s.6 Inland Fisheries Act 2010.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

12. Disputing those fears, MOWI asserts that its intended salmon production will be highly sustainable. It also deposes to an estimated capital cost of construction of the proposed Salmon Farm at €8.9 million,20 an expected biannual turnover of €26 million and operating costs of €12.2 million and to an expected 8 fulltime employees in the Salmon Farm and 15 more at its Donegal processing plant to process its production. MOWI intends single-generation21 22-month production over a 2-year cycle in which 850,000 smolts22 are introduced to the Site in October/November. Depending on growth rate, salmon will be harvested progressively over 6 months between March and August of the second year. The Site will be fallowed for at least 2 months before restocking.23 The Maximum Allowable Biomass (“MAB”24) of 2,800 tonnes will allow a total harvest of about 3,500 tonnes of salmon.25 MOWI intend 2-year single-generation production cycles asynchronous as between its Shot Head Site and its existing salmon farms at Roancarraig/Aghabeg in Bantry Bay. This will allow annual harvesting overall as each site will harvest every second year in alternate years. MOWI’s commercial object is better continuity of supply of farmed salmon. This is MOWI’s very reason for the proposed Shot Head Salmon Farm.26

13. MOWI has agreed not to act on the Aquaculture Licence and the Foreshore Licence pending determination of these Proceedings.

Judicial Review is not an Appeal on the Merits

14. As a general observation it seems to me that, despite their protestations to the contrary, much effort was spent by the Applicants in seeking to lure me into adjudication on the merits of the Impugned Decisions, cunningly disguised as issues of law, and into preferring the views of one or some of the competing experts on those issues over others’. Typically, disagreement as to merits was cloaked in assertion of irrationality. So, for example, Gregory Forde, deponent for IFI, asserts27 that it was irrational of ALAB to rely on the hydrological isolation of the proposed Shot Head Salmon farm from other farms in Bantry Bay as preventing cross-infestation of sea lice as between those farms when, in his opinion, wild salmon can swim throughout the bay and acquire lice regardless of hydrological isolation. Whether such a consideration outweighs or undermines the effects of hydrological isolation specifically or ALAB’s decision more generally is, in my view, a matter of expert judgment of merit not of irrationality. The same may be

20 Unless otherwise indicated, all monetary figures in this judgment are approximate.

21 Stocking a salmon farm with only a single generation of salmon at any particular time is an element of sea lice risk reduction as it enables periodic

fallowing of the site as opposed to its continuous operation.

In the wild, smolt are young salmonids – salmon or sea-trout – in the process of their first migration to sea. They have grown at about 2.5 or 3 years

old to the stage of moving from the freshwater river of their birth to the saltwater ocean The process involves physiological changes to allow adaptation

to saline waters. Typically these changes occur during time spent in brackish estuaries. MOWI produces smolts at hatcheries elsewhere.

23 Fallowing to break lice cycles is an element of sea lice risk reduction. See ALAB determination §6.5.2 & Aquaculture Licence Schedule 4.

24 The maximum weight of live fish allowed on site at any given time.

25 Once salmon stocks approach MAB, harvesting must start so MAB is not exceeded.

26 EIS §1.3 §3.2.1, p148.

27 Affidavit 30 September 2022 §8(iii).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

said as to his criticism of the Marine Institute’s methodology of counting lice. Such issues were for ALAB, not the court, to consider and resolve.

15. It bears recollection that,

“Judicial review is not available as a remedy to correct errors or to review decisions so as to render the High Court a Court of Appeal from the decisions complained of …28 The system of judicial review is radically different from the system of appeals. When hearing an appeal, the Court is concerned with the merits of the decision under appeal. When subjecting some administrative act or order to judicial review, the Court is concerned with its legality. On an appeal, the question is “right or wrong?”. On review, the question is “lawful or unlawful?”29.

16. As Humphreys J put it, strikingly, in Holohan30 in judicial review, the court can’t quash for unreasonableness merely if it considers that the decision, on its merits was, even, “clearly wrong”:

“……. the court cannot decide that the exercise by a decision-maker of a discretion, or a finding as to fact, is simply wrong (or even clearly wrong) on the merits, if there is material to support it and if the

conclusion is reached by a logical process, without factual error and supported by reasons, and does not disproportionately interfere with rights.

The underlying reason why a court should not be permitted to find that an administrative or executive decision is wrong, or even clearly wrong, is the separation of powers. To do so the court would, as Lord Brightman said, “be itself guilty of usurping power””

17. A striking – though not unusual – feature of these proceedings and the dispute underlying them is that highly qualified experts, each to a greater or lesser extent expressing confidence in their view, are ranged on either side of the argument and have sworn affidavits accordingly. I emphasise that the Courts cannot take sides in, much less resolve, disagreements as to the substantive merits and demerits of salmon farming – either generally or as to the specifics of the Aquaculture Licence in question in these proceedings. I am confined by law to deciding only the legality – as opposed to the merits – of the Impugned Decisions to grant the Aquaculture Licence and the Foreshore Licence. While, for the purpose of this judgment and elucidation of the legal issues, I must describe many of the scientific and practical concepts in dispute, I cannot, and am not to be read as purporting to, resolve them.

28 See State (Abenglen Properties) v Dublin Corporation [1984] I.R. 381.

29 See, amongst many authorities, Dunne v Minister for Fisheries and Forestry [1984] 1 I. R. 230, at p.237. Costello J citing Wade's Administrative Law (5th

ed., p.34), Dunnes Stores v An Bord Pleanála [2016] IEHC 226 – cited in North Meath Wind Farm Ltd v An Bord Pleanála [2018] IEHC 107. Redrock

Developments Limited v An Bord Pleanála [2019] IEHC 792, Monkstown Road Residents’ Association v An Bord Pleanála [2022] IEHC 318.

30 Holohan v An Bord Pleanála [2017] IEHC 268 citing Chief Constable of North Wales Police v Evans [1982] 1 W.L.R. 1155. Cited to this effect in Fernleigh

v ABP & Ironborn [2023] IEHC 525. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

ALAB

18. ALAB is established by the 1997 Act.31 Its primary function is to determine appeals by persons aggrieved by decisions of the Minister on aquaculture licence applications and like matters. ALAB must decide such appeals as if the application for the licence had been made to it in the first instance. In doing so, it must hear objectors and prescribed bodies, may require production to it of information and documents and must conduct Environmental Impact Assessments32 (“EIA”), and Appropriate Assessments33 (“AA”) as required. It may hold oral hearings and it may refer questions of law to the High Court. Its decisions operate to nullify the appealed decisions of the Minister and result, in the case of a decision to grant, in the issuing of an Aquaculture Licence entitling the recipient to conduct aquaculture in the place or waters specified, notwithstanding the existence of any public right to fish in any waters to which the licence relates. ALAB may impose conditions on such licences.

19. ALAB is entitled by statute to deal with multiple appeals of a Ministerial decision on an Aquaculture Licence application as a single appeal and did so in this case as to the 14 appeals of the Minister’s decision. Hence, I will refer to the Appeal, singular, as encompassing all 14.

20. Notably, attempts to improperly influence ALAB in its consideration and decision of appeals is an offence.34 ALAB’s members are required by statute to declare and disclose their interests as relevant. Importantly, and as I will consider in greater detail, while the members of ALAB are drawn from what might be called constituencies of those interested in Aquaculture, its members are not delegates of and may not report back to those interests – they must, as it were, leave their interests at the door when deciding appeals.

21. This brief and incomplete account of the statutory scheme applicable to ALAB, and analogy with both statutory schemes such as that applicable to An Bord Pleanála and the nature of decisions made by bodies such as An Bord Pleanála, suffice to establish, which was not disputed, that ALAB is a quasi-judicial body on which duties of independence and impartiality are laid and it must conduct its affairs accordingly.

22. Dr Graham Saunders, marine ecologist, was on 7 April 2016 appointed consultant Technical Advisor to ALAB as to the MOWI Appeal. He retained that position to the point of issuing his final report and recommendations to ALAB in December 2020.

31 As a body corporate with perpetual succession and an official seal.

32 In accordance with whichever Directive on the assessment of the effects of certain public and private projects on the environment applies – see further

below.

33 within the meaning of the Habitats Directive 92/43/EEC on the conservation of natural habitats and of wild fauna and flora.

34 s.31 of the 1997 Act.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

The Geography & WFD Waterbodies

23. A general understanding of the geography will assist:

Figure 1 Bantry Bay35

• the Shot Head Salmon Farm Site is marked “x”.

• Shot Head lies just north of the Salmon Farm Site.

• North of Shot Head lies Trafrask Bay, into which the Dromagowlane/Trafrask river system36 flows. The river mouth is 2.5 km by sea from the proposed Salmon Farm.37

• North west of Shot Head lies Adrigole Harbour. For purposes of the Water Framework Directive (“WFD”)38, Adrigole Harbour is a separate water body.

• Gearhies is roughly south southeast of the Site on the southern shore of the Bay.

• Bantry Bay is divided into two water bodies for purposes of the WFD. The Shot Head site is in Outer Bantry Bay. Inner Bantry Bay lies at the head of the Bay essentially between Whiddy Island and the mainland to the East.

• Glengarriff Harbour lies generally to the East and North-East of the Shot Head Site. Glengarriff Harbour and Woodland SAC is in this area, about 10km from the Site. Its conservation interests include seals and otters.

• Berehaven lies west of the Site and is essentially the area of water between Bere Island and the mainland to the north and west. For purposes of the WFD, Berehaven is a separate water body.

• The Waterfall site is in Berehaven. The Roancarrig/Aghabeg site lies between the Waterfall site and the Shot Head site but in the Bay rather than in Berehaven.

35 This map is found in Exhibit GF2 to the affidavit of Gregory Forde sworn 23 September 2021 in the IFI Proceedings.

36 This river system is variously and interchangeably referred to, in whole or in part, as the Dromagowlane/Trafrask river system, the Dromagowlane River

and the Trafrask River. It consists of three confluent waterways discharging as a single waterway to Trafrask Bay – an inlet of Bantry Bay north of Shot

Head. Nothing seems to turn on the nomenclature used.

37 sEIS 2018 p4.

38 See below.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Outside the Bay lie various Wild Birds Directive Special Protection Areas (“SPA”) to which I will refer as required.

24. The Outer Bantry Bay water body identified39 for purposes of the WFD 40 amounts to 27,618 hectares.41 It has always had High Water Quality Ecological42 Status within the meaning of the WFD. Berehaven, separately identified for WFD purposes, has Good Water Quality Ecological Status. Adrigole Harbour, Glengarriff Harbour and Inner Bantry Bay are water bodies separately identified as having no assigned Water Quality Ecological Status.

The Proceedings & Structure of Judgment – a brief note

25. The Aquaculture and Foreshore licence applications were made in May 2011 and continued to March 2022. Notably, the Proceedings agitate about43 63 Core Grounds of judicial review. They have generated no less than 53 affidavits and about 12,000 pages of exhibits. 13 sets of written submissions run to 168,564 words44 and the Authorities folder contains 240 items. This position is generated by high controversy – pitting even the state agency, IFI, responsible for inland fisheries (including salmon angling) against the state agency, ALAB (and in substance the Minister), responsible for aquaculture licensing, with the former state agency even making the “ugly allegation”45 of objective bias against the latter. That controversy is superimposed on a Byzantine regulatory regime46 which was, with environmental law generally, in a state of considerable flux over the lengthy duration of the process from licensing application to the impugned decisions. The trial took 15 days – of which all but part of one consisted of legal argument. Against that background I hope that the regrettable length of this judgment may be thought tolerable.

26. 3 deponents were cross-examined.47 They were,

• John Murphy, a director of SWI. He is a retired fishery manager and holds a BSc in wildlife biology and an MSc in environmental sustainability. He deposes to professional familiarity, which I have no reason to

39 See Natura Impact Statement July 2020 Fig 2.3.

40 See below.

41 Transcript Day 8 p6. There is a discrepancy here with the 23,000ha indicated above but nothing turns on it.

My emphasis need merely be noted here. My reason for it will become apparent in due course.

43 There is appreciable overlap and duplication and the number in part turns on the categorisation of sub-grounds.

44 A brief websearch suggests that novels average 70,000 to 120,000 words. One website suggests that anything over 110,000 words is considered too

long for a fiction novel. While these websearch results are to be taken with a pinch of salt, they suffice to give a general impression.

45 Shatter v Guerin [2019] IESC 9; [2021] 2 IR 415, §30.

46 A flavour of the regime can be seen in the plea by SWI’s plea that “The impugned decision is invalid because ALAB failed to carry out an assessment

of the structural adequacy of salmon cages to prevent escapes, contrary to Article 6(3) of the Habitats Directive, as implemented by Regulation 42(8) of

the European Communities (Birds and Natural Habitats) Regulations 2011 as amended), and Articles 2, 3 and 8a of the EIA Directive or their precursors,

as implemented by Regulation 3 of the Aquaculture (Licence Application) Regulations 1998 (S.I. No. 236 of 1998) as amended by the Aquaculture

(Licence Application) (Amendment) Regulations 2010 (S.I. No. 280 of 2010), Aquaculture (Licence Application) (Amendment) (No. 2) Regulations 2010

(S.I. No. 369 of 2010) and Aquaculture (Licence Application) (Amendment) Regulations 2012 (S.I. No. 301 of 2012) and European Union (Environmental

Impact Assessment) (Aquaculture) Regulations 2012 (S.I. No. 410 of 2012) and / or as required by Article 3(1) of the Aquaculture Appeals

(Environmental Impact Assessment) Regulations 2012 (S.I. No. 468 of 2012).”

47 Transcript Day 5.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

doubt, without necessarily accepting his views, with the risks posed to wild salmon by salmon farming activities.

• Dr Neill Bass, principal of Watermark,48 consultant to MOWI and, inter alia, primary author of its 2011 EIS49 and 2018 Supplemental EIS (“sEIS” – also referred to at trial as the “Watermark Report”), and

• Dr Ciar O’Toole is the full-time Technical Advisor to ALAB. She took up her position in October 2020 - a little before Dr Saunders completed his final report in December 2020. She holds a bachelor’s degree in marine science, a master’s degree in marine environmental protection and a PhD in zoology – for which her thesis related to Atlantic salmon and brown trout. She came to ALAB from some years working for the Marine Institute50 (“MI”) researching anadromous51 fish – in particular the ecology and life histories of salmonids in freshwater. However, and despite suggestion to the contrary in cross-examination, I am happy that Dr O’Toole is expert for purposes of receipt of her evidence as to marine/tidal/salt water environments as they bear on salmon. As to the Appeal, she is Dr Saunders’ successor but had no interaction with him as to the Appeal. She, not Dr Saunders, swore the expert replying affidavits for ALAB.

She said, and I accept, that she was not chosen to do so to shield Dr Saunders from cross-examination.

27. I am grateful to all counsel for the manner and substance of their submissions and for their courtesy and co-operation with me and each other as to the practicalities of trial.

28. As to the structure of the judgment, the table of contents above will, I hope, assist.

Brief Chronology

29. A remarkable feature of the process leading to the grant of the Impugned Licences is its duration from May 2011 to March 2022. It will be necessary to consider that process in some detail, but it will assist comprehension to first summarise a bare outline of that process, as follows:

48 I will generally refer to Dr Bass as opposed to Watermark.

49 Environmental Impact Statement 2011. Such documents are, since the 2014 EIA Directive took effect, termed an Environmental Impact Assessment

Report (“EIAR”).

50 Scientific advisors to the Department of Agriculture, Food and the Marine.

51 Migratory fish species that move between fresh and salt water – such as salmon and sea trout. Salmon typically spend their first two or three years in

freshwater and then head to sea for about a year to 18 months before returning to their home river to spawn. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Date Event May 2011 Dr Bass for MOWI produces an EIS52 as to the proposed aquaculture project – for 12 cages.

June 2011 MOWI applies to the Minister for an Aquaculture Licence and a Foreshore Licence.53 The single combined application was accompanied, inter alia, by the EIS.

The Application and EIS describe the project as

• consisting of 12 cages (plus 2 cages for periodic temporary use54),

• those cages occupying 5.88 hectares of foreshore plus anchors and moorings, occupying an overall 19.20 hectares,

• being 45% of the proposed licensed foreshore area of 42.5 hectares.

(The proposed total licensed foreshore area stayed constant as between the 12-cage and the later 18-cage proposals. The total area is variously stated in the papers as 40 ha, 42.5ha and 42.49 ha but that was not an issue of dispute in the Proceedings.)

The Application states that the farm will hold a maximum of 2,800 tonnes of salmon at a peak stocking density of 10kg/m3 – allowing a total harvest of 3,500 tonnes per 2- year cycle.

Consultation with the public and prescribed bodies ensues.

2014 230,000 salmon escape from a salmon farm in Bantry Bay (not MOWI’s)

September The Minister approves a draft EIA55 explicitly in accordance with the 2011 EIA 2015 Directive,56 the Aquaculture Licence Regulations 199857 and the Aquaculture EIA Regulations 2012.58

The EIA describes the project as consisting of 12 circular cages, each of 128m circumference, plus 2 additional cages, on a footprint of 5.88 hectares.

It concludes that “the salmon farm as proposed is not likely to have a significant negative impact on the environment” and that “there are no substantial environmental grounds for refusing to approve this application”.

The Minister determines to grant the Aquaculture Licence and the Foreshore Licence.

The resultant drafts of both are in evidence.

Environmental Impact Statement within the meaning of the EIA Directive 2011/92/EU on the assessment of the effects of certain public and private

projects on the environment. Since the amending 2014 directive this document is termed an Environmental Impact Assessment Report (“EIAR”).

53 On a single pre-printed form encompassing both applications.

54 During biennial grading /harvesting for 4 months (between months 16 to 20 i.e. February to June of year 2 of the production cycle).

55 The document put in evidence as that draft EIA is dated 12 June 2015.

56 EIA Directive 2011/92/EU on the assessment of the effects of certain public and private projects on the environment.

57 S.I. No. 236 of 1998, Aquaculture (Licence Application) Regulations, 1998.

58 S.I. No. 410 of 2012, European Union (Environmental Impact Assessment) (Aquaculture) Regulations 2012.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Date Event The draft Aquaculture Licence envisages 12 cages only, in 2 rows of 6, and stipulates that, apart from a feed barge, “No other floating structures may be moored for extended periods at the site.” It does not provide for the 2 additional temporary cages for which licences had been sought. October 2015 et 14 Appeals of the determination to grant the Aquaculture Licence ensue. seq One appeal is by MOWI.59 Inter alia, it argues against limiting cage numbers and for flexibility to adopt evolving best practice in this regard. By this time, MOWI say, best practice to produce the same quantum of salmon requires 16 cages plus 2 additional cages.

The remaining appeals are by the Applicants in the Proceedings and many of the Notice Parties in the proceedings.

Consultation with the public, various technical advisors to ALAB and with prescribed bodies ensues – as do numerous procedures embarked upon by ALAB to investigate various, primarily environmental, matters by way of soliciting further reports and information/responses from participants.

ALAB considers the Appeals at over 50 meetings of its board members between 20 October 2015 and 24 June 2021. November 2015 RPS60 report entitled “Water Quality Modelling for all existing & currently proposed salmon farm sites in Bantry Bay”, for MOWI

• Surprisingly, given it post-dates MOWI’s appeal, RPS assumes 1261 precisely geolocated cages.

• This is surprising not least as:

o a reduced modelling cell size of 20m2 is expressly used in the vicinity of the site “so that discharges from individual salmon pens could be discerned”.62

59 MOWI summarises its appeal as requesting that

(i) the licence permit a MAB of 2,800 tonnes on site at any time during the growing cycle.

(ii) no harvesting limitations be imposed.

(iii) the dimensions of floating facilities should not be specified in such detail and that the number of cages be increased to 18.

(iv) that CIFT be substituted as the applicant for the licence. See Affidavit of Catherine McManus 6 July 2022.

60 Environmental consultants to MOWI.

Figure 4.2: Proposed pen locations, Shot Head, Bantry Bay & p28.

pp8-9.

At p28 it states: “Discharges were incorporated into the model as arising from a point source at each pen centre, i.e. Shot Head and Roancarrig / Ahabeg

12 pens each …” Figure 4.2: Proposed pen locations, Shot Head, Bantry Bay depicts and attributes a national grid location to each of only 12 cages. Figure

4.2 is described as “showing the proposed arrangement of the fish pens at Shot Head, as used to define dispersion sources”.

p30 states: “Pen centre point sources were used for dispersal simulations from all sites.”

p36 states as to phosphorus discharge: “As for the nitrogen model, the discharge was included within the model from a source at each pen centre.”

p43 states as to Biological Oxygen Demand that the BOD source was “introduced at each pen centre”.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Date Event o “Discharges were incorporated into the model as arising from a point source63 at each pen centre, i.e. Shot Head ….. 12 pens each.”64

This RPS Water Quality Report is the subject of further detailed examination below.

MOWI responds to the objectors’ Appeals – enclosing the RPS Report as, it asserts, confirming the findings of the EIS.

8 April 2016 ALAB writes to all concerned recording that,

• It had carefully considered, taken legal advice upon and rejected allegations by Appellants that ALAB members were affected by conflicts of interest.

• ALAB now proposed to consider the substance of the appeals.

• S.56(2)(a) of the 1997 Act required it to endeavour to determine appeals within 4 months of receiving them. ALAB was unable to do so but will try to do so by 30 November 2016.

Notes:

• IFI say this s.56 time extension letter was invalid as the 4-month period had expired

2 months earlier on 7 April 2016.

• This was the first of many such s.56 time extension letters. Some gave reasons for the extensions. Some, it is argued, are invalid because they did not.

26 October MOWI Shot Head Integrated Pest Management /Single Bay Management Plan 2016

31 December ALAB Technical Advisor’s Interim Report (Dr Saunders) 2016 • Inter alia, Dr Saunders confirms that the RPS report of November 2015 assumed only 12 cages and that a s.47 request65 had ensued as to the impact on the seabed of an additional 4 cages.66

• In reply, MOWI had confirmed that the unchanged MAB of 2,800 tonnes would be maintained across the additional cages and that the change in cage configuration will only result in more diffuse waste outputs, with lower concentrations distributed over a slightly greater footprint area.

p49 records that the settleable solids analysis is based on a near-field or a far-field “Allowable Zone of Effect” – the near-field AZE being equivalent to an

area bounded by a line 25m beyond the pen footprint and the latter being equivalent to an area bounded by a line 100m beyond the pen footprint. P52

records that solids discharges were simulated … located at the centre of each pen … where 12 pens are used / proposed, a total of 24 sources were used,

i.e. one source per pen relating to feed and a second relating to faeces.

p61 records that in the particle tracking modelling approach used for EmBZ “particles were released from the pen centres” and “The 100m boundary

from the site required by the EQS is indicated by the black outline around the pens.”

P63 states of Alphamax® lice treatment that “Dispersion from pens is modelled here since it provides a worst case” and was modelled on the basis of

consecutive dosing of individual cells. It states that in Figure 5.40 “The solid line around the pens denotes the EQS boundary.”

p66 states that sea lice dispersal modelling was based on “sources placed in the centre of each pen.”

63 This was described as a conservative assumption as in reality discharges would be diffusely from each pen.

64 p28.

65 A request for further information by ALAB to MOWI under s.47 of the Fisheries (Amendment Act) 1997.

66 Saunders interim report 31/12/16 p75 & p94.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Date Event

• Dr Saunders accepted that would be the likely outcome of the requested increase in cage numbers and, since the impacts would be confined to well within the licence area, he considered the issue of negligible concern. May & IFI made a semi-quantitative and inexhaustive electrofishing survey of salmonids in September the Dromagowlane/Trafrask river system and included it in a s.47 reply67 to ALAB68 2017 shortly before the resumed oral hearing.69 February, Oral Hearing – chaired by ALAB Member Prof Owen McIntyre. Prof McIntyre’s report September & on the oral hearing recommends that: November 2017

• ALAB should issue an aquaculture licence – conditional on the outcomes of the following:

1.

• Before deciding the licence application ALAB should:

o (1) require an sEIS on ▪ sea lice risks to wild salmonids migrating from/to the Dromagowlane and Trafrask Rivers, ▪ any resulting implications for local Freshwater Pearl Mussel70 (“FwPM”)

populations, ▪ potential impact of salmon farm waste on water quality, having particular regard to the required maintenance of WFD ‘good water status’.

o (2) do desk-top studies, which may indicate the need for supplemental AA screening of risks to:

▪ The otters of the Dromagowlane and Trafrask catchments, – including the potential impact of declining wild salmon stocks;

▪ The common seal population in the Glengarriff Harbour & Woodland SAC;

and ▪ Birds in nearby SPAs.

o (3) “make every effort to consider the potential impacts of large-scale farmed salmon escapes.” February 2018 Dr Tom Gittings71 reported to ALAB by “Bird Expert's Report: Briefing Note Bird impact assessment”.

67 s.47 of the Fisheries (Amendment) Act 1997.

68 Dated 6 September 2017.

69 See below.

70 Margaritifera Margaritifera.

71 Consultant to ALAB.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Date Event

• This was a desktop review (not a full AA Screening) as to the possible need for AA.

• He considered MOWI’s EIS and the Minister’s EIA to be inadequate as to wild birds and that “further AA Screening” was required.

• Of some note, Dr Gittings considered the Storm Petrel – an SCI72 of The Bull and The Cow Rocks SPA.73 Inter alia he states:

o “…. in the absence of any more detailed information on Storm Petrel foraging ranges, the possibility that the site is within the likely core foraging range distance of the Storm Petrel SCI of the Bull and the Cow Rocks SPA cannot be ruled out.”74 And the Storm Petrel is “screened in for assessment”.75

Note: While it could be clearer, this observation seems to be for the purposes of Dr Gittings’ own “preliminary screening”76 with a view to focussing his report. It does not appear to consist in an opinion that AA is required as to the Storm Petrel.

Thereafter, his report devotes considerable attention to the Storm Petrel.

o “Overall, therefore, it is very unlikely that there is any significant spatial overlap between foraging Storm Petrels from the Bull and the Cow Rocks SPA and the proposed fish farm site. Therefore, development of the proposed fish farm is unlikely to have any impacts on the Storm Petrel SCI of the Bull and the Cow Rocks SPA.”77 o As to cumulative impacts, “Storm Petrel …. are unlikely to regularly occur within Bantry Bay (see above), so these species are not discussed in this section.”78

• Dr Gittings concludes: “The present briefing note largely contains the information required for this screening. However, based on the assessment presented here, a stage 2 Appropriate Assessment of the potential impact of Gannet mortalities on the Gannet SCI of the Bull and the Cow Rocks SPA may be required.”

Note: As will have been seen, of Dr Gittings’ conclusion does not identify the Storm Petrel for AA.

1 October 2018 SWI letter to ALAB

72 Species of Conservation Interest – within the meaning of the Wild Birds Directive. A species for the protection of which an SPA is designated as such.

73 Special Protection Area – within the meaning of the Wild Birds Directive.

74 p8.

75 p10.

76 p6.

77 p20.

78 p21.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Date Event

• ALAB is clearly unable to conclude in AA as a certainty that there would be no impact on any European Site or protected fauna or flora.

• MOWI, the Minister and ALAB cannot keep on seeking some scientific basis for such certainty.

• “ALAB cannot keep on adjourning and deferring a decision indefinitely. While there may be a statutory basis for the deferral of a decision, the licence Applicants should not be indulged with continuous adjournments.”

• The licence application should be immediately refused “without further delay”.

April 2019 Dr Olivia Crowe79 submitted an AA Screening report80 to ALAB as to bird conservation interests.

She recommends AA as to specific SCI and SPAs:

SCI SPA

• Fulmar • Beara Peninsula SPA,

• Iveragh Peninsula SPA,

• Deenish Island and Scariff Island SPA

• Gannet • The Bull and The Cow Rocks SPA

• Skelligs SPA

• Guillemot • Iveragh Peninsula SPA

She explicitly does not recommend AA as to the Storm Petrel as she considers spatial overlap between it and the Site to be highly unlikely – it was last recorded in Bantry Bay in 1995.

14 May 2019 The Marine Institute (the “MI”) confirms and explains to ALAB its view that in the Dromagowlane/Trafrask River the brown trout, not migratory salmonids, acts as host for the FwPM glochidia larvae.

July 2020 MOWI submits a finalised81 Natura Impact Statement82 (“NIS”) to ALAB.

September MERC report83, to ALAB to inform AA, 2020 • concludes that the Salmon Farm will not impact adversely on SCI species or SPA conservation objectives.

ALAB

• notifies the public that it is performing AA.84

• publishes the NIS85 of July 2020 and the MERC report of September 2020.

79 Consultant to ALAB.

As terminology varies, I will used the self-explanatory terms “AA Screening” and will use the terms “AA” and “Appropriate Assessment” for the

substantive study consequent on “screening in” a European Site – the process sometimes termed “Stage 2 Appropriate Assessment”.

81 There were earlier versions which were not published and are not in evidence.

82 To inform AA and as contemplated by the Habitats Regulations 2011.

83 Environmental Expert Consultants to ALAB.

84 And that the specific SCIs and SPAs of concern in AA are Fulmar (Beara Peninsula SPA, Iveragh Peninsula SPA, Deenish Island and Scariff Island SPA),

Gannet (The Bull and The Cow Rocks SPA and Skelligs SPA) and Guillemot (Iveragh Peninsula SPA).

85 Natura Impact Statement to inform appropriate assessment within the meaning of the Habitats Directive.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Date Event

• invites submissions from the public, the parties to the Appeal and prescribed bodies.

8 December Dr Saunders, as technical advisor to ALAB, issues his final report.

2020 • He “can offer no substantive technical reasons to refuse the current licence application.”

Note: this is a crude account of a complex report, to which I will return.

28 May 2021 Dr Ciar O’Toole86 reports to ALAB,

• on her inspection of the Dromagowlane/Trafrask River system.

• her agreement with the MI that resident brown trout rather than migratory sea trout and salmon were, in that river system, the majority of hosts for FwPM glochidia larvae.87

• her opinion that a decline in the migratory salmonid populations of the river system would not negatively impact on any remaining FwPM populations in that system.

• her agreement with MERC that effects of the Salmon Farm in combination with a licenced but inactive mechanical kelp harvesting activity in Bantry Bay on SCI species or conservation objectives of any Natura 2000 Site88 were highly unlikely.

ALAB records its AA Conclusion

• that the Salmon Farm will have no significant adverse effect on the integrity of the conservation objectives of European Sites or on their SCIs.

24 & 29 June ALAB determines the Appeals89 2021 • in favour of granting the Impugned Aquaculture Licence,

• recording EIA and AA of the proposed Salmon Farm.

• and, inter alia, “on the basis of an increase in the number of cages from 14 to 18 to facilitate current best practice”.

26 January 2022 ALAB issues the Aquaculture Licence

• inter alia, on the basis of 18 cages on a footprint of 8.82 hectares.

The Minister signs the Foreshore Licence on 26 January 2022,

29 March 2022 but recalls it and reissues it on 29 March 2022 having amended it to explicitly record that it had issued on foot of the Minister’s determination of 5 September 2015.

30. The following, inter alia, are notable in the foregoing:

86 ALAB Technical Advisor.

Salmonid hosting of these larvae is essential to the Freshwater Pearl Mussel life-cycle.

88 Special Protection Areas – within the meaning of the Wild Birds Directive and Special Areas of Conservation within the meaning of the Habitats Directive.

I will use this term in preference to “European Site”.

89 In exercise of its powers under s.40(4)(b) of the 1997 Act – “(b) determining the application for the licence as if the application had been made to the

Board in the first instance.” The determination reads in part: “Having considered all the foregoing, the Board determined at its meeting on 24 June 2021,

pursuant to section 40(4)(b) of the Act, to determine the application for the licence as if the application had been made to the Board in the first instance

and TO GRANT a licence to the Applicant for the proposed activity on the Site in accordance with the draft licence prepared by the Minister, but subject

to the varied and amended Terms and Conditions as set out in this Determination.” [emphasis added] SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Between the making of the licence applications in June 2011 and the issuing of the Foreshore Licence in March 2022, almost 11 years elapsed.

• Between the making of the Appeals in October 2015 and the issuing of the Foreshore Licence in March 2022, almost 6.5 years elapsed.

• The Foreshore Licence of March 2022 envisaged 18 cages on a footprint of 8.82 hectares. The Minister’s determination of September 2015 on foot of which it issued had envisaged 12 cages on a footprint of 5.88 hectares. That represents a 50% increase in the number of cages and cage footprint area.90 However, the maximum allowable biomass91 of 2,800 tonnes permitted by the Aquaculture Licence was unchanged – as was the maximum stocking density of 10kg/m3. Also unchanged was the overall Foreshore Licence area.

Effects on Wild Salmonids and FwPM – EIA Only & Licensing Despite Significant Effects

31. As it has the scope to cause confusion, an initial observation as to the identification of the environmental procedures bearing on impacts on wild salmonids (the Atlantic Salmon and the Brown/Sea Trout92) and the FwPM may assist. Those who oppose the Shot Head salmon farm say it will incubate large numbers of lice which will infest and cause increased mortality and morbidity in wild salmonids in the vicinity of the farm93 and that the resultant diminution of the population of wild salmonids in the nearby Dromagowlane/Trafrask River system will diminish the availability of salmonid hosts for FwPM larvae – which hosts are essential to the life-cycle of, and recruitment94 to, the FwPM population in that river system.

32. Accordingly it is vital to understand the level of legislative protection from sea lice of the wild Atlantic Salmon, the Brown/Sea Trout95 and the FwPM with which ALAB was concerned in making its Impugned Determination. There is no dispute but that risks to the FwPM, the wild Atlantic Salmon and the Brown/Sea trout were proper subjects of assessment in EIA in this case.

33. However, it might have been thought that they would also attract assessment under other more protective and demanding rubrics such as AA or as strictly protected species for purposes of Article 12. In the particular circumstances of this case, that is not so. Only EIA arises. That is significant as whereas, for example, AA can compel refusal of development consent, EIA is primarily procedural and does not as a matter of law compel refusal of development consent. I will briefly explain how that position arises.

90 (8/82 – 5.88)/5.88 x 100 = 50%.

91 In substance, the maximum weight of salmon allowed onsite at any point in time.

92 Both are the same species – the sea trout is an anadromous form of brown trout.

93 Leaving aside for present purposes identification of the vicinity.

94 Natural population increase by reproduction.

95 Both are the same species – the sea trout is an anadromous form of brown trout.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

34. The wild Atlantic Salmon is listed in Annex II to the Habitats Directive as a species – “Salmo Salar (only in fresh water)” – whose fresh water habitats may merit designation as Special Areas of Conservation (“SAC”). Not all such habitats need be designated – Member States have a discretion to select some and not others.96 The Dromagowlane/Trafrask River system in which wild salmon are found, and which were the subject of evidence in this case, are not SACs so designated for their salmon habitats. And that nondesignation is not challenged in these proceedings. Nor is Bantry Bay so designated – not could it be as it is a saltwater environment. AA arises only as to effects on the integrity of SACs97 having regard to their conservation objectives. The wild Atlantic Salmon is not a species strictly protected under Article 12 and Annex IV of the Habitats Directive. Accordingly it does not fall within the particular mention of the importance of strictly protected species in EIA which is articulated in Namur-Est98 and is found in Recital

11 and Article 3.1.b of the EIA Directive 2014. Though, by Article 14 and Annex V of the Habitats Directive, its taking and exploitation in freshwater may be subject to management99 in the present case, while any likely significant effects on wild Atlantic Salmon require assessment in EIA, they do not arise as requiring AA Screening or AA or as requiring strict protection under Article 14 of the Habitats Directive.

35. The Brown/Sea Trout – “Salmo Trutta” – is not listed in the Habitats Directive at all so no question AA or strict protection arises in that regard.

36. Perhaps surprisingly, as it is critically endangered across Europe,100 and its recruitment is highly sensitive to environmental conditions, the legal position of the FwPM is essentially the same as that of the wild Atlantic Salmon. The FwPM is listed in Annex II of the Habitats Directive but the Dromagowlane/Trafrask River system is not designated an SAC on their account. The FwPM is not listed in Annex IV of the Habitats Directive as requiring strict protection under Article 12 – though, by Article 14 and Annex V of the Habitats Directive, its taking and exploitation in freshwater may be subject to management. Accordingly and like the wild salmon the FwPM does not fall within the particular mention of the importance of strictly protected species in EIA which is articulated in Namur-Est101 and is found in Recital 11 and Article 3.1.b of the EIA Directive 2014.

37. The Courts are used to litigation as to SACs designated as such by reason of their FwPM habitats such that AA of projects which might affect them is required. But this is not such a case. Accordingly, and perhaps a little unusually in the experience of the courts, in this particular case, risk to the FwPM is not assessable in AA and is assessable only in EIA.

96 Sweetman v An Bord Pleanála [2016] IEHC 374. The limits of that discretion need not detain us here.

97 And, as is also relevant to the case but not at this point, special protection areas designated under the Wild Birds Directive.

98 Case C-463/20, Namur-Est Environnement ASBL v Région wallonne. Opinion of Kokott AG of 21 October 2021.

99 to ensure that the taking of specimens in the wild and their exploitation is compatible with their being maintained at a favourable conservation status.

100 sEIS p71.

101 Case C-463/20, Namur-Est Environnement ASBL v Région wallonne. Opinion of Kokott AG of 21 October 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

38. Given that EIA must be comprehensive,102 the foregoing is not at all to suggest that the criterion of significance of risk to the FwPM and the consideration of such risks are not crucially informed by its being a critically endangered species. Likewise, concerns as to the diminution in its populations will inform the requirement of comprehensive EIA as to risks to wild salmon.

39. However it is important to remember that EIA is essentially procedural as it relates to the making of development consent decisions. As the Supreme Court said in FitzPatrick,103 EIA “informs, rather than determines, the planning decision which should or may be made.” Its purpose is to ensure that the development consent decision informed by EIA is taken with full knowledge and public awareness of the likely significant environmental effects of the project. What is required is a development consent decision made with the fullest environmental knowledge, consistent with the precautionary principle and the purpose of the EIA Directive that projects may be designed such that they have the least possible effect on the environment – FitzPatrick. It must be comprehensive and draw reasoned conclusions on the risks of significant environmental impact. And these conclusions must be taken into account in deciding whether and in what terms to grant a development consent – Case C-261/18.104 EIA law requires that significant adverse environmental effects be mitigated. But, once these requirements of lawfulness105 are satisfied, and though EIA may result in refusal in a particular case as a matter of the judgment of the competent authority that the residual risk is unacceptable, EIA law does not preclude consent for a project likely to cause significant harm to the environment if the decision-maker rationally considers that the residual risk of harm acceptable. There is no rule of EIA that residual significant adverse environmental effect requires refusal of permission. And within those requirements of lawfulness, it is the competent decision-maker who gets to decide what risk is acceptable or unacceptable – even if others vehemently, and rationally, disagree.

40. The net result of the foregoing position as to environmental assessment is that, in this case and as to any risk to wild Salmon, the Brown/Sea Trout or the FwPM, only EIA applies. And as to such risks and given the essentially procedural nature of EIA just described above, as long as a comprehensive EIA is done and is

102 “as complete an assessment as possible” – Case C-50/09 Commission v Ireland, EU:C:2011:109, §§35, 37-41; Case C-404/09 Commission v Spain,

Opinion of Kokott AG, 28 June 2011; Case C-463/20 Namur-Est Environnement ASBL v Région wallonne, Opinion of Kokott AG of 21 October 2021.

Fitzpatrick v An Bord Pleanála & Apple [2019] 3 IR 617 – citing Connelly v An Bord Pleanála [2018] IESC 31, [2018] 2 I.L.R.M. 453 and Kelly v An Bord

Pleanála [2014] IEHC 400.

104 Case C-261/18 European Commission v Ireland, Opinion of Pitruzzella of 13 June 2019. “The main purpose of this assessment is to gather the

information necessary to enable the competent authorities of the Member States to identify, during the development consent procedure for such

projects, the environmental aspects liable to be adversely affected and thus make an informed decision on whether to grant or refuse the relevant

planning consents. The logic underlying Directive 85/337 is undoubtedly the prevention of environmental damage, and, as part of that logic, the obligation

to carry out a prior assessment of the environmental effects of a project is justified by the fact that, at a decision-making level, it is necessary for the

competent authorities to take effects on the environment into account at the earliest possible stage in all the technical planning and decision-making

processes, with a view to preventing ‘the creation of pollution or nuisances at source, rather than subsequently trying to counteract their effects’.

However, it is also apparent from the text of the directive that the environmental impact assessment is also intended to enable the competent authorities

of the Member States, in the event of development consent being granted, to make the consent subject to compliance with conditions that reduce the

adverse effects of the project on the environment and, more generally, to take measures to ensure that the resulting structure is used in accordance

with criteria of sound environmental management.”

105 There are other requirements of lawfulness – for example as to public participation – but they need not be listed here.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

considered by ALAB in granting the Impugned Licence, ALAB may licence the Shot Head salmon farm if it rationally considers any risk to the wild Atlantic Salmon, the Brown/Sea Trout and/or the FwPM acceptable.

Some Practical Notes

41. The Aquaculture Licence is in somewhat unusual form. It lists many terms and conditions arranged by topic in numbered paragraphs to which reference can easily be made. These appear to be standard conditions. But it contains also a “Schedule 5” to which no reference is made in the body of the licence to which it is scheduled. Schedule 5 is headed “Additional conditions applicable to the licence”. It is not clear why that format is used. If these were conditions highly specific to the licence in question, as opposed to standard conditions found in the body of the licence, clerical convenience might explain recourse to the Schedule. But in fact, many of the conditions in Schedule 5 seem more or less generic or standard. While this is not a fundamental issue, it would certainly ease the comprehension and understanding of the licence if its conditions were consistently grouped thematically by topic and the logic of the use of such a schedule were expressed in the body of the licence.

42. More practically, Schedule 5 contains at least 25 separate conditions106 distinguished, one from another, only by the use of bullet points. While, as this judgment will illustrate, bullet points are very useful in small numbers, it would have assisted in easily identifying, referring to and citing the Schedule 5 conditions had they been numbered or the like. In this particular context, pagination of the licence would also have assisted.

43. For reasons no longer relevant, the pleadings in all 3 proceedings have been repeatedly amended. What follows, and all references to pleadings in this judgment, reflect latest iterations.

44. It should be borne in mind that, while I have endeavoured to segment this judgment by topic, in appreciable degree the grounds and topics overlap as between the three sets of proceedings.

LEGISLATION

45. It is impractical to set out all relevant statutory provisions at this point. However, the following are notable for purposes of this judgment. Though of considerable importance in the case, I will leave any account of the Water Framework Directive to the section of the judgment addressing the case in that regard.

106 Arguably more – depending how you count them.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Fisheries (Amendment) Act 1997

46. Given the multifarious grounds of judicial review at issue and the centrality of the 1997 Act to the regulation of aquaculture and the appeal process, many provisions of the 1997 Act are relevant to the Proceedings – as is, in a more general sense, the overall statutory scheme of the 1997 Act. So it is necessary to set out the relevant terms of the 1997 Act at greater length than might otherwise be desirable in a judgment.

1997 Act107 - Section Relevant content & notes thereon108 S.3 - Interpretation • “aquaculture” means the culture or farming of any species of fish, …..

• “aquaculture licence” means a licence granted under section 14 to engage in aquaculture or operations in relation to aquaculture, ….

• “licensing authority” means (a) the Minister, … or (c) ALAB .. S.6 - Regulation of Engaging in unlicensed aquaculture is an offence. aquaculture. S.7 - Aquaculture licences. The licensing authority may licence aquaculture at a specified place or in specified waters if satisfied that it is in the public interest to do so.

There is a general power to impose conditions on licences - without prejudice to which, conditions may address listed specific matters, including:

• Specification of boundaries or limits of the place or waters.

• Limits on stocking.

• Operational practices, including fallowing.

• Record-keeping.

• Measures for preventing escapes of fish.

• Protection of the environment and control of discharges.

• Environmental, water quality and biological monitoring.

Note: Of some significance, “licensing authority” encompasses both the Minister and ALAB.109 S.10 - Licence Applications. Ministerial regulations shall govern licensing procedures and, inter alia, require EIA and provide for public and prescribed bodies’ participation in the licensing process. S.12 - Determination of The licensing authority may licence applications. (a) grant a licence or a variation of a licence, or

(b) refuse to grant a licence.

107 As amended.

108 Not verbatim. Irrelevant content omitted.

109 s.3 Interpretation.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1997 Act107 - Section Relevant content & notes thereon108 S.13 - Period for The Minister shall endeavour to determine a licence application within determination of licence four months from the date on which all applicable application applications. regulations requirements have been met.

Where it appears to the Minister that it would not be possible or appropriate, because of the particular circumstances of the application, to determine it within those four months, the Minister must notify all participants of the reasons why that would not be possible or appropriate and shall specify the date before which the Minister intends that the application shall be determined.

The Minister shall then take such steps as are open to ensure that the application is determined before that date. S.14 - Granting licences. Where the Minister’s determination to grant an aquaculture licence is appealed to ALAB, the Minister shall not grant the licence pending determination of the appeal. S.15 - Duration of licences. An aquaculture licence shall specify its duration – not exceeding 20 years. S.16 - Effect of licence. Aquaculture licences license licensees to carry on, in accordance with the licence, such operations as are specified in the licence. S.22 - ALAB. S.22 establishes ALAB. Ss.23, 24 & 26. The Minister appoints ALAB members from nominees of prescribed Membership, Tenure & organisations which, in the Minister’s opinion, are110 Remuneration of ALAB. • concerned with promotion of, or representative of persons carrying on, aquaculture;

• concerned with conservation, development and protection of wild fisheries;

• representative of professions or occupations relating to physical planning and development;

• representative of persons concerned with the protection and preservation of the environment and amenities;

• concerned with general economic development;

• concerned with community development.

ALAB members

• are appointed for a maximum of 5 years.

• may be re-appointed for a second or subsequent term.

110 Descriptions somewhat abbreviated.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1997 Act107 - Section Relevant content & notes thereon108

• are paid such remuneration as the Minister determines. S.26 - Efficient ALAB’s chairperson is to ensure the efficient performance of its performance of ALAB functions. functions. S.28 - ALAB Meetings & ALAB makes decisions by majority vote. procedure S.31 Communications with ALAB members and staff for the purpose of influencing it improperly as to its consideration and decision of an appeal are prohibited.

ALAB members must notify ALAB of any such communication and must not entertain it. S.32 - Secretary of ALAB. The secretary to ALAB is an officer of the Minister – a seconded civil servant. Ss.33 & 34 - Declaration of Declarations of interests relevant generally to the functions of ALAB or to interests. particular matters considered by ALAB are required of its members, secretary, employees, consultants and advisers.

Recusal from decisions involving conflict of interest is required.

The register of interests is open to public inspection. S.35 - Consultants and ALAB may engage consultants and advisers. They are to be identified in advisers. ALAB’s annual report. S.35C - Provision of services To enable ALAB to perform its functions, the Minister may provide by Minister to ALAB and by services (including staff) to ALAB, on such terms and conditions ALAB to Minister. (including payment for such services) as may be agreed.

ALAB may likewise provide services to the Minister. Ss.37 & 38 - Accounts & ALAB is to keep accounts which the Comptroller and Auditor General is Reports to audit and report.

ALAB reports annually, and also on request, to the Minister. S.40 - Appeals. (1) & (3) A person aggrieved by the Minister’s decision of an aquaculture licence application has 1 month from its publication to appeal to ALAB.

ALAB may not consider late appeals.

(4) (b) ALAB shall determine the appeal as if the licence application had been made to ALAB in the first instance.

(6) ALAB’s decision of an appeal immediately annuls the Minister’s

decision.

(8) (a) ALAB’s decision of an appeal shall state the main reasons and considerations on which it is based.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1997 Act107 - Section Relevant content & notes thereon108 S.41 - Provisions as to appeals. (1) The validity of an appeal depends, inter alia, on its stating in full the grounds of the appeal and the reasons, considerations and arguments on which they are based.

Appeals shall be accompanied by such documents, particulars or other information as the appellant considers appropriate.

(3) Appellants may not elaborate on or make further submissions on or submit further documents, particulars or other information on their appeals and ALAB shall not consider any such.111 S.43 - Submission of (1) ALAB shall, as soon as practicable, send a copy of the notice of appeal documents, etc., to ALAB by to the Minister. Minister. (2) The Minister shall, within 14 days thereafter submit to ALAB copies of

(a) the aquaculture licence application and of any drawings, maps, particulars, evidence, EIS, other written study or further information received from the applicant for the licence,

(b) any report prepared for the Minister as to the licence application,

(c) the Minister’s decision of the licence application.

Note: S.43 clearly implies that, while the appeal is determined de novo, ALAB may have regard to the Minister’s decision and the documents which informed it. S.44 - Submissions or (1) ALAB shall, as soon as practicable send a copy of the notice of appeal observations by other parties to the other parties to the appeal. to appeal. (2) & (4) The Minister and each other party except the appellant have 30 days to respond. ALAB shall not consider late responses, elaborations on responses or further submissions.

Note: S.44 clearly implies that the Minister may make submissions to ALAB as to how it should determine an appeal. S.45 - Submissions or ALAB shall set a period112 within which non-parties can make observations by persons other submissions. Save as to EIA, ALAB shall not consider late submissions, than parties to appeal. elaborations on submissions or further submissions by non-parties. S.46 - Power of ALAB to ALAB may solicit from a party to the appeal or anyone who has made request submissions or submission in the appeal, submissions as to any matter which has arisen observations. in relation to the appeal, where ALAB is of the opinion that, in the particular circumstances of an appeal, it is appropriate in the interests of justice to do so.

ALAB must specify a time limit for response and state that failing response within the time-limit it will determine the appeal without further notice.

111 Without prejudice to sections 46 & 47.

112 Not less than 30 days from its publication for the notice of appeal.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1997 Act107 - Section Relevant content & notes thereon108 Again, elaboration and further submission is prohibited. S.47 - Power of ALAB to Where ALAB is of the opinion that any document, particulars or other require submission of information is or are necessary to enable it to determine an appeal, it documents, etc. shall serve on a party or any person who has made submissions a notice,

(a) requiring specified documents, particulars or other information within a specific time limit

(b) stating that failing their provision within that time-limit, it will determine the appeal without further notice.

Failure to comply with such a notice is an offence. S.48 - Powers of ALAB where At any time after a s.46 or s.47 notice has expired, ALAB may without s.46 or s.47 notice served. further notice, determine the appeal.

Note: On a strict interpretation, there is an internal conflict within s.47 and between s.47 and s.48. ALAB determines that the information is necessary to determine the appeal. Yet, failing its provision within the set time limit, it is nonetheless to determine the appeal.

It would seem to follow that the necessity cannot be absolute. S.49 - Oral hearings. ALAB has absolute discretion to hold an oral hearing. S.50 - Other matters may be ALAB, in determining an appeal, may take into account relevant113 taken into account in matters other than those raised by the parties or by any person who has determining appeals. made submissions but must invite their response114 to such matters.

Again, ALAB cannot consider late responses. S.56 - Duty and objective of (1) ALAB shall, as far as practicable, ensure that appeals are dealt with ALAB. and determined expeditiously and that all steps are taken to avoid unnecessary delays.

(2) Without limiting the generality of subsection (1) but subject to subsections (3) and (4), ALAB should endeavour to ensure that every appeal is determined within four months from its receipt of the appeal.

(3) Where it appears to ALAB that it would not be possible or appropriate, because of the particular circumstances of an appeal, to determine an appeal within that four months it shall, before the expiration of that four months,115

• notify the parties to the appeal and all persons who made submissions of the reasons why the appeal would not be determined within that four months and

113 Matters to which, under section 61, it may have regard.

114 Either in writing or at an oral hearing.

115 Emphasis added.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1997 Act107 - Section Relevant content & notes thereon108

• specify the date by which it intends that the appeal shall be determined.

(4) Where such a notice has been served, ALAB shall endeavour to ensure that the appeal is determined before the date specified. S.59 - Reports to the Board. Those who perform inspections or hold oral hearings shall report thereon and make recommendations to ALAB – which ALAB shall consider before determining the appeal S.61 - Matters to which The Minister and ALAB, in deciding a licence application or appeal, shall licensing authority shall have take account116 of, inter alia, regard in determining • the suitability of the place or waters for the aquaculture proposed, aquaculture licence • existing or potential other beneficial uses of and any statutory status117 applications and appeals. of the place or waters

• likely effects of the proposed aquaculture on the economy of the area in which it will take place,

• likely ecological effects and likely effects on the environment generally in that vicinity. S.62 - General policy The Minister may issue general directives as to policy in relation to directives aquaculture as (s)he considers necessary.

ALAB shall, in performing its functions, have regard to such directives. S.73 - Judicial review of ALAB • Questioning of an ALAB determination on an appeal must be by Decisions judicial review under O.84 RSC.

• The High Court, in determining such an application for judicial review, shall act as expeditiously as possible consistent with the administration of justice.

• The High Court’s determination of such an application for judicial review shall be final and no appeal shall lie there from save, o with the leave of the High Court, where its decision involves a point of law of exceptional public importance and appeal is desirable in the public interest, or o where the appeal involves a question of the constitutionality of a law. S.77 - Recapture of escaped • IFI may take such action as it considers necessary to recapture stock stock. which has escaped from a licensed facility.

• The Minister may authorise the licensee or any person to take specified action to recapture such stock.

116 as appropriate in the circumstances of the particular case.

117 including under a development plan under the Planning and Development Act 2000.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1997 Act107 - Section Relevant content & notes thereon108 S.82 - Application of The Minister, in considering a Foreshore Licence application made in Foreshore Acts to connection with aquaculture shall have regard to any licensing authority aquaculture. decision as to the aquaculture licence.

Foreshore Act 1933

47. Relevant content of the 1933 Act includes the following.

Foreshore Act 1933118 - Relevant content & notes thereon119 Section S.1B - ‘appropriate Generally, for Foreshore Act purposes, the appropriate Minister is the Minister Minister’ for the Environment, Community and Local Government.120 However, as to aquaculture, the appropriate Minister is the Minister for Agriculture, Fisheries and Food.121 Note: The practical effect is that the same minister issues both the foreshore licence and the aquaculture licence. S.3 - Power to grant (1) The appropriate Minister may, if of the opinion that it is in the public foreshore licences. interest to do so, grant a licence to any person to place or erect any articles, things, structures, or works in or on State foreshore or to use or occupy such foreshore for any purpose.

(1B) Where either appropriate Minister is considering granting a foreshore licence (s)he shall consult the other Minister before doing so.

(7) Foreshore licences shall contain such covenants, conditions, and agreements as the appropriate Minister shall consider proper or desirable in the public interest and shall agree upon with the person to whom such licence is granted. S.12 - Structures A District Judge may order the removal of any structure – including structures, unlawfully erected on articles or equipment used or capable of being used for aquaculture122 – State foreshore. erected on State foreshore without lawful authority. Non-compliance is an offence. Ss. 13A & 19A See below, under the heading “EIA Legislation”.

118 As amended.

119 Not verbatim. Irrelevant content omitted.

120 Now The Minister For Environment, Climate And Communications.

121 Now the Minister.

122 Fisheries (Amendment) Act 1997 (23/1997), s.67.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Foreshore Act 1933118 - Relevant content & notes thereon119 Section EIA of foreshore proposals.

EIA Legislation – General, Aquaculture & Foreshore Licensing

48. For convenience, if somewhat inaccurately as the EIA Directive 2014123 took effect as an amendment of the 2011 EIA Directive, I will refer to the to 2011 EIA Directive as amended in 2014 as the “EIA Directive 2014”. References in this judgment to “the EIA Directive” encompass both the 2011 and 2014 iterations – to which much is common.

49. By the EIA Directive, EIA is required to inform “development consent” for specified types of project likely to have significant effects on the environment by virtue, inter alia, of their nature, size or location. Article 1 of the EIA Directive defines “development consent” as “the decision of the competent authority or authorities which entitles the developer to proceed with the project”. Depending on circumstance, a given project may require more than one development consent, each of which triggers an EIA obligation.

50. §1(f) Annex II of the EIA Directive in its 1985,124 2011 and 2014 iterations listed and lists “intensive fish farming” as a type of project for which Articles 2 and 4(2) require EIA where the project is likely to have significant effects on the environment – automatically where above a threshold set in domestic legislation and, where below such threshold, by way of case-by-case analysis (EIA screening) determining the presence or absence of such likelihood.

51. Pursuant to §1(f) Annex II of the EIA Directive 1985, the 1989 EIA Regulations125 listed “Seawater salmonid breeding installations with an output which would exceed 100 tonnes per annum”. The 1999 EIA Amendment Regulations126 amended this to “Seawater fish breeding installations with an output which would exceed 100 tonnes per annum”. That description also appears in in Part 10, Art 93, Schedule 5 Part 2 §1(f) PDR 2001127 which, by S.13A of the 1933 Act, was applied to EIA of projects the subject of Foreshore Licence applications.

123 Directive 2014/52/EU on the assessment of the effects of certain public and private projects on the environment. It amended Directive 2011/92/EU

on the assessment of the effects of certain public and private projects on the environment. The time limit for transposition of the 2014 Directive expired

on 16 May 2017. I consider below the relevant transitional provisions of the 2014 EIA Directive.

124 Directive 85/337/EEC, on the assessment of the effects of certain public and private projects on the environment – superseded by the 2011 Directive.

125 European Communities (Environmental Impact Assessment) Regulations (1989. S.I. No. 349/1989) 1st Schedule Part II §1(f).

126 S.I. No. 93 of 1999.

127 Planning and Development Regulations 2001 (S.I. No. 600 of 2001).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

52. Articles 1(2)(g)(iv), 3 and 8a of the EIA Directive 2014 provide as follows:

• Article 1(2)(g)(iv) – requires that EIA include a reasoned conclusion on the significant environmental effects of the project, taking into account the competent authority’s128 examination of o The EIAR.129 o Any supplementary information provided by the developer. o Any information received through consultations with the public and public authorities. o Where appropriate, its own supplementary examination.

• Article 3 – states the obligation in EIA to “identify, describe and assess in an appropriate manner, in the light of each individual case, the direct and indirect significant effects of a project”.

• Article 8a – prescribes, inter alia, that a decision to grant development consent include the reasoned

conclusion referred to in Article 1(2)(g)(iv) and that the competent authority be satisfied that the reasoned

conclusion is still up-to-date when deciding to grant development consent.

53. Articles 5, 6, 8, 9, and 11 of the EIA Directive 2014 impose obligations as to

• 5 – EIAR content.

• 6 – participation of the public and of authorities likely to be concerned by the project by reason of their specific environmental responsibilities or local and regional competences.

• 8 – the EIA must be taken into account in the development consent procedure.

• 9 – publication of the development consent decision and the main reasons and considerations on which it is based.

• 11 – availability of judicial review.

54. Article 9a of the EIA Directive 2014 obliges Member States to ensure that its competent authorities perform their EIA duties “in an objective manner and do not find themselves in a situation giving rise to a conflict of interest”.

128 The competent authority is the authority conducting the EIA.

129 Environmental Impact Assessment Report to be prepared by the developer seeking development consent as part of the EIA Process. See EIA Directive

2014 Article 1(2)(g)(i) and Article 5. Before the 2014 Directive the equivalent document was an environmental Impact Statement – “EIS”.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

EIA & Foreshore Licences

Foreshore Act 1933130 Relevant content & notes thereon131

- Section S.1 – Definitions (EIA) S.1 of the 1933 Act defines EIA.

• That definition changed in 2021132 by way of Ireland’s belated transposition of the 2014 EIA Directive, which amended the 2011 EIA Directive.

• Before and after 2021 that definition included “examination, analysis and evaluation” to identify, describe and assess the direct and indirect effects of a proposed development.

• While the wording differs before and after 2021, in substance both versions require EIA in an appropriate manner, in light of each individual case and in accordance with the applicable EIA Directive.

• However, it is notable that in all iterations of EIA law since at least the 2011 EIA Directive, the intending developer must provide, inter alia and by way of an EIS/EIAR, o “at least: (a) a description of the project comprising information on the site, design and size of the project;”133 ▪ The 2014 Directive added “and other relevant features”.

o “1. A description of the project, including in particular: (a) a description of the physical characteristics of the whole project and the land-use requirements during the construction and operational phases;”134 ▪ The 2014 Directive added “a description of the location of the project”.

Note: S.13A, S.19A and S.19B generally prescribe the EIA process.

S.13A - EIA of certain Note:

proposals relating to • S.13A is lengthy and complex.135 Only a brief account is possible here.

the foreshore. • S.13A was inserted by the 1989 EIA Regulations136 and thereafter amended both before and after the licence application was made in the present case.

• As to EIA of projects the subject of foreshore licence applications, the following, inter alia, amended S.13A:

As amended.

131 Not verbatim. Irrelevant content omitted.

132 By European Union (Foreshore Act 1933) (Environmental Impact Assessment) (Amendment) Regulations 2021 (S.I. No. 145 of 2021), reg. 4(a), (b).

133 EIA Directive 2011 Article 5(3).

134 EIA Directive 2011 Annex IV.

135 In the LRC Consolidated version to 25/3/21 and including annotations, S.13A runs to almost 9 pages. 24 amendments are noted as having occurred in

1990, 1998, 1999, 2009, 2010, 2012 and 2021.

136 European Communities (Environmental Impact Assessment) Regulations 1989 (S.I. No. 349 of 1989) Regulation 13(c).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Foreshore Act 1933130 Relevant content & notes thereon131

- Section o effect was given to the 2011 EIA Directive by the Foreshore EIA Regulations of 2012137 and 2014138 o effect was given to the 2014 EIA Directive by the Foreshore EIA Regulations of 2021.139 13A(1)(a) The appropriate Minister140 shall, in accordance with paragraph (b), before decision of a “relevant application”, ensure EIA of projects likely to have significant effects on the environment by virtue, inter alia, of their nature, size or location.

(b)(i)(II) EIA is required where the proposed development would be of a class specified141 and exceeds the relevant quantity, area or other limit specified.

13A(5) defines “relevant application” as meaning applications under Sections 2, 3, 10 and 13 of the 1933 Act – including for a foreshore licence under s.3.

13A(6)142 states that “relevant application” “does not include an application for an aquaculture licence143 … that is accompanied by” an EIS.

Notes: I will return to this issue but it bears noting at this point that:

• Only such “relevant applications” require EIA.

• Schedule 5, Part 2, §1(f) PDR 2001 specifies “Seawater fish breeding installations with an output which would exceed 100 tonnes per annum.” A literal reading of S.13A(1)(a) & (b) would require that foreshore licences for such installations be informed by EIA.

• S.13A(6) is puzzling. It is not apparent that an aquaculture licence application could have fallen within the general definition of ‘relevant application’ so as to require explicit exclusion from it, as S.13A(5) defines “relevant application” in terms confined to applications under the Foreshore Act 1933 – which aquaculture licences are not.

• Accordingly, on one view of a literal interpretation of S.13A(6) it has no effect as it addresses an issue which simply never arises given the terms of S.13A(5).

• The State says its effect is to absolve the Minister of EIA when considering a foreshore licence application where an associated aquaculture licence application is subjected to EIA.

European Union (Environmental Impact Assessment) (Foreshore) Regulations 2012 (S.I. No. 433 of 2012).

138 European Union (Environmental Impact Assessment and Appropriate Assessment) (Foreshore) Regulations 2014. (S.I. No. 544 of 2014).

139 European Union (Foreshore Act 1933) (Environmental Impact Assessment) (Amendment) Regulations 2021 (S.I. No. 145 of 2021)

140 S.1B – ‘appropriate Minister’- Generally, for Foreshore Act purposes, the appropriate Minister is the Minister for the Environment, Community and

Local Government. As to aquaculture, the appropriate Minister is the Minister for Agriculture, Fisheries and Food.

141 in Schedule 5, Part 2, PDR 2001.

142 Inserted (30.09.2009) by European Communities (Foreshore) Regulations 2009 (S.I. No. 404 of 2009), reg. 3(b).

143 Within the meaning of the Fisheries (Amendment) Act 1997.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Foreshore Act 1933130 Relevant content & notes thereon131

- Section S.13A(1)(cc)144 Inter alia, an EIAR in a foreshore licence application must include a description of the proposed development comprising information on its site, design, size and other relevant features. S.13A(2) 13A(2), inter alia, provides for public consultation in EIA of a foreshore licence application. 19A. Procedure as to S.19A inter alia requires certain relevant • The foreshore licence applicant to publish notice of an application with which applications.145 an EIA has been submitted and of arrangements for public inspection thereof and to send the Minister, inter alia, a map marked so as to identify clearly the land or structure to which the application relates.

• The Minister to make information, including that map, available to the public.

EIA & Aquaculture Licences – 2014 Directive – Transitional Provisions & their Transposition

55. Article 3 of the Aquaculture Appeals EIA Regulations 2012146 generally required ALAB, in considering an appeal, to perform EIA within the meaning of the 2011 EIA Directive, of aquaculture likely to have significant effects on the environment by virtue, inter alia, of its nature, size or location. It specifically required EIA by ALAB of seawater salmonid breeding installations.147 That ALAB was required to perform EIA in the Appeal is not in dispute.

56. But there is dispute as to which of the 2011 and 2014 EIA Directives properly applied to the subject licence applications and also whether Ireland correctly transposed the transitional provisions of the 2014 EIA Directive.

57. Article 2 of the amending EIA Directive 2014 stipulated its transposition to domestic law by 16 May 2017 and Article 3 stipulated transitional provisions as follows:

144Inserted (25.03.2021) by European Union (Foreshore Act 1933) (Environmental Impact Assessment) (Amendment) Regulations 2021 (S.I. No. 145 of

2021), reg. 5(a)(ii)-(v), (b)(iii), (iv), (vi), (viii), (ix)(II).

145 Substituted (15.01.2010) by Foreshore and Dumping at Sea (Amendment) Act 2009 (39/2009), s.13, commenced as per s. 1(4).

146 Aquaculture Appeals (Environmental Impact Assessment) Regulations 2012 (S.I. No. 468 of 2012).

147 As specified in Regulation 5(1)(i) of the Aquaculture (Licence Application) Regulations 1998 (S.I. 236 of 1998) as amended by the Aquaculture (Licence

Application) (Amendment) Regulations 2010 (S.I. No. 280 of 2010). SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“1. Projects in respect of which the determination referred to in Article 4(2) of Directive 2011/92/EU148 was initiated before 16 May 2017 shall be subject to the obligations referred to in Article 4 of Directive 2011/92/EU prior to its amendment by this Directive.

2. Projects shall be subject to the obligations referred to in Article 3 and Articles 5 to 11 of Directive 2011/92/EU prior to its amendment by this Directive where, before 16 May 2017:

(a) the procedure regarding the opinion referred to in Article 5(2) of Directive 2011/92/EU149 was initiated; or

(b) the information referred to in Article 5(1) of Directive 2011/92/EU150 was provided.”

58. In the event, the 2014 Directive was, in Ireland, transposed late: as to Aquaculture Licensing by the 2019 Regulations151 amending the 2012 Regulations.152 Article 8 of the 2019 Regulations provided that the 2012 Regulations “will continue to apply to licence applications received on or before 16 May 2017 without the amendments contained in Regulations 3 to 7 of these Regulations.”

59. As to transitional provisions, an issue arises whether Article 8 of the 2019 Regulations properly reflects Article 3.2(b) of the amending 2014 Directive.153 The State says that given that Article 3.2(b) in effect refers to the provision of an EIS and as, where EIA arises, the EIS accompanies the licence application (and did in the present case), no difficulty arises.

Inland Fisheries Act 2010

IFA 2010154 - Relevant content & notes thereon155 Section S.6 - Inland S.6 establishes IFI to perform functions as to the inland waters of the State and the sea to Fisheries the 12-mile territorial limit.

Ireland S.7 - Functions The principal function of IFI is the protection, management and conservation of the of IFI. inland fisheries resource.

148 i.e. EIA Screening.

i.e. EIS Scoping.

150 i.e. Information to be provided by the developer in its EIS/EIAR.

151 Aquaculture Appeals (Environmental Impact Assessment) (Amendment) Regulations 2019 (S.I. No. 276 of 2019); European Union (Aquaculture

Appeals) (Environmental Impact Assessment) Regulations 2019 (S.I. No. 341 of 2019).

152 Aquaculture Appeals (Environmental Impact Assessment) Regulations 2012 (S.I. No. 468 of 2012).

153 i.e. Directive 2014/52/EU amending Directive 2011/92/EU on the assessment of the effects of certain public and private projects on the environment.

154 As amended.

155 Not verbatim. Irrelevant content omitted.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

IFA 2010154 - Relevant content & notes thereon155 Section Its general functions include to

• promote, support, facilitate and advise the Minister on the conservation, protection, management, marketing, development and improvement of inland fisheries, including sea angling,

• develop and advise the Minister on policy and national strategies relating to inland fisheries including sea angling, and

• ensure implementation and delivery of policy and strategies agreed with the Minister.

Its obligations include to

• protect the inland fisheries resource,

• promote and encourage the management, conservation, protection, development and improvement of the fisheries not in its possession or occupation,

• encourage and develop, inter alia, angling for salmon and trout,

• have regard in performing its functions to Habitats law156 and to the need for the sustainable development of the inland fisheries resource (including the conservation of fish and other species of fauna .. and the biodiversity of inland water ecosystems), S.8157 IFI is a statutory consultee on aquaculture licence applications.158 S.12 With a view to representing the public interest in inland fisheries matters, the qualifications for appointment to IFI include experience or capacity in aquaculture and/or environmental/biodiversity matters.

Habitats Regulations 2011

60. Relevant content of the Habitats Regulations 2011159 includes the following. The regulations were amended while MOWI’s applications were in train. I have included some such amendments below.

2011 Relevant content161 Regulations160 Art.2(1) - • “Appropriate Assessment” (“AA”) means AA as referred to in Article 6(3) of the Interpretation Habitats Directive;162

156 The section cites the European Communities (Natural Habitats) Regulations 1997 (S.I. No. 94 of 1997).

157 And Schedule 2, Part 7.

158 By amendment of Regulation 10(1) of the Aquaculture (Licence Application) Regulations 1998.

159 European Communities (Birds and Natural Habitats) Regulations 2011 (S.I. No. 477 of 2011 as amended).

160 As amended.

161 Not verbatim. Irrelevant content omitted.

162 Directive 92/43/EEC on the conservation of natural habitats and of wild fauna and flora (as amended).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

2011 Relevant content161 Regulations160

• “Natura Impact Statement” (“NIS”) means a report comprising the scientific examination of a plan or project and the relevant European Site or European Sites, to identify and characterise any possible implications of the plan or project individually or in combination with other plans or projects in view of the conservation objectives of the site or sites, and any further information including, but not limited to, any plans, maps or drawings, scientific information or data required to enable the carrying out of an Appropriate Assessment; Art.42 - (1) Subject to Regulation 42A,163 AA screening of a project for which an application for AA Screening & consent is received, and which is not directly connected with or necessary to the AA management of a European Site, shall be carried out by the public authority to assess, in view of best scientific knowledge and of the conservation objectives of the site, if that project, individually or in combination with other plans or projects is likely to have a significant effect on the European site.

(2) AA screening shall precede consent for a project.

(3) At any time after an application for consent for a project,164 a public authority may direct the applicant to furnish —

(a) an NIS

(b) any additional information the public authority considers necessary for the purposes of this Regulation.

(4) (Failure to submit NIS – application deemed withdrawn)

(5)(a) An NIS shall, in addition to addressing the issues referred to in the interpretation contained in Regulation 2(1), include such information or data as the public authority considers necessary, and specifies, to enable it to ascertain if the plan or project will affect the integrity of the site.

(5)(b) (NIS to include any

• alternative solutions considered and reasons why they were not adopted,

• imperative reasons of overriding public interest to proceed despite adverse effects,

• compensatory measures proposed).

(6) The public authority shall determine that an AA is required where, after screening, it cannot be excluded, on the basis of objective scientific information that the project, individually or in combination with other plans or projects, will have a significant effect on a European site.

(7) The public authority shall determine that AA is not required where it can be excluded on the basis of objective scientific information following screening, that the

163 Inserted by Regulation 6(a) of European Union (Birds and Natural Habitats) (Amendment) Regulations 2021 (S.I. No. 293 of 2021).

164 Emphasis added.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

2011 Relevant content161 Regulations160 plan or project, individually or in combination with other plans or projects, will have a significant effect on a European site.

(8)(a) (The public authority shall give notice of a determination that AA is required, including its reasons, to (i) the applicant, (ii) ….. any person who made submissions or observations … (iii) …. any party to an appeal or referral.)

(b) (Where AA is required the public authority may direct in that notice that an NIS is required.)

(11) AA shall include a determination, pursuant to Article 6(3) of the Habitats Directive, whether a project would adversely affect the integrity of a European site and the AA shall be done before a decision is taken to approve, undertake or adopt a project.

(12) In AA the public authority shall take into account …—

(a) the NIS,

(b) any other plans or projects that may, in combination with the plan or project under consideration, adversely affect the integrity of a European Site,

(c) any supplemental information furnished,

(d) if appropriate, any additional information sought by the authority and furnished by the applicant in relation to a NIS,

(e) any information or advice obtained by the public authority,

(f) if appropriate, any written submissions or observations made to the public authority in relation to the application for consent for proposed plan or project,

(g) any other relevant information.

(13) A public authority may, for the purposes of AA and if it considers it appropriate, invite the opinion of the general public and, if it does so, it shall take such steps for that purpose as it considers necessary.165

(13) (a) Before AA is determined, the public authority shall carry out a public consultation and publish a notice of the proposed plan or project ….

(b) (Details to be set out in notice)

(c) (Application documents to be available to the public)

(d) (Public entitled to make submissions or observations)

(e) (Public authority to have regard to such submissions or observations) 166

(16) A public authority shall consent to a project only having determined that it shall not adversely affect the integrity of a European site.

(17) Subject to any other provision of these Regulations —

(a) Consent for a project may be given where a public authority has modified the project or attached conditions to the consent where the authority is satisfied to do so

165 Replaced by Regulation 6(b) of S.I. No. 293 of 2021 – European Union (Birds and Natural Habitats) (Amendment) Regulations 2021.

166 Inserted by Regulation 6(b) of S.I. No. 293 of 2021 – European Union (Birds and Natural Habitats) (Amendment) Regulations 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

2011 Relevant content161 Regulations160 having determined that the project would not adversely affect the integrity of the European site if effected in accordance with the consent and the said modifications or conditions.

(b) A public authority shall not consent to a project containing conditions, restrictions or requirements purporting to —

(i) permit deferral of the completion of AA screening or AA until after the consent has been given,

(ii) accept an incomplete NIS, or

(iii) permit or facilitate avoidance of compliance with Article 6(4) of the Habitats Directive.

(18)(a) A public authority shall make available for inspection any determination that it makes in relation to a project and provide reasons for that determination.

(24) As to a project likely to affect more than one European Site, the AA screening and AA shall address the impact of the project, individually or in combination with other plans and projects, on each of the sites likely to be affected.

SWI PROCEEDINGS – OUTLINE OF PLEADINGS.

SWI – Reliefs Sought

61. By its last amended Statement of Grounds,167 SWI seeks to quash both ALAB’s determination of 29 June 2021 to grant the Aquaculture Licence and, consequentially, the Aquaculture Licence dated 26 January 2022. By trial it became apparent that various pleaded grounds were not pursued.

62. SWI also impugned various legislative provisions as follows, seeking declarations accordingly:

• S.22 of the 1997 Act - as unconstitutional. However, despite an insufficiently thoroughgoing process of amendment of the statement of grounds, it is adequately clear that the challenge is not to s.22, which establishes ALAB, but to s.23 which provides for the composition of and appointment to ALAB. The essential point is one of structural bias.

167 Dated 20 December 2022.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• SWI CG9d - Schedule 5 Table 9 of the Environmental Objectives (Surface Waters) Regulations 2009168 as ultra vires in setting a nutrient limit of 170 µg/l for nitrogen in high quality status coastal waters – as incompatible with Annex V, Table 1.2.4 WFD.169 This was not pursued.

• Article 19 of the Control of Dangerous Substances in Aquaculture Regulations 2008170 – as ultra vires the Minister. This was not pursued.

• The EQS171 for Emamectin Benzoate172 (“EmBz”) set by the Minister – as invalid in effecting requirements of the WFD and the EQS Directive 2008173 as set without giving them unquestionable binding force contrary to Article 4(3) TEU174 and Article 288 TFEU.175 This was not pursued.

63. SWI pleads Core Grounds,176 in effect 21, which I summarise as follows:

SWI Grounds – Domestic

• SWI CG3177 – ALAB o failed to determine the Appeal as though the aquaculture licence application had been made to it in the first instance – contrary to S.40(4) of the 1997 Act. o fettered its discretion in the determination of the Appeal. o failed to finalise the conditions it intended to impose. o Breached its duty to state reasons for its decision – contrary to s.40(8) of the 1997 Act.

SWI CG3 also pleads s.40(5) of the 1997 Act, the relevance of which is not apparent.

• SWI CG4 – ALAB

168 European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009). Schedule 5 sets criteria for calculating

surface water ecological status and ecological potential. Table 9 sets Physico-chemical conditions supporting the biological elements. Under the heading

“Nutrient Conditions”, it sets limits for Dissolved Inorganic Nitrogen. Table 9 was replaced by the European Union Environmental Objectives (Surface

Waters) (Amendment) Regulations 2019 (S.I. No. 77 of 2019).

169 Annex V – 1.2. Normative definitions of ecological status classifications – Table 1.2. General definition for … coastal waters – 1.2.4. Definitions for high,

good and moderate ecological status in coastal waters. Physico-chemical quality elements – General conditions – High Status – “Nutrient concentrations

remain within the range normally associated with undisturbed conditions.”

170 European Communities (Control Of Dangerous Substances In Aquaculture) Regulations 2008 (S.I. No. 466 of 2008). Article 19 empowers the Minister

to establish water quality standards in respect of a particular substance or class of substances respect of an area subject to an aquaculture licence, or an

adjacent area. Different standards may be established for different substances or classes of substance and for different areas.

171 Environmental Quality Standard.

172 Emamectin benzoate is, as used in aquaculture, a chemical pesticide used to kill sea lice. It is commonly known by its trade name “Slice”.

Directive 2008/105/EC on environmental quality standards in the field of water policy.

174 The Treaty on European Union – Article 4(3) cites the principle of sincere cooperation and requires that Member States take any appropriate measure,

general or particular, to ensure fulfilment of the obligations arising out of the Treaties or resulting from the acts of the institutions of the Union.

175 The Treaty on the Functioning of the European Union – Article 288 states inter alia that a directive shall be binding, as to the result to be achieved,

upon each Member State to which it is addressed, but shall leave to the national authorities the choice of form and methods.

176 Due to their thematic rearrangement in amendment of the statement of grounds, SWI’s Core Grounds are no longer in numerical order.

177 “SWI” stands for Salmon Watch Ireland. “CG3 stands for “Core Ground 3”. The number “3” is taken from the Statement of Grounds. Like coding is

used in what follows. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o repeatedly requested further information which was not necessary for the purpose of determining the appeal, contrary to s.47 of the 1997 Act.178 o failed to determine the Appeal expeditiously contrary to s.56 of the 1997 Act.

o adopted a procedure that was not fair and equitable, contrary to the Aarhus Convention Article 9(4).

• SWI CG8 – The Impugned Decision is invalid for impossibility as ALAB authorised asynchronous stocking of fish farms in Bantry Bay but required compliance with DAFM Monitoring Protocol No. 3 For Offshore Finfish Farms – Sea Lice Monitoring and Control, which requires single bay management and synchronous fallowing.

• SWI CG12a – ALAB is structurally biased as o it is appointed in accordance with s.23 of the 1997 Act which is invalid as breaching the principles of constitutional justice. o by s.44 of the 1997 Act, the Minister is involved in the appeal against his own decision before a Board which he appointed. o the Minister’s policy in favour of aquaculture renders him objectively biased in the appointment of the Board.

SWI Grounds – EU Law

• SWI CG5 – ALAB did its EIA under the 2011 EIA Directive, contrary to Article 3 of the amending 2014 EIA Directive and Article 3 TEU, even though the information required to be submitted pursuant to Article 5(1)

of the EIA Directive was not submitted until after 16 May 2017. ALAB thereby failed to comply with Articles 1(2)(g)(iv),179 3 and 8a of the 2011 EIA Directive as amended by 2014 EIA Directive.

• SWI CG6 – ALAB’s EIA failed to o comply with Article 3 of the Aquaculture (Licence Application) Regulations 1998180 as required by Article 3(1) of the Aquaculture Appeals EIA Regulations 2012, (i.e. the definition of EIA181) interpreted in accordance with Recital 2 and Articles 1, 2, 3 of the EIA Directive, as to the risk from ▪ sea lice, given MOWI’s assertion that they are neutrally buoyant.

178 47.—(1) Where ALAB is of the opinion that any document, particulars or other information is or are necessary to enable it to determine an appeal,

………….

179 The Grounds actually cite “Article 2(iv)(g)”. There is no such Article and its citation is clearly a typo. The identity of the relevant Article 1(2)(g)(iv) is

readily discernible.

180 S.I. No. 236 of 1998.

181 S.I. No. 468 of 2012. There is no Article 3(1) of the 2012 Regulations. Article 3(a) amends Article 3(1) of the 1998 Regulations by inserting a definition

of EIA as “an assessment, to include an examination, analysis and evaluation, carried out by the Minister that shall identify, describe and assess in an

appropriate manner, in light of each individual case and in accordance with Articles 4 to 11 of Council Directive No. 2011/92/EU on the assessment of

the effects of certain public and private projects on the environment, the direct and indirect effects of a proposed development ….”

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

▪ amoebic gill disease. (Amoebic gill disease was not mentioned in written submissions and received only passing mention at trial. I will not consider it further.)

o state adequate reasons contrary to s.40(8) of the 1997 Act.

• SWI CG7 – ALAB failed to assess the structural adequacy of salmon cages to prevent escapes, contrary to o Article 6(3) of the Habitats Directive, as implemented by Article 42(8) of the Habitats Regulations 2011.182 o Articles 2, 3 and 8a of the EIA Directive.183

• SWI CG9a.1 – ALAB o contrary to Article 4 WFD, allowed a discharge to waters that would ▪ breach “emission limits” for nitrogen and EmBz, and ▪ eliminate the benthic community beneath the fish farm, and

• SWI CG9a.2 – ALAB o contrary to Article 4 WFD, allowed a discharge to waters without ▪ providing for an appropriate mixing zone for the area of non-compliance as required pursuant to Article 4 of the EQS Directive 2008184 and / or Article 51 of the Environmental Objectives (Surface Waters) Regulations 2009,185 or ▪ considering whether to provide for such a zone.

I will disregard this ground as ALAB and MOWI have clarified that the licence provides for no mixing zones.

I will elaborate briefly later in this judgment.

• SWI CG9b – Any mixing zone applied was not applied in accordance with Article 51 of the Environmental Objectives (Surface Waters) Regulations 2009 which purports to implement Article 4 of the EQS Directive

182 European Communities (Birds and Natural Habitats) Regulations 2011 as amended. Article 42(8) reads:

(8)(a) Where, in relation to a plan or project for which an application for consent has been received, a public authority makes a determination that an

Appropriate Assessment is required, the public authority shall give notice of the determination, including reasons for the determination of the public

authority, to the following—

(i) the applicant,

(ii) if appropriate, any person who made submissions or observations in relation to the application to the public authority, or

(iii) if appropriate, any party to an appeal or referral.

(b) Where a public authority has determined that an Appropriate Assessment is required in respect of a proposed development it may direct in the notice

issued under subparagraph (a) that a Natura Impact Statement is required.

183 or their precursors, as implemented by Regulation 3 of the Aquaculture (Licence Application) Regulations 1998 (S.I. No. 236 of 1998) as amended by

the Aquaculture (Licence Application) (Amendment) Regulations 2010 (S.I. No. 280 of 2010), Aquaculture (Licence Application) (Amendment) (No. 2)

Regulations 2010 (S.I. No. 369 of 2010) and Aquaculture (Licence Application) (Amendment) Regulations 2012 (S.I. No. 301 of 2012) and European Union

(Environmental Impact Assessment) (Aquaculture) Regulations 2012 (S.I. No. 410 of 2012) and / or as required by Article 3(1) of the Aquaculture Appeals

(Environmental Impact Assessment) Regulations 2012 (S.I. No. 468 of 2012).

184 Directive 2008/105/EC on environmental quality standards in the field of water policy.

185 European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

2008.186 I will disregard this ground as ALAB and MOWI have clarified that the licence provides for no mixing zones.

• SWI CG12b – ALAB is appointed according to a formula that, o favours the grant of licences over their refusal, o gives rise to an inability to perform its duties in an objective manner, or without the appearance of a conflict of interest, and thereby contravenes Article 9a of the EIA Directive187 and Article 47 CFREU.188 So, all ALAB acts breach Article 288 TFEU189 and Article 4(3) TEU.190

• SWI CG16 – The Impugned Decision is invalid for irrationality and/or inadequacy of reasons, as based on a misrepresentation or clear error of fact as to breach of EQSs for inorganic nitrogen.

SWI Grounds – Validity

• SWI CG9c – Article 51 of the Environmental Objectives (Surface Waters) Regulations 2009191 fails to adequately transpose Article 4 of the EQS Directive 2008192 and establish a proper and lawful mixing zone regime, contrary to Article 288 TFEU and Article 4(3) TEU. I will disregard this ground as ALAB and MOWI have clarified that the licence provides for no mixing zones.

• SWI CG12c – ALAB is structurally biased, and s.23 of the 1997 Act193 breaches, o the principles of constitutional justice,

186 Directive 2008/105/EC on environmental quality standards in the field of water policy.

187 Art 9a provides in part: “Member States shall ensure that the competent authority or authorities perform the duties arising from this Directive in an

objective manner and do not find themselves in a situation giving rise to a conflict of interest.”

188 Charter on Fundamental Rights of the European Union. Article 47 – Right to an effective remedy and to a fair trial

Everyone whose rights and freedoms guaranteed by the law of the Union are violated has the right to an effective remedy before a tribunal in compliance

with the conditions laid down in this Article.

Everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal previously established by law. Everyone

shall have the possibility of being advised, defended and represented.

Legal aid shall be made available to those who lack sufficient resources in so far as such aid is necessary to ensure effective access to justice.

The Treaty on the Functioning of the European Union. Art 288 in part reads: A directive shall be binding, as to the result to be achieved, upon each

Member State to which it is addressed, but shall leave to the national authorities the choice of form and methods.

190 The Treaty on the European Union – Art 4.3 reads:

3. Pursuant to the principle of sincere cooperation, the Union and the Member States shall, in full mutual respect, assist each other in carrying out tasks

which flow from the Treaties.

The Member States shall take any appropriate measure, general or particular, to ensure fulfilment of the obligations arising out of the Treaties or resulting

from the acts of the institutions of the Union.

The Member States shall facilitate the achievement of the Union's tasks and refrain from any measure which could jeopardise the attainment of the

Union's objectives.

191 European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009). Article 51 relates to mixing zones.

192 Directive 2008/105/EC on environmental quality standards in the field of water policy. The Grounds refer to Directive 2008/15 but as that is irrelevant

as relating to standards and procedures for returning illegally staying third-country nationals, the typo is clear – as, from the context, is the intended

citation of Directive 2008/105/EC.

193 The Minister appoints ALAB members.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o Articles 34.1,194 40.3.1,195 and 15.4.2196 of the Constitution, o Article 9a of the EIA Directive,197 due to conflict of interest.

• SWI CG12d – ALAB is appointed according to a formula that, o favours the grant of licences over their refusal, o gives rise to an inability to perform its duties in an objective manner, or without the appearance of a conflict of interest, and thereby contravenes Article 9a of the EIA Directive and Article 47 CFREU,198 such that S.23 of the 1997 Act must be set aside for breach Article 288 TFEU and Article 4(3) TEU.

IFI PROCEEDINGS – OUTLINE OF PLEADINGS.

IFI – Reliefs Sought

64. By its last amended Statement of Grounds,199 IFI seeks to quash,

• ALAB’s determination of 29 June 2021 to grant the Aquaculture Licence.

• the Aquaculture Licence dated 26 January 2022.

• the Foreshore Licence granted by the Minister.

It does so on, in effect, 12 Core Grounds which I summarise as follows:

IFI Grounds – Domestic

• IFI CG1 – ALAB purported to determine the Appeal de novo but in fact determined it by amending the Minister’s determination and draft licence.

Art 34.1: Justice shall be administered in courts established by law by judges appointed in the manner provided by this constitution, and, save in such

special and limited cases as may be prescribed by law, shall be administered in public.

195 Art 40.3.1: The State guarantees in its laws to respect, and, as far as practicable, by its laws to defend and vindicate the personal rights of the citizen.

196 Art 15.4.2: Every law enacted by the Oireachtas which is in any respect repugnant to this Constitution or to any provision thereof, shall, but to the

extent only of such repugnancy, be invalid.

197 Art 9a provides in part: “Member States shall ensure that the competent authority or authorities perform the duties arising from this Directive in an

objective manner and do not find themselves in a situation giving rise to a conflict of interest.”

198 Charter on Fundamental Rights of the European Union. Article 47 – Right to an effective remedy and to a fair trial

Everyone whose rights and freedoms guaranteed by the law of the Union are violated has the right to an effective remedy before a tribunal in compliance

with the conditions laid down in this Article.

Everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal previously established by law. Everyone

shall have the possibility of being advised, defended and represented.

Legal aid shall be made available to those who lack sufficient resources in so far as such aid is necessary to ensure effective access to justice.

199 Dated November 2022.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• IFI CG2 – The 6 years ALAB took to determine the Appeal was excessive and ALAB failed to give reasons for its extensions of time.

• IFI CG3 – ALAB’s failed, in breach of natural justice and irrationally, to circulate for response, reports as to

the presence of wild salmonids in the Dromagoulane/Trafrask River and their role in the propagation of the FwPM population of that river. ALAB fell into serious error as to the diminution in the population of migratory wild salmonids in that river and resultant impact on that FwPM population.

• IFI CG4 – ALAB’s irrational reliance on factual and scientific conclusions without proper factual or scientific

basis in the face of contrary evidence, including: o A sea lice dispersal model erroneously premised on the neutral buoyancy of sea lice. o A conclusion of no negative impact on wild salmonids – for which it failed to give adequate reasons. o Failure to require single or synchronous bay management.

• IFI CG5 – ALAB’s conduct of the appeal evinced objective bias by o the large number of time extensions granted, o directing the production of further documents, o bespeaking additional reports, thereby collecting the material to justify its pre-determined decision to grant the Aquaculture Licence.200

• IFI CG6 & CG7.1 – The Aquaculture Licence should be quashed as contingent on ALAB’s determination of the Appeal. The Foreshore Licence should in turn be quashed as contingent on the Aquaculture Licence.

• IFI CG7.2 – The Minister failed to consult the Minister for Housing, Local Government and Heritage prior to issuing the Foreshore Licence. This ground was barely advanced at trial by no more than a formal mention.

It does not feature in IFI’s written submissions. I reject it as, in substance, not argued.

• IFI CG7.3 & 7.4 – The Minister, in granting the Foreshore Licence in 2022. o erred in relying on the Ministerial determination of September 2015 as the basis of the grant o in relying on that determination as the basis of the grant, failed to have regard to a relevant consideration, being ALAB’s decision of 2021 to grant the Aquaculture Licence for an aquaculture activity changed from that which the Minster had decided to licence in 2015. o failed to have regard to the public interest or to record his consideration thereof and give reasons accordingly.

200 The Statement of Grounds says “determined to grant the appeal” but the correct meaning is clear in context.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

IFI Core Grounds – EU Law

• IFI CG9 – ALAB failed to perform a valid EIA in failing to assess cumulative dangers to marine life.

• IFI CG10 – ALAB failed to apply the precautionary principle and S.61 of the 1997 Act.201

SWEETMAN PROCEEDINGS – OUTLINE OF PLEADINGS

PS202 – Reliefs Sought

65. By their last amended Statement of Grounds,203 the Sweetman Applicants seek to quash:

• The Minister’s determination on 18 September 2015204 to grant MOWI an aquaculture licence.

• ALAB’s like decision on appeal made on 29 June 2021.

• The Minister’s determination on 15 September 2015205 to grant MOWI a foreshore Licence.

• The Foreshore Licence issued by the Minister on 30 March 2022.

66. The Sweetman Applicants also seek a declaration that Article 8 of the Aquaculture Appeals (EIA) (Amendment) Regulations 2019206 (which addresses the transition from the 2011 to the 2014 EIS Directive) is invalid as incompatible with the State’s obligations, under Arts 2(1) and 3 of the EIA Directive 2014, to apply Articles 3 and 5 to 11 of the EIA Directive 2014 to projects as to which, at 16 May 2017, scoping under Article 5(3) had not been initiated and the information referred to in Article 5(1) had not been provided.

201 ALAB, in deciding a licence application or appeal, shall take account of the suitability of the place or waters for the aquaculture proposed, existing or

potential other beneficial uses of and any statutory status of the place or waters, likely effects of the proposed aquaculture on the economy of the area

in which it will take place, likely ecological effects and likely effects on the environment generally in that vicinity.

Peter Sweetman.

203 Pursuant to order made 21 November 2022.

204 Note – in fact made 5 September 2015.

205 Note – in fact made 5 September 2015.

206 Aquaculture Appeals (Environmental Impact Assessment) (Amendment) Regulations 2019 (S.I. No. 276 of 2019). These regulations purport to

transpose requirements of the 2014 EIA Directive by amending the “Principal Regulations” being the Aquaculture Appeals (Environmental Impact

Assessment) Regulations 2012 (S.I. No. 468 of 2012). Article 8 states: “The Principal Regulations will continue to apply to licence applications received on

or before 16 May 2017 without the amendments contained in Regulations 3 to 7 of these Regulations.” SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

PS Core Grounds – Domestic

• PS CG3 – The Aquaculture Licence is invalid as ALAB, in November 2015, accepted the RPS Water Quality Report 2015 from MOWI contrary to Ss.40(4),207 41(3)&(4),208 and/or s.44(2)&(4)209 of the 1997 Act.

• PS CG4 – The Aquaculture Licence is invalid and contrary to s.47(1) of the 1997 Act210 in that: o ALAB recorded in its minutes its opinion that documents relating to the escape of 230,000 fish from a salmon farm in Bantry Bay in 2014 were necessary to enable it to determine the appeal. o Therefore ALAB was obliged to and did serve a s.47 notice requiring the Minister to submit those documents to it. o ALAB withdrew the notice before the documents were provided. o The decision to withdraw the notice was unreasonable and unreasoned.

• PS CG5 – The Aquaculture Licence is invalid as made following a process which ALAB extended unreasonably to more than 5.5 years (68 months), contrary to its duty set out in s.56 of the 1997 Act to ensure, as far as practicable, that appeals are dealt with and determined expeditiously and that all steps are taken to avoid unnecessary delays.

• PS CG1 – The Aquaculture Licence is invalid and contrary to S.40(4) of the 1997 Act211 in that ALAB determined the Appeal without determining it as if the licence application had been made to ALAB in the first instance.212

• PS CG2 – If the Court finds that ALAB’s actions negated annulment under s.40(6) of the 1997 Act213 of the Minister’s decision as to the Aquaculture Licence, then the Minister’s decision of 18 September 2015214 as to the Aquaculture Licence is invalid and was made contrary to fair procedures and natural justice and the public participation rights afforded by EU law.215

207 ALAB shall determine the appeal as if the licence application had been made to ALAB in the first instance.

208 Appellants may not elaborate on or make further submissions on or submit further documents, particulars or other information on their appeals and

the Board shall not consider any such.

209 The other parties to the appeal have 30 days from notice of the appeal to respond. ALAB shall not consider late responses, elaborations on responses

or further submissions.

210 “Where the Board is of the opinion that any document, particulars or other information is or are necessary for the purpose of enabling it to

determine an appeal, it shall” require its submission to the Board.

211 ALAB shall determine the appeal as if the licence application had been made to ALAB in the first instance.

212 I have omitted reference to irrelevant elements of this Ground as pleaded.

213 “(6) The determination …. of an appeal shall annul the decision or action of the Minister immediately the determination is made.”

214 In fact, 5 September 2015.

215 I have omitted reference to irrelevant elements of this Ground as pleaded.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

PS Core Grounds – EU Law

• PS CG6 – ALAB breached Article 3(5) and 3(6)216 of the Aquaculture Appeals EIA Regulations 2012,217 as it failed to require MOWI to produce information necessary to correct inadequacies in the EIS. o The particulars of this plea amount merely to a statement that “much of the information that normally would be provided to an EIA process by the person seeking the consent was actually provided by the decision maker, often through the offices of the MI who are shared advisors with the Department on these appeals.” I reject this ground as inadequately pleaded.

• PS CG7 – ALAB relied on an NIS that did not comply with Articles 2(1) and 42(5)(a)218 of the Habitats Regulations 2011219 o The particulars of this plea in essence complain that the AA did not include consideration of effects on the Storm Petrel or the Common Seal or site-specific bird counts or in-combination effects with other aquaculture and future kelp harvesting.

• PS CG8 & 9 – ALAB failed, in breach of Article 42(1) 220 and 42(6) 221 of the Habitats Regulations 2011 to o ensure proper AA Screening. o determine that AA was required where it could not be excluded on AA Screening and on the basis of objective scientific information that the Salmon Farm, individually or in combination with other plans or projects, would have a significant effect on the Glengarriff Harbour and Woodland SAC and the Bull and Cow Rocks SPA.

216 The grounds refer to Articles 5 and 6. But the typo and its correction are evident: These 2012 regulations have only 3 articles. The only substantive

article is Article 3. Article 3 (5) & (6) read as follows:

(5) The Board shall require the production by the appellant of any additional or supplemental information that it considers necessary to enable it to make

an assessment.

(6) …. If the environmental impact statement (and other material) is inadequate, then the Board shall serve a notice (hereinafter ‘a request for further

information’) which sets out the manner in which the information is inadequate and requires the appellant to submit further information to remedy

these inadequacies.

217 Aquaculture Appeals (Environmental Impact Assessment) Regulations 2012 (S.I. No. 468 of 2012).

2(1) – “Natura Impact Statement” means a report comprising the scientific examination of a plan or project and the relevant European Site or European

Sites, to identify and characterise any possible implications of the plan or project individually or in combination with other plans or projects in view of the

conservation objectives of the site or sites, and any further information including, but not limited to, any plans, maps or drawings, scientific information

or data required to enable the carrying out of an Appropriate Assessment;

42(5)(a) – A Natura Impact Statement shall, in addition to addressing the issues referred to in the interpretation contained in Regulation 2(1), include

such information or data as the public authority considers necessary, and specifies in a notice given under paragraph (3), to enable it to ascertain if the

plan or project will affect the integrity of the site.

219 European Communities Birds and Natural Habitats Regulations 2011 (S.I. No. 477 of 2011).

220 Article 42(1) states the obligation to carry out AA Screening “to assess, in view of best scientific knowledge and in view of the conservation objectives

of the site, if that plan or project, individually or in combination with other plans or projects is likely to have a significant effect on the European site.”

221 “(6) The public authority shall determine that an Appropriate Assessment of a plan or project is required where the plan or project is not directly

connected with or necessary to the management of the site as a European Site and if it cannot be excluded, on the basis of objective scientific information

following screening under this Regulation, that the plan or project, individually or in combination with other plans or projects, will have a significant effect

on a European site.” SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• PS CG9A – The 18-cage project envisaged by the Aquaculture Licence and for which the Foreshore Licence issued in March 2022 is not the same project as the 12-cage project for which the Minister carried out EIA or AA screening, such that in issuing the Foreshore Licence, the Minister breached, o The EIA Directive as transposed into s.13A(1)(a)&(b) of the Foreshore Act, 1933. o The Habitats Directive as transposed into Article 42(6) of the Habitats Regulations 2011.

• PS CG9B222 – The Impugned Foreshore Decisions are invalid as: o the Minister failed, in breach of s.3(1) of the 1933 Act, to consider properly or at all whether it was in the public interest to grant a foreshore licence for 18 cages occupying 8.82 hectares and occupying together with their anchors and moorings 24.6 hectares, being 60% of the proposed licensed foreshore area, in circumstances where the Minister’s determination and draft foreshore licence of 2015 was for 12 cages occupying 5.88 hectares and occupying together with their anchors and moorings 19.20 hectares, being 45% of the proposed licensed foreshore area.223 o contrary to Article 6(2) of the EIA Directive224 as transposed to s.13A(2)225 and s.19A226 of the 1933 Act, there was no public consultation as to the environmental impacts of the larger development. o and any public consultation in relation to the larger development during the aquaculture licence appeal process could not have been taken into account in the foreshore development consent process to September 2015, contrary to Article 8 of the EIA Directive227 and s.13A(2) of the 1933 Act. o Any EIA associated with the Foreshore Licence issued in March 2022 failed to comply with Annex IV of the EIA Directive228 as transposed into s.13A(1)(cc)229 and 13A(2AC)230 of the 1933 Act in that there was no description of the physical characteristics of the whole project, including land-use requirements.

• PS CG9C231 – the Impugned Foreshore Decisions breach s.3(1) of the 1933 Act232 as o the Minister purported to grant a foreshore licence on 15 September 2015, o without authorising MOWI to place any material or to place or erect any articles, things, structures, or works in or on such foreshore,

222 Note, in its original form this ground consisted of a single sentence of no less than 354 words. I have had difficulty understanding it and hope I have

succeeded in rationalising its meaning somewhat.

223 The proposed licensed foreshore area stayed constant as between both the 12-cage and the 18-cage proposal.

224 “…. to ensure the effective participation of the public concerned ….”

inter alia, provides for public consultation in EIA of a Foreshore licence application.

226 Inter alia requires,

The Applicant to publish of notice of a foreshore licence application with which an EIA has been submitted and of arrangements for public inspection

thereof and to send the Minister, inter alia, a map marked so as to identify clearly the land or structure to which the application relates.

The Minister to make information, including that map, available to the public.

227 Article 8 – The results of consultations and the information gathered pursuant to Articles 5 to 7 shall be duly taken into account in the development

consent procedure.

228 Content of EIS/EIAR.

229 The grounds cite s.13A(CC), but it is clear in context that intended reference is to s.13A(1)(cc) as to the necessary content of an EIAR.

230 An applicant for a foreshore licence for sub-threshold development requiring EIA may describe any features of the development or measures envisaged

to avoid or prevent significant adverse effects on the environment.

231 Note, in its original form this ground consisted of a single sentence of no less than 242 words. I have had great difficulty understanding it and hope,

but am not confident, that I have succeeded in rationalising its meaning somewhat.

232 Ministerial power to grant foreshore licences.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o and the purported authorisation to do such things did not occur until the decision to issue the foreshore licence made on 30 March 2022 or alternatively until the commencement of the Impugned Foreshore Licence on 26 January 2022.

o S.3(1) of the 1933 Act “does not provide for the granting of a foreshore licence without authority or the authorisation of activities on the foreshore without the Minister granting a licence for those activities and any attempt to do so was an impermissible delegation of Ministerial powers”, o and “one effect of this was that the Foreshore Licence was in fact not granted with authority until more than 6 years after its environmental impacts were assessed”, o “and the public was not promptly informed of the content of the decision and any conditions attached thereto and the main reasons and considerations on which the decision was based”, o contrary to Article 9 of the EIA Directive233 as transposed by s.13A(2)(d)234 of the Foreshore Act 1933 as amended.

• PS CG9D – The Impugned Foreshore Decisions breach of s.13A of the 1933 Act235 and o Article 2(1) of the EIA Directive and as the Minister failed to ensure that EIA preceded his decision on the foreshore licence application. o Article 9 of the EIA Directive and as the Minister failed to inform the public of the main reasons and considerations for that decision.

• PS CG9E & 13A – The Impugned Foreshore Decisions breach Article 6(3) of the Habitats Directive and Article 42(1) of the Habitats Regulations 2011 as the Minister failed to screen for AA the project to place

18 cages occupying 8.82 hectares together with their anchors and moorings occupying an overall area of

24.6 hectares of foreshore being 60% of the proposed licensed foreshore area.

• PS CG9F – The Impugned Foreshore Decisions breach of s.82 of the 1997 Act,236 because the Minister, in considering the foreshore licence application, failed to have regard to ALAB’s decision as to the Aquaculture Licence. As a result, and contrary to Article 9 of the EIA Directive, the Minister failed to inform the public of the main reasons and considerations on which the decision to grant the foreshore licence was based, and/or thereby failed, contrary to Article 11 of the EIA Directive, to facilitate access to a review procedure to challenge237 the substantive or procedural legality of decisions, acts or omissions.

Obligation to publicise decision to grant or refuse development consent including the main reasons and considerations on which it is based.

234 The grounds cite s.13A(d) but it is clear in context that intended reference is to s.13A(2)(d) which reads “Where the appropriate Minister makes a

decision … he or she shall … inform the … public thereof and publish … a notice stating —(i) that the appropriate Minister has made a decision to grant

or, as the case may be, refuse consent for the relevant application, (ii) the main reasons and considerations on which the decision … is based, …”

235 13A – EIA of proposals as to the foreshore.

236 The Minister, in considering a Foreshore Licence application made in connection with aquaculture shall have regard to any licensing authority decision

as to the aquaculture licence.

237 Before a court of law or another independent and impartial body established by law.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• PS CG9G – The Impugned Foreshore Licence issued on 30 March 2022 is invalid as purportedly replacing an earlier version that contained a so called ‘clerical error’. The Minister had no power to issue a replacement licence in this manner or without public participation, contrary to fair procedures and the public participation provisions of the EIA Directive.

PS Core Grounds 10 – 13 – Habitats Directive

ALAB’s Impugned Aquaculture Licence decision is in breach of Article 6(3) of the Habitats Directive as transposed by

• PS CG10 – Art 42(1) of the Habitats Regulations 2011 by reason of failing to conduct AA screening.

• PS CG12a – Art 42(1) of the Habitats Regulations 2011 by conducting AA screening which took account of measures intended to avoid or reduce harmful noise effects of the project on seals, a conservation interest of the Glengarriff Harbour and Woodland SAC, foraging outside the SAC.

• PS CG11 – Art 2(1) of the Habitats Regulations 2011 by reason of failing to conduct AA.

• PS CG12b – Art 2(1) of the Habitats Regulations 2011 by granting development consent without being certain that its conditions guarantee that Acoustic Deterrent Devices permitted outside the current consent process will not adversely affect the integrity of the Glengarriff Harbour and Woodland SAC.

• PS CG13 – Art 2(1) of the Habitats Regulations 2011, in failing in its AA to make an explicit and detailed statement of reasons, capable of dispelling all reasonable scientific doubt as to the effects of the Proposed Development, for rejecting, o Dr Gittings’ findings as to the effect of navigational lighting on Storm Petrels from the Bull and The Cow Rocks SPA. o Mr Coram’s findings as to the possibility of acoustic deterrents causing hearing damage to seals from the Glengarriff Harbour and Woodland SAC.

PS Core Ground – WFD

• PS CG14 – the Aquaculture and Foreshore licence decisions are in breach of o Art 4(1)(a)(i) to (iii) WFD in that ALAB and the Minister failed to consider properly or at all if the project may, ▪ cause a deterioration of the status of Adrigole Harbour, and ▪ otherwise jeopardise the attainment of good surface water status or good ecological potential and good surface water chemical status.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o Article 4(A) of the Surface Waters Regulations 2009238 in that ALAB failed to take all reasonable steps to ensure that protected areas239 achieve compliance with any applicable standards and WFD objectives.

PS Core Grounds – EIA Directive

• PS CG15 – ALAB’s determination of June 2021 to grant the Aquaculture Licence, in assessing the project under the 2011 EIA Directive,240 breaches Articles 3 and 5 to 11 of the 2014 EIA Directive241 as no scoping opinion was given and the information referred to in Article 5(1) of the EIA Directive242 was not provided before 16 May 2017, so the 2014 Directive applied.243

• PS CG16 – ALAB’s determination of June 2021 to grant the Aquaculture Licence, breaches Articles 3 and 5 to 11 of the 2011 EIA Directive.

It is as well so set out the particulars of that plea here as PS CG16 is uninformative. The Sweetman Applicants allege failure to adequately assess, o impacts sea lice from the fish farm on wild salmon or to require MOWI to submit site specific studies.244 o impacts on the FwPM or to require MOWI to collect site specific information and submit reports.245 o impacts of a major accident such as a fish escape and to acquire the necessary information.246

238 European Communities Environmental Objectives (Surface Waters) Regulations 2009 as amended – including by the European Union Environmental

Objectives (Surface Waters) (Amendment) Regulations 2012 & 2019. Article 4 states that “A public authority shall, in so far as its functions allow, comply

with the requirements of these Regulations and take all reasonable steps including, where necessary, the implementation of programmes of measures

to: (a) ensure, in so far as its functions allow, that—

(i) surface water bodies comply with the relevant environmental quality standards specified in the Schedules contained in these Regulations, and

(ii) protected areas achieve compliance with any standards and objectives laid down for such areas at the latest by 22 December 2015 unless otherwise

specified in the national legislation under which the individual protected areas have been established.

239 European Communities Environmental Objectives (Surface Waters) Regulations 2009, Article 3: “protected areas” means areas designated as requiring

special protection under specific Community legislation for the protection of their surface water and groundwater or for the conservation of habitats and

species of European sites directly dependent on water and listed in the register established by the Agency in accordance with Article 8 of the 2003

Regulations. The Surface Waters Regulations 2009 do not identify what are the “2003 Regulations” but it is clear from the context that they are European

Communities (Water Policy) Regulations 2003 (S.I. No. 722 of 2003) made to give further effect to the WFD. Article 8 of the 2003 Regulations requires

the EPA to keep a register of protected areas in accordance with Article 6 WFD. Article 6 requires registration of all areas lying within each river basin

district which have been designated as requiring special protection under specific Community legislation for the protection of their surface water and

groundwater or for the conservation of habitats and species directly depending on water. The register must include bodies of water used for the

abstraction of water intended for human consumption within the meaning of Article 6 WFD and protected areas covered by Annex IV WFD – i.e., Rivers,

Lakes, Transitional waters, Coastal waters and Artificial and heavily modified surface water bodies.

240 i.e. the 2011 EIA Directive unamended by the amending 2014 EIA Directive.

241 i.e. the 2011 EIA Directive as amended by the amending 2014 EIA Directive.

242 i.e. both the 2011 and 2014 versions.

243 Article 3 of the 2014 EIA Directive provides transitional arrangements as between the 2011 and 2014 directives. Inter alia it provides that projects

will remain to be considered under Articles 3 and 5 to 11 of the 2011 Directive if before 16 May 2017, either an EIA scoping decision under Article 5(2)

had issued (not relevant here) or the information referred to in Article 5(1) of the 2011 Directive was provided. Article 5(1) stipulates the minimum

information to be contained in an EIS/EIAR. In its 2011 iteration, it included at least (a) a description of the project comprising information on the site,

design and size of the project; (b) a description of the measures envisaged in order to avoid, reduce and, if possible, remedy significant adverse effects;

(c) the data required to identify and assess the main effects which the project is likely to have on the environment; (d) .. outline of the main alternatives

studied by the developer …; (e) a non-technical summary of the information referred to in points (a) to (d) and also information identified in Annex IV.

244 Citing Art. 3(1)(a) 2011, Art. 3(1)(b) 2014, Art 5(1), Art 5(3).

245 Citing Art. 3(1)(a) 2011, Art. 3(1)(b) 2014, Art 5(1), Art 5(3).

246 Citing Art. 3(1), 3(2) 2014, Art 3(1) 2011.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o the structural integrity of the proposed structures and the failure to acquire the necessary information in that regard.247 o the impact of 18 cages when only 12 were applied for initially and only 12 were modelled in the RPS report 2015248 o the introduction of information not notified to the public. (I interpret this as a plea of unfair failure to circulate that information to the public for comment. This information is identified as249 ▪ the “Freshwater Pearl Mussel Report”250 – (the “O’Toole FwPM Report 2021”251).

▪ the DMP Statistical Solutions report of May 2021 on risk of Gannet mortality due to cage net entanglement.252 ▪ the “Kelp Report”253 – (the “O’Toole Kelp Report 2021”254).

o impacts of freshwater abstraction using well boats and the impacts of dumping used wash water from the well boats in the licensed area.255 (This was not pursued).

o impacts on human health.256

The Sweetman Applicants allege failure to adequately assess, failure to prepare a Non-Technical Summary of the EIS.257 – but this point is derivative of the substantive pleas set out above.

• PS CG17 – ALAB o erred in law and acted contrary to fair procedures by allowing a situation of objective bias, o acted contrary to Article 9(a) of the 2014 EIA Directive by failing to perform its duties arising from the EIA Directive in an objective manner and by performing them in a situation giving rise to a conflict of interest.

• PS CG17A – This repeats PSCG9A258 and PSCG9B259 above (Ministerial failure to consider whether the 18- cage project was in the public interest, to screen it for EIA and AA, to consult the public on it and to

247 Citing Art 3(1)(d) 2014, 3(1(c) 2011.

248 Article 3(1).

249 In fact it is identified by way of example but that is insufficient to render other breaches pleaded.

250 So identified in ALAB’s Determination.

251 As I will refer to it.

252 entitled “Population Viability Analysis of The Impacts of Additional Mortality Due to Fish Net Entanglement in Gannets from Bull and Cow Rocks SPA”.

253 So identified in ALAB’s Determination.

254 As I will refer to it.

Citing Article 3(1)(a) to (e ) 2014; Art 3(1) to (c ) 2011; Art 5(1) to 5(3).

256 Citing Article 3(1)(a) EIA Directive.

257 Art 5(1) 2014; Art 5(3) 2011.

258 PS CG9A – The 18-cage project envisaged by the Aquaculture Licence and for which the Foreshore Licence issued in March 2022 is not the same

project as the 12-cage project for which the Minister carried out EIA or AA screening such that in issuing the Foreshore Licence, the Minister breached.

The EIA Directive as transposed into s.13A(1)(a)&(b) of the Foreshore Act 1933. The Habitats Directive as transposed into Article 42(6) of the Habitats Regulations 2011.

259 PS CG9B – The impugned foreshore decisions are invalid as:

• the Minister failed, in breach of s.3(1) of the 1933 Act, to consider properly or at all whether it was in the public interest to grant a foreshore licence

for 18 cages occupying 8.82 hectares and occupying together with their anchors and moorings 24.6 hectares, being 60% of the proposed licensed

foreshore area, in circumstances where the Minister’s determination and draft foreshore licence of 2015 was for 12 cages occupying 5.88 hectares

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

consider it in EIA, as the 18-cage project was not considered in the determination of September 2015 on foot of which the Foreshore Licence issued in 2022.)

• PS CG17B – This repeats PSCG9C above. (The Minister purported to grant a foreshore licence on or about

15 September 2015 without authorising MOWI to develop on the foreshore, and the purported authorisation to develop did not occur until the Impugned Foreshore Licence issued in 2022, i.e. more than 6 years after its environmental impacts were assessed, such that the public was not promptly informed of the content of the decision and any conditions attached thereto and the main reasons and considerations on which it was based, contrary to Article 9 of the EIA Directive260 and s.13A(2)(d) of the Foreshore Act 1933.

• PS CG17C – This repeats PSCG9D above (Ministerial failure to ensure that the 18-cage project occupying

8.82 hectares of foreshore, with anchors and moorings occupying an overall area of foreshore amounting to 24.6 hectares, being 60% of the proposed licensed foreshore area was subjected to EIA contrary to Article 2(1) of the EIA Directive.

PS Core Ground – Transposition / Invalidity

• PS CG18 – Art 8 of the Aquaculture Appeals EIA Regulations 2019261 is incompatible with the State’s obligations under Arts. 2(1) 3 of the EIA Directive 2014 to apply Art. 3 and Arts. 5 to 11 of that Directive to projects where, before 16 May 2017: (a) the scoping opinion procedure of Art. 5(2) of EIA Directive 2011 had not been initiated; or (b) the information referred to in Article 5(1) of EIA Directive 2011 (i.e. the EIS)

had not been provided.

and occupying together with their anchors and moorings 19.20 hectares, being 45% of the proposed licensed foreshore area. contrary to Article

6(2) of the EIA Directive as transposed to s.13A(2) and s.19A of the 1933 Act, there was no public consultation as to the environmental impacts of

the larger development.

• and any public consultation in relation to the larger development during the aquaculture licence appeal process could not have been taken into

account in the foreshore development consent process to September 2015, contrary to Article 8 of the EIA Directive259 and s.13A(2) of the 1933

Act.

• Any EIA associated with the issuing of the foreshore licence in March 2022 failed to comply with Annex IV of the EIA Directive259 as transposed into

s.13A(CC)259 and 13A(2AC)259 of the 1933 Act in that there was no description of the physical characteristics of the whole project, including land-

use requirements.

260 Obligation to publicise decision to grant or refuse development consent including the main reasons and considerations on which it is based.

261 Aquaculture Appeals (Environmental Impact Assessment) (Amendment) Regulations 2019 (S.I. No. 276 of 2019). Transitional Arrangements 8. The

Principal Regulations will continue to apply to licence applications received on or before 16 May 2017 without the amendments contained in Regulations

3 to 7 of these Regulations. (The Principal Regulations are the Aquaculture Appeals (Environmental Impact Assessment) Regulations 2012 (S.I. No. 468 of

2012)). SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

PLEADINGS IN JUDICIAL REVIEW

67. Despite the 63 Core Grounds and further complicating matters, numerous pleading objections were raised, some in written submissions, others orally at trial. At my request, after the trial had ended the parties co-operated in producing a table identifying the pleading objections and the documents, including the trial transcript references, relevant thereto. It will give some idea of the controversy to observe that, by my count and though there is perhaps a number of ways one could count them, no less than 31 pleading issues are identified in that schedule. I will address them as necessary in the contexts in which they rise. Meanwhile, and to reduce the need for their repeated repetition, I should say a little of the overarching rules of pleading in judicial review.

68. Order 84, rule 20(3) of the Rules of the Superior Courts, provides that:

“It shall not be sufficient for an applicant to give as any of his grounds for the purposes of paragraphs

(ii) or (iii) of sub-rule (2)(a) an assertion in general terms of the ground concerned, but the applicant should state precisely each such ground, giving particulars where appropriate, and identify in respect of each ground the facts or matters relied upon as supporting that ground”.

69. O.84, r.20(3) was inserted into the Rules following the judgment of the Supreme Court in AP v DPP. Murray CJ said that: “In the interests of the good administration of justice it is essential that a party applying for relief by way of judicial review set out clearly and precisely each and every ground upon which such relief is sought. The same applies to the various reliefs sought.” Denham J said that the order granting leave to seek judicial review “determines the parameters of the grounds upon which the application proceeds. The process requires the applicant to set out precisely the grounds upon which the application is to be advanced. On any such application the High Court has jurisdiction to allow an amendment of the statement of grounds, if it thinks fit. Once an application for leave to appeal has been granted the basis for the review by the court is established.” Hardiman J referred to the “absolute necessity for a precise defining of the grounds on which relief is sought”.

70. Murray CJ also observed that if at trial a new argument emerges, “The court of trial of course may, in the particular circumstances of the case, permit these matters to be argued, especially if the respondents consent, but in those circumstances the applicant should seek an order permitting any extended or new ground to be argued ...”. At trial in this case there was no application to amend the Statement of Grounds.

71. Denham J in AP held that the High Court judge had “addressed issues outside the grounds granted for the judicial review, in the absence of any order, or consent, to amend the statement of grounds. In this he

262 [2011] 1 IR 729.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

fell into error. A court, including this Court, is limited in a judicial review to the grounds ordered for the review

on the initial application, unless the grounds have been amended.”

72. Baker J in the Supreme Court held in Casey263 that in judicial review the pleadings,

• set the parameters, define and fix the issues in dispute between the parties and those to be determined by the court.

• define and limit the jurisdiction of the court to decide the issues pleaded.

• confine the evidence at trial to the matters relevant to those issues.

73. These rules have been considered in many cases since. In a Ballyboden TTG judicial review264 the Supreme Court cited Cooney v Browne265 as asking whether “the pleading in question is so general or so imprecise that the other side cannot know what case he will have to meet at the trial”. In ETI266 an account was given of the caselaw as to the strictness of the pleading rules in judicial review. Humphreys J returned to the issue in O’Donnell.267 General pleas will not ground a specific and detailed complaint. What is required,

on a “fair and reasonable” – not overly strict or narrow – reading, is clarity, precision and particularity such

that the plea is “acceptably clear”.268 If, on such an interpretation of the pleadings, a matter is found not to

have been pleaded, the rule is strict – it is excluded from consideration at trial. New grounds can be introduced only by amendment of the statement of grounds – that cannot be done merely by affidavit. A fortiori, it cannot be done for the first time in submissions, written or oral. These rules apply equally to opposition papers.

CONDITIONS IN LICENCES – VALIDITY & INTERPRETATION – GENERAL NOTE

Conditions – Certainty & Enforceability

74. Conditions in licences cannot merely be aspirational or lists of merely desirable actions or outcomes. They are not mere guidance, suggestions or recommendations. They must be in clear terms and take effect as certain, binding and enforceable legal obligations. These criteria, set out in guidance then-current, were

263 Casey v Minister for Housing [2021] IESC 42.

264 Ballyboden Tidy Towns Group v An Bord Pleanála [2022] IESC 47 (Supreme Court, Woulfe J, 15 November 2022).

265 Cooney v Browne [1984] I.R. 185, Henchy J.

266 Environmental Trust Ireland v An Bord Pleanála & Cloncaragh Investments [2022] IEHC 540 §211 et seq.

267 O’Donnell v An Bord Pleanála & Drumakilla [2023] IEHC 381.

268 E.g. St. Audoen’s National School v An Bord Pleanála [2021] IEHC 453, Sweetman v APB & Bord na Mona [2021] IEHC 390, Atlantic Diamond Ltd v An

Bord Pleanála [2021] IEHC 322, Sherwin v An Bord Pleanála & CWTC [2023] IEHC 26, Dunne & Ors v Kildare County Council & Ors [2023] IEHC 73.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

approved in Ashbourne.269 They have been repeated in similar guidance since as amongst the “basic criteria” on which imposition of planning conditions should be considered.270

75. Though the point may not have been made in these terms to date, the principle of legal certainty, notable as a general (though not absolute) principle of EU law and essential to the rule of law generally, illustrates why that is so. Legal certainty requires that law, and public law legal instruments creating permissions, privileges, licenses and the like, must be clear in their terms, and be reliable, predictable and enforceable in their effect. This is for the benefit of both the beneficiary of the instrument and of those others reliant on it to enforce its terms.271 A licensee is entitled to clarity as to what acts and omissions will and will not place it in breach of the terms and conditions of its licence and, potentially, at risk of criminal conviction of such breach272 and revocation of the licence.273 Also entitled to such clarity are those others – public bodies and others – such as those with standing to seek to enforce the obligations imposed on the licensee by the licence. Notably, this latter group includes those seeking to advance the high level of protection of the environment which is an aim of EU law and policy.274

76. Absolute certainty is impossible but, put simply and as far as practical, by the terms of any such licence all legitimately interested parties need in practice and are entitled in law to know where they stand. As Noonan J said in the planning context in Aherne,275 “Clearly a condition which is unenforceable must … be ultra vires the Board.” As Browne276 says: “... a condition must be precise. Breach of a condition creates a liability to enforcement action and thus it is only fair that a developer should know with certainty what it is that is required of him by the condition.” This is so not least as engaging in aquaculture other than in accordance with an aquaculture licence is a criminal offence – ss.6 and 65 of the 1997 Act. Legal certainty is not partisan: it benefits all. So, conditions must be clear and precise – as far as language can reasonably achieve.

77. It is vital that licensing and similar authorities keep this need for clarity and precision in mind in drafting licence conditions as, by imposing such conditions they may, intentionally or unintentionally and successfully or unsuccessfully and in any event confusingly, transmute or attempt to transmute the generally

269 Ashbourne Holdings Ltd v An Bord Pleanála & Cork County Council [2003] 2 IR 114.

270 Development Management Guidelines for Planning Authorities June, 2007 §7.3; OPR Practice Note PN03, Planning Conditions, October 2022. A similar

view is taken in the UK – DOE Circular 11/95 (WO 35/95) states that conditions should only be imposed where they are, inter alia, enforceable.

(Butterworths Planning Law Service/Division C Obtaining planning permission/Section 9 Conditions and limitations/A Conditions and limitations:

summary).

E.g. Case C-212/80 – Salumi.

272 S.65(1) of the 1997 Act provides that “A person who contravenes or fails to comply with a term of a licence or a condition subject to which it is

issued shall be guilty of an offence”.

273 S.68(1) of the 1997 Act provides inter alia that “….. the Minister may, in his or her discretion and, …. without compensation to the licensee, revoke

an aquaculture licence if the Minister (a) is satisfied that there has been a breach of any condition specified in the licence, …”

274 Article 191 TFEU.

275 Aherne v An Bord Pleanála [2015] IEHC 606 (High Court, Noonan J, 6 October 2015).

276 Simons on Planning Law, 3rd Ed’n, 2021 (Browne) §4-236.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

desirable and voluntary into legal obligations enforceable in criminal law.277 If they do so, it will be presumed intentional. The presumption of validity of administrative decisions implies interpretation of conditions, where possible, to avoid their invalidity278 and hence, so as to comply with the requirement that conditions be enforceable. The courts presume the licensor’s intention to impose, by its conditions, binding and enforceable legal obligations. All this requires very considerable care and thought in drafting licence conditions.

78. Browne279 says that the courts take a pragmatic approach to the interpretation of conditions and a condition will only be void for uncertainty if it can be given no sensible or ascertainable meaning, and not merely because it is ambiguous.

Conditions – Incorporation by Reference of Standards and the Like

79. Use, in drafting licence conditions, of the technique of incorporation of a document, “wholesale” as it were, by reference, is proper, convenient and useful. Nonetheless, great care is needed in drafting and imposing such conditions. They should not be imposed without careful consideration of the text of the document in question to ensure clarity as to exactly what legal obligations are to be imposed and are not to be imposed. See recently in this regard Fernleigh.280 The document – e.g. guidelines or protocols – may well not have been drafted, intended or promulgated to produce binding and enforceable legal obligations and may be expressed, for good reason, in more or less vague terms giving greater or lesser discretion or even in contemplation only of voluntary compliance. Often they are intended not to produce binding and enforceable legal obligations but are intended to state principles or guidelines to which decision-makers must merely have regard. The Department of Agriculture has produced many “Protocols” as to aquaculture on, it seems, a non-statutory basis and in terms an many respects understandably flexible and non-specific. The incorporation by reference in a condition of an inherently and properly imprecise document may well prove problematic. While it is tempting to incorporate guidelines, protocols and the like holus-bolus as reliably representing adopted policy and good practice, in truth, far from being a substitute for thought by the decision-maker, they require, often, greater thought. Such conditions can prove to have conveyed an illusory first impression of having imposed obligations and can result in confusion as to whether obligations have in fact been imposed. Such conditions can pose very particular difficulties in judicial review, where an interpretation of a condition is necessary to a decision – perhaps as to the validity of such condition. None of this is intended to discourage incorporation by reference – it is to emphasise the care required in doing so. In

277 In a somewhat different context, I described a similar effect in Jennings v An Bord Pleanála [2023] IEHC 14 §391 whereby SPPRs can transmute narrative

content of planning guidelines from “have regard to” obligations to obligations to comply and the consequent necessity of careful drafting of Planning

Guidelines and SPPRs.

278 Simons on Planning Law, 3rd Ed’n, (Browne) §4-237 says that “The courts take a pragmatic approach to the interpretation of conditions and it seems

that a condition will only be void for uncertainty if it can be given no meaning or no sensible or ascertainable meaning, and not merely because it is

ambiguous.” citing Irish Asphalt Ltd v An Bord Pleanála, unreported, High Court, Costello J, 28 July 1995 and Houlihan v An Bord Pleanála, unreported,

High Court, Murphy J., 4 October 1993.

279 Simons on Planning Law, 3rd Ed’n, (Browne) §4-237 – citing Irish Asphalt Ltd v An Bord Pleanála, unreported, High Court, Costello J., 28 July 1995.

280 Fernleigh Residents Association v An Bord Pleanála [2023] IEHC 525.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

that context one can only express concern at, specifically as to the enforceability, of Condition 12.1 of the Aquaculture Licence which requires that “The Licensee shall at all times comply with all .. Protocols applicable to aquaculture operations.”

Conditions – Severability

80. Though ALAB does not, MOWI relies281 on Condition 12.3 of the Aquaculture Licence, which reads as follows:

“If any condition or part of a condition in this licence is held to be illegal or unenforceable in whole or in part, such condition shall be deemed not to form part of this licence but the enforceability of the remainder of this licence is not affected.”

81. In my view the efficacy of such a condition is highly dubious. In effect it purports to deem each and every condition in a licence or permission unnecessary to its grant. Taken to absurdity it could validate an entirely unconditional licence. As a general proposition, conditions are imposed precisely because they are necessary to the grant and proper operation of the licence. That is not to say that particular conditions may not be severable. But if they are, it will be because that is discernible at law – not because the decisionmaker indiscriminately deems dispensable every condition it has imposed. Browne282 observes that

“Where a condition is ultra vires, it appears that the entire permission will be quashed where that condition is an essential part of the grant of permission and cannot be severed from the grant without materially altering its terms. …… “the whole planning permission must go; … The good part is so inextricably mixed up with the bad that the whole must go.”283 However …. the invalidity of a condition on occasion need not render the licence or permission invalid. That is so if the condition is severable from the licence or permission. Whether a condition is severable is decided on established legal principles. The key criterion is whether the impugned condition impacts on the character of the permission or whether the condition can be effectively severed without compromising the validity of the vestigial permission.”

82. Browne284 cites FPH Properties285 to the effect that unless it can be demonstrated that the decisionmaker would have granted the relevant permission subject only to the other conditions, if it had been

281 MOWI written submissions in SWI Proceedings.

282 Simons on Planning Law, 3rd ed’n (Browne), §4-241 et seq.

283 Citing Lord Reid in Kingsway Investments (Kent) Ltd v Kent County Council [1971] A.C. 72.

284 Simons on Planning Law, 3rd ed’n (Browne) §§4-244 & §249.

285 State (F.P.H. Properties S.A.) v An Bord Pleanála [1987] I.R. 698 at 711.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

advised that the impugned condition was invalid, the impugned condition is not severable. A court cannot rewrite the permission.286

83. In light of those principles of law, I cannot see how a decision-maker can regard and by condition deem, every condition it imposes on a licence as discretely severable from it. The resultant possible permutations of the terms of the effective licence would be unacceptably myriad, uncertain and unpredictable. ALAB was correct not to rely on Condition 12.3. More generally, I may as well say now that, as to the various licence conditions canvassed below, I cannot see that it can be reliably said that ALAB would have granted the licence without any of them and I will consider severability no further.

PART 2 – SPECIFIC GROUNDS OF CHALLENGE

AQUACULTURE LICENCE APPEAL – DE NOVO

84. ALAB’s decision was made explicitly pursuant to its jurisdiction under s.40(1)(b) of the 1997 Act to determine the Appeal as if the licence application had been made to ALAB “in the first instance”. All Applicants plead and submit that ALAB didn’t do so. The respondents reply, in essence, that this challenge is of form rather than substance.

85. Putting the “first instance” requirement another way, ALAB, if it exercises, as it did here, its jurisdiction under s.40(1)(b) of the 1997 Act, must decide the appeal “de novo” – in Hynes287 McGuinness J

held that, “The Board is required to determine the application as if it had been made to it in the first instance.

This means that it is determining the matter de novo and without regard to anything that had transpired before the Planning Authority.” In my view the authorities establish that while the phrase “without regard to anything that had transpired before the Planning Authority” may be correct and generally so as to de novo appeals, yields to the specific statutory scheme which may provide otherwise and did in the case of the 1997 Act. In that context, it does not seem to me that SWI’s reliance on Sliabh Luachra288 in this respect much adds to its case on this issue

86. In general terms, the obligation to decide de novo means that ALAB is bound in no way or degree by the Minister’s decision and must consider anew all aspects of the licence application and of the decision to be made on foot of it. What is essential to a de novo appeal is that all relevant issues are, as it were, up for

decision afresh and that the tribunal must approach them all with an open mind and a tabula rasa. In

286 It was concluded in that case that, having regard to the contents of the inspector’s report and the residual condition, the permission could not stand

with the conditions severed as this would be tantamount to rewriting the permission.

287 Hynes v An Bord Pleanála [1998] IEHC 127.

288 Sliabh Luachra Against Ballydesmond Windfarm Committee v Bord Pleanála [2019] IEHC 888, §42.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

particular, it must not regard itself as merely reviewing the decision appealed to see if it agrees with or should depart from it. That is a distinct, though connected, requirement from the identification of the materials which it may consider in coming to its de novo decision. However, all depends on the statutory scheme in this regard.

87. Before considering what a decision as if at first instance – de novo – amounts to, I should say that I accept the general proposition that a failure by ALAB to decide the Appeal as if at first instance would constitute breach of a mandatory statutory obligation and, ordinarily at least, certiorari would issue in consequence.289

88. Fitzgibbon290 was cited to me as to what a de novo appeal amounts to. It concerned the scope of an appeal to the High Court from a determination or direction against a solicitor by the Law Society under s.11(1) of the Solicitors (Amendment) Act 1994, as to which appeal the High Court may make “such order as it thinks fit”. The Supreme Court held that such appeals were not to be heard de novo. In so holding, Clarke J considered the various types of appeal commonly encountered. Importantly, the governing statute must prevail over such typology and the typology is often most relevant where, as is regrettably common, the statute is unclear as to the proper type and scope of appeal.

89. Clarke J identified the two critical characteristics of a de novo appeal as, first that the decision taken at first instance is “wholly irrelevant” and, second, that the appeal body is required to come to its own conclusions on the evidence and materials properly available to it. He observed that the evidence and materials which were properly before the first instance body are not “automatically” properly before the appeal body. The word “automatically” is notable. Importantly, Clarke J also considered that what happened at first instance is not necessarily entirely irrelevant or inadmissible in a de novo appeal. The default position in the case of a de novo appeal is that the parties will again present to the appellate body whatever evidence or materials they consider necessary for their case. But that is so “in the absence of any specific rule to the contrary”.291 Clarke J considered “In summary … that the use of the term “de novo appeal” or similar terminology, carries with it a requirement that the appellate body exercise its own judgment on the issues before it without any regard to the decision made by the first instance body against whom the appeal lies.” Notably, it is fresh “judgment on the issues” which is required.

90. It is important to repeat however that the foregoing principles are subject to any specific provisions of the statute governing the appeal. Such provisions of the 1997 Act are in my view significant. First, and as stated, s.40(1)(b) of the 1997 Act requires ALAB to determine the licence application as if it had been made

to ALAB “in the first instance” – that is “similar terminology”, to use Clarke J’s phrase, to “de novo”.

289 SWI correctly cites McAnenley v Bord Pleanála [2002] 2 IR 763, Environmental Trust Ireland v Bord Pleanála [2022] IEHC 540, §145.

290 Fitzgibbon v Law Society of Ireland [2016] 2 I.L.R.M.

291 And subject to certain other general exceptions not relevant to the present case.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

91. However, the 1997 Act also provides as follows:

i. That appellants “state in full the grounds of the appeal and the reasons, considerations and arguments on which they are based first”.292

It seems inevitable that such grounds will consist, at least in part, of a critique of the Minister’s

decision. Indeed grounds of appeal would be unnecessary (as, for example, they are unnecessary in an appeal from the Circuit Court in a civil matter to a de novo decision of the High Court) if the de novo nature of the appeal required merely that ALAB completely ignore the Minister’s decision. That an appellant be encouraged to refocus on the important issues by identifying grounds of appeal – perhaps abandoning some issues decided by the Minister and addressing new ones thrown up by that decision – can only enure to better and more efficient public administration and reduce waste of time.

ii. That the Minister will send to ALAB copies of293

▪ The aquaculture licence application and all documents which accompanied it, including any EIS and any further information received from the applicant.

▪ Any report prepared for the Minister in relation to the application.

That would seem to include not merely expert/technical reports but such as civil servants’ recommendations to the Minister as to the decision (s)he should make.

▪ Any document recording the decision of the Minister.

It is difficult to see why that would be required to be sent on if ALAB had to ignore it.

iii. That the Minister may make submissions or observations in writing to the Board in relation to the appeal.294

This may seem an unusual provision – for example, it would be thought strange, at least ordinarily, were a judge to make to a court considering an appeal of his/her judgment submissions seeking to

292 s.41(d) 1997 Act.

293 s.43(2) 1997 Act.

294 s.44(2) 1997 Act – though in this case the Minister made no such submissions, he did reply to s.46 notices on 27 April 2018 and 16 November 2020.

See also below as to the making and withdrawal by ALAB of a s.47 request to the Minister. See Affidavit of Ultan Waldron sworn 13 July 2013 in the SWI

Proceedings. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

have his/her judgment upheld. However that such a provision is unusual does not imply that it is inherently objectionable, much less repugnant to the Constitution. It is perhaps understandable in the Aquaculture Licensing process given the dual role of the Minister in both deciding licence applications and making and implementing policy relevant to the decision of licence applications.

Whether or not that is so, more importantly it is what the 1997 Act provides. It would be strange and not, as far as I can see, required by the 1997 Act, nor would it be practical or assist in teasing out the real issues, were the Minister obliged to refrain from framing his/her submissions, at least in part, as supporting his/her decision in the licence application. Indeed, it would be entirely unreal. That is so not least when, as I have observed, the Appellant’s grounds of appeal are likely, in practice and to at least some degree, to have been framed as a response to the Minister’s decision, to which grounds one would expect the Minister to respond in his/her submissions or observations.

92. None of the foregoing is to dilute ALAB’s obligation to decide the appeal de novo – an obligation I am happy to re-emphasise. However it does make clear that in doing so, ALAB is not obliged to ignore what has gone before – including the terms of the Minister’s decision.

93. I also accept ALAB’s submission that a distinction is to be drawn between the quasi-judicial duty to make a “judgement on the issues” and the administrative and drafting task of reducing those judgements to writing. While care must be taken to ensure no slippage as to that distinction, it is inevitable in licensing of many types that similar issues repeatedly arise and result in similarly-worded terms and conditions in licences. Indeed, as long as the need to respond to the particular circumstances of each application is kept firmly in mind and as long as the preparedness to draft bespoke conditions is maintained, there is much to be said for the use of standard terms and conditions, the meanings of which come to be understood generally by repeated use, experience and, perhaps, authoritative judicial interpretation. Decision-makers will also be conscious that in departing from well-worn wording they may be considered to have likewise departed in some way from a well-understood meaning of that well-worn wording and thereby to have provided a hook on which to hang a legal challenge. Appreciation of the predictable content of such conditions may aid all protagonists in participating in the licensing process. In short, departures from form should not be made for the sake of form by way of demonstration of de novo decision-making. It seems to me entirely unsurprising that a licence granted by ALAB on appeal and by way of de novo decision may be in very similar terms to the draft licence prepared by the Minister.

94. And, human nature being what it is, (even administrators and quasi-judicial decisionmakers are human apparently) at least considerable use of standard terms and conditions is not merely unobjectionable and arguably desirable – it is probably inevitable. Nor should ALAB be forced to reword the terms and conditions of a Minister’s licence, with which it happens to agree as a matter of its judgment de novo, merely to be seen to have made its judgment de novo. That would prioritise form over substance.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

95. Returning to Clarke J’s two critical characteristics of a de novo appeal, and remembering that they are general propositions which must yield to the particular statutory scheme at issue, the first, that the

decision taken at first instance is “wholly irrelevant”, clearly is inapplicable to an appeal to ALAB under the 1997 Act. That is so given the explicit obligation on the Minister to forward to ALAB not merely a copy of his/her decision but all reports to the Minister which informed it. The requirement of grounds of appeal and the Minister’s entitlement to make submissions to ALAB also seem to me to disapply this characteristic – at very least when taken with the obligation on the Minister to forward a copy of his/her decision and those reports to ALAB.

96. Clarke J’s second critical characteristic of a de novo appeal is that the appeal body must reach its own conclusions on the evidence and materials properly before to it. That requirement is invariably applicable in a de novo appeal. The indicia and evidence of the satisfaction of that criterion will vary with the type of decision under appeal, the statutory scheme and the type of output resultant on the decision under appeal. Here that output is a highly technical and detailed licence to be issued in a field in which standardisation of the terms of such licences is to be expected, even while ensuring their applicability to the particular circumstances of the case.

97. In my view, and as to satisfaction of that second criterion, it is far more important in this case to consider the substance of the decision-making process in which ALAB engaged as opposed to a textual comparison of the Minister’s draft licence and the licence issued by ALAB. Nor is a formalistic analysis of ALAB’s determination the important issue here.

98. ALAB’s determination clearly and repeatedly harks back to the Minister’s decision. Given the statutory scheme as described above,295 that is of itself unobjectionable.

99. In Part 4 of its determination, as to EIA, ALAB refers to “the EIA” in terms clearly referable to the Minister’s EIA and to which the Applicants pointed as intimating a lack of a de novo decision – indeed a failure to itself do an EIA. It would have been preferable, to avoid confusion, to refer to it explicitly as the Minister’s EIA. However it is also clear, reading Part 4 as a whole, that while its conclusion could have more clearly stated that it had itself done an EIA, ALAB in fact did so. ALAB’s recorded determination that MOWI’s 2011 EIS was inadequate, its requirement of an sEIS and its inviting submissions and observations thereon, clearly support the view that ALAB did its own EIA. §§4.4 and 4.5 record what are clearly ALAB’s own conclusions on the issues of sea lice, risk to the freshwater pearl mussel and wild birds. At §6.8.4 ALAB returns to the issue of EIA, stating that “appeals expressed dissatisfaction with the adequacy of the 2011 EIS and the EIA. Part 4 of this Determination outlines the steps taken by the Board to ensure a complete Environmental Impact Assessment process was followed”. ALAB follows with a repetition of its own

295 s.43(2) 1997 Act.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“satisfaction” in terms set out in §4.6 and undeniably reflecting Articles 2 and 3 of the EIA Directive. While

the words “the EIA” must refer to the Minister’s EIA, it is perfectly clear that ALAB is stating that it did its

own EIA. Similar content is found at §7.4.1 of the determination.296

100. In Parts 2, 4 and 5 (as to AA) of its determination, ALAB cites its regard to, and reliance on, numerous and significant reports and other information post-dating the Minister’s decision: notably and inter alia, the Technical Advisor’s Interim Report of December 2016 (Saunders), the Otter Screening Report of November 2017, the Seal Screening Report of February 2018, the Bird Impact Assessment Report of February 2018 (Gittings), the sEIS dated April 2018, the AA Screening Report as to Birds of April 2019 (Crowe), the MI Screening Reports of June 2018 and September 2020, the NIS297 dated July 2020, the MERC AA report of September 2020 and supplemental Briefing note of May 2021, the Freshwater Pearl Mussel Report of May 2021 (O’Toole), the Kelp Harvesting, In-Combination Effects, Report of May 2021 (O’Toole), the Technical Advisor’s Final Report of December 2020 (Saunders), in addition to numerous listed contributions by parties and prescribed bodies and s.46 and s.47 responses. ALAB’s determination also refers to the fact that an oral hearing had been held and records that ALAB had regard to the Chairman’s report on that oral hearing and to the listed written submissions to that oral hearing.

101. §5.6.5 of ALAB’s determination explicitly records that ALAB did its own AA and required an NIS of MOWI for that purpose. MOWI had not submitted an NIS to the Minister. The determination also cites ALAB’s own AA Conclusion Statement of May 2021.

102. Not only did ALAB, in making its decision, explicitly invoke its jurisdiction under s.40(1)(b) of the 1997 Act to determine the Appeal as if the licence application had been made to ALAB “in the first instance”,

at §6.8.8 of its determination it explicitly noted that it was determining “the application for the licence as if

the application had been made to the Board in the first instance.”

103. It is true that ALAB’s determination at §6.1.3 refers to the Board’s imposing “an additional condition”. I accept that the sense is that the condition is “additional” to those which the Minister had imposed. ALAB’s determination at §6.5.2 reads that it “determines that Schedule 4 of the licence be amended by removing the specified details concerning production and substituting the following:” I accept that the

words “amended”, “removing” and “substituting” refer to the Minister’s decision – as are similar words in

§6.5.3. Those words would have been better not used as creating ambiguity as to whether ALAB had made its decision de novo. I respectfully recommend to ALAB that it might consider how better to ensure the

296 “The 2011 EIS submitted by the Applicant was considered inadequate in parts by the Board. See Part 4 of this Determination for details of this and of

subsequent reports requested, along with consideration by the Board before their acceptance of the Environmental Impact Assessment process as

complete and adequate and that the proposed aquaculture activity will not have significant effects on the environment, by virtue of, inter alia, its nature,

size or location.”

297 Natura Impact Statement within the meaning of the Habitats Directive.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

avoidance of any impression of a decision taken other than de novo – if only to avoid challenges, even if unsuccessful, on this account.

104. But, for reasons set out above, having regard to

• the statutory structure, including provision to ALAB of the Minister’s decision

• the reports to the Minister and

• the requirement of grounds of appeal and the entitlement of the Minister to make submissions in the Appeal and having regard more generally to

• the detailed content of the determination,

• the obvious centrality to that determination of many documents post-dating the Minister’s

decision and

• ALAB’s explicit invocation in the Impugned Decision of its obligation to decide the Appeal de novo, I do not consider that these counter-suggestive references, taken by themselves or with any other information before me, undermine the proposition that ALAB determined the appeal de novo.

105. I therefore reject Core Grounds IFI CG1, SWI CG3 (as to this issue) and PS CG1.

BIAS, FAIR PROCEDURES, FURTHER INFORMATION PROCESSES DELAY & FACILITATION OF MOWI – INTRODUCTORY NOTE

106. In terms of strict legal analysis, allegations against ALAB as they relate to respectively, alleged bias, alleged unfair procedures, alleged misuse of the statutory further information processes of the 1997 Act and alleged breach of ALAB’s duty of expedition imposed by the 1997 Act, under the general umbrella of an allegation of excessive facilitation of MOWI in remediating its licence application with a view to granting it, are distinct and separate grounds upon which judicial review is sought and by reference to which the claims for certiorari of the Aquaculture Licence are to be considered.

107. However, on the facts of the present case these issues are in reality entangled as the essential allegation is that:

• ALAB prejudged the licence application by determining at an early stage to grant it despite significant defects in the application.

• Motivated by that prejudgment, ALAB operated its statutory procedures unfairly – in essence it “bent over backwards”, as it were, to repair or allow the repair of the application so that the aquaculture licence could be granted.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• This unfair process took the form of repeated notices to MOWI and to others, pursuant to s.46 and s.47 of the 1997 Act and repeated extensions of time under s.56 of the 1997 Act to allow the processes initiated by such notices to take their course to MOWI’s advantage.

• ALAB retained its own experts to “plug the gaps” in MOWI’s application.

• As a matter distinct from the issues of motivation and unfairness, it is alleged that ALAB breached its statutory duties in that: o The “interests of justice” did not require the s.46 notices. o The information sought in the s.46 notices was not “necessary for the purpose of enabling it to determine an appeal” – that being the criterion as opposed to “necessary for the purpose of enabling it to grant the licence”.

108. These issues are further entangled in that the applicants for judicial review have characterised events in the processing of the aquaculture licence application, in particular the issuing of the s.46, s.47 and s.56 notices as bias (even though those events are internal to the process) or as evidence internal to the process rendering apparent bias external to the process. And, in the alternative, those events internal to the process are characterised as denoting unfair procedures.

109. As it seems to me that, at least broadly, the sequence posited by the allegations against ALAB proceed from, first, the issue of bias, I should consider that issue first. However, as the issues are entangled I will inevitably, if incidentally, in considering the bias issues, touch on issues of fairness of procedures and breach of statutory duty.

BIAS – OBJECTIVE & STRUCTURAL

110. All the applicants allege objective bias. They vary in their particulars but their pleas have much in common.

Bias – Introduction & the Law of Objective Bias

111. Quasi-judicial bodies such as ALAB are bound by important duties of impartiality as between, and independence of, all concerned – Cairde Chill An Disirt.298 It is of the essence of the rule against bias is that

decision-makers must keep an “open mind” as to their decisions to the point in time of decision – e.g.

298 Cairde Chill An Disirt v An Bord Pleanála, Shannon Explosives et al [2009] IEHC 76 ([2009] 2 I.L.R.M. 89).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Orange #2299, Nurendale,300 Murphy v DPP301 and Ulster Bank v McDonagh.302 As O’Donnell CJ recently observed in Kelly,303 the rule against bias is one of fair procedures, “derived from the fundamental rule of nemo iudex in causa sua”. It seems fair to suggest that bias also offends the other fundamental rule of fair procedures – audi alteram partem – in the sense that a biased decision-maker with a closed mind is unlikely to give a fair hearing to the party against whom (s)he is biased – or there may arise a reasonable apprehension to that effect.

112. In general terms and as a starting point, it is vital that ALAB must be as open to refusing a licence as to granting one.

113. While it must have regard to government policy, ALAB is not an executive instrument of that policy. It is a very different thing – an independent arbiter as between the relevant protagonists. Though the interests counterpoised before ALAB are somewhat, but not entirely, different to those at issue in Reid304 and those typically before An Bord Pleanála, the following observation of McKechnie J in Reid illustrates and applies to the position of ALAB:

“... An Bord Pleanála … is, and importantly is seen as, objectively impartial in its decision making function, positioned as it is between the body's interest in ultimately advancing the public good and the landowner's vested interest in preserving his constitutionally protected rights. Such intervention of an independent third party inspires overall confidence in the process and, unless made practically impossible or exceedingly difficult by compelling countervailing circumstances, should be provided for.”

114. Also applicable to ALAB, as to the requirements of impartiality, independence, keeping an open mind and, generally, avoidance of actual bias or the appearance of bias, is the following observation of O’Donnell in Balz305 – though he was speaking of the necessity to consider submissions and explain their nonacceptance:

“This is fundamental not just to the law, but also to the trust which members of the public are required to have in decision making institutions if the individuals concerned, and the public more generally, are to be expected to accept decisions with which, in some cases, they may profoundly disagree, and with whose consequences they may have to live.”

299 Orange Ltd v Director of Telecoms (No. 2) [2000] 4 IR 159 Citing Locabail (U.K.) Ltd. v Bayfield Properties Ltd. [2000] 2 W.L.R. 870; [2000] 1 All E.R. 65.

300 Nurendale Ltd t/a Panda Waste Services v Dublin City Council [2013] 3 IR 417, §158 et seq.

301 Murphy v Director of Public Prosecutions [2021] IESC 75, [2021] 2 I.L.R.M. 377.

302 Ulster Bank Ireland DAC v McDonagh [2022] IECA 180.

303 Kelly v The Minister for Agriculture, Fisheries and Food, et al [2021] 2 IR 624.

304 Reid v Industrial Development Agency [2015] IESC 82, [2015] 4 I.R. 494. As to the exercise of compulsory purchase powers by the IDA.

305 Balz v An Bord Pleanála [2019] IESC 90 (Supreme Court, O'Donnell J, 12 December 2019).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

115. So, the nature of the decisions made by ALAB on aquaculture licence appeals is such that the law of bias applies thereto.

116. The allegations against ALAB are of objective (also called apparent) – not actual – bias. Barniville J in CHASE306 cited, as the overriding principle of the law of objective bias, the principle stated by Lord Hewart C.J. in Sussex Justices:307

“… it is …. of fundamental importance that justice should not only be done, but should manifestly and undoubtedly be seen to be done.”

In an observation as to judicial proceedings, but just as applicable to quasi-judicial bodies such as ALAB and perhaps to salmon farming in particular, Denham J in Wellwoman308 said:

“In cases such as this where many reasonable people in our community hold strong opinions, it is of particular importance that neither party should have any reasonable reason to apprehend bias ….”

117. The test of objective bias is, unsurprisingly, “strictly objective”309 and is set out in Orange #2.310 A

“… decision will be set aside on the ground of objective bias where there is a reasonable apprehension or suspicion that the decision maker might have been biased, i.e. where it is found that, although there was no actual bias, there is an appearance of bias."

As McKechnie J says in Nurendale:311

“It need not be proved that there is actual prejudgment or bias; a reasonable suspicion or real likelihood will suffice312 … the decision will be stood down because of the perception that, in the absence of the factor constituting bias, the decision may have been different.”

118. All the cases agree that the test remains strictly objective. As was said in O'Callaghan313

306 Cork Harbour Alliance For A Safe Environment v An Bord Pleanála [2021] IEHC 203 – Adding “Almost all of the leading Irish cases cite this dictum”.

307 Rex v Sussex Justices, Ex Parte McCarthy [1924] 1 KB 256.

Dublin Wellwoman Centre Ltd v Ireland [1995] 1 ILRM 408.

309 O'Callaghan v Mahon [2007] IESC 17, [2008] 2 IR 514, Fennelly J.

310 Orange Ltd v Director of Telecoms (No. 2) [2000] 4 IR 159, at p. 186 – confirmed by Murray J. in Spin Communications Ltd v IRTC [2001] 4 IR 411, at p.

431 and by McKechnie J in Nurendale Ltd t/a Panda Waste Services v Dublin City Council [2013] 3 IR 417.

311 Nurendale Ltd t/a Panda Waste Services v Dublin City Council [2013] 3 IR 417.

312 Citing E.H. Cochrane v Ministry of Transport [1987] 1 N.Z.L.R. 146, at p. 149; Committee for Justice and Liberty v National Energy Board [1978] 1 S.C.R.

369, at p. 394 (dissent of Granpre J.)).

313 O'Callaghan v Mahon [2007] IESC 17, [2008] 2 IR 514.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“The appearance of what is being done is critical. … the test refers to a reasonable apprehension by a reasonable person, who has knowledge of all the facts, who sees what is being done. It is this reasonable person's objective view which is the test. … It is not the apprehension of a party.”

In similar vein, it was said in Goode Concrete314 that “As it is an objective test, it does not invoke the apprehension of a judge, or any party: it invokes the reasonable apprehension of a reasonable person, who is in possession of all the relevant facts.”

119. Barniville J in CHASE added, citing McGuinness J in Bula #6315 as to objective bias based on prior association but, I think, in terms applicable to objective bias generally, that “the apprehension of bias must be both reasonable and realistic” – though the phrase serves to emphasise a single criterion rather than add

a second. The reasonable person is not “oversensitive”, “will pay attention to the wood, not the trees” (i.e.

look at the “big picture”) and, as to the facts, knows “both those which tended in favour and against the possible apprehension of a risk of bias.” 316

120. Irvine J in Penfield Enterprises317 stressed that it was “very clear from the leading decisions… that each case must turn upon its own particular facts and circumstances”. Allen J in Kemper observes that “The authorities on bias are, … necessarily highly fact specific.” Barniville J said likewise in CHASE.318

121. Irvine J in Penfield Enterprises also considered that the party alleging objective bias bears the onus of proving it. That is undoubtedly so, including in judicial review.319 But in the specific context of judicial review, it does not relieve the respondent decision-maker of its duty of candour to put on the table, face up, all cards relevant to the issues in the case – Student Transport Scheme.320

122. McKechnie J, citing the dissenting judgment of Hardiman J in O’Callaghan,321 considered that the views of the parties are “pragmatically” admissible evidence of objective bias.322 But Fennelly J in

Goode Concrete v CRH plc [2015] IESC 70 [2015] 3 IR 493.

315 Citing McGuinness J in Bula Ltd v Tara Mines Ltd (No. 6) [2000] 4 IR 412 – a case Barniville J describes as “The leading authority in this jurisdiction now

on objective bias …” At issue in Bula was prior association between the decision-maker, or a member of the decision-making body, and one of the parties

to the dispute or issue.

316 Fennelly J in Kenny v Trinity College Dublin [2008] 2 IR 40. Also, O'Callaghan v Mahon [2007] IESC 17, [2008] 2 IR 514 – the reasonable fair-minded

objective person is someone is not “unduly sensitive” and Davidson v Scottish Ministers [2004] UKHL 34. All cases cited by Barniville J in

Cork Harbour Alliance For A Safe Environment v An Bord Pleanála [2021] IEHC 203 §116.

317 Commissioner of An Garda Síochána v Penfield Enterprises Ltd [2016] IECA 141 cited by Barniville J in Cork Harbour Alliance For A Safe Environment v

An Bord Pleanála [2021] IEHC 203 §126.

318 Cork Harbour Alliance For A Safe Environment v An Bord Pleanála [2021] IEHC 203, §95.

319 E.g. Royal Brompton & Harefield NHS Foundation Trust v JCPCT [2012] EWCA Civ 472, §126.

320 Student Transport Scheme Ltd v Minister for Education [2021] IESC 35.

321 O'Callaghan v Mahon [2007] IESC 17, [2008] 2 IR 514.

322 McKenzie J in Nurendale Ltd t/a Panda Waste Services v Dublin City Council [2013] 3 IR 417 §165.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

O’Callaghan, Allen J in Kemper323 and Barniville J in CHASE324 seem to disagree – as I respectfully tend to. Not

least, Hardiman J’s view seems to imply inquiry whether the party whose views are to be admitted in evidence is a reasonable person. And no matter how reasonable a party may be in a general sense, it is difficult to see him/her as objective in a context in which he/she is an interested – often highly motivated and partisan – protagonist. That is why, for example, the opinion of a party on matters calling for expertise is inadmissible in evidence even if he/she is otherwise an expert.

123. The Sweetman Applicants cite Chronopost SA325 – a “state aid” case – for its statement of the duty of a tribunal to be objectively impartial – to offer guarantees sufficient to exclude any legitimate doubt that none of its members show bias or personal prejudice. They do so in the specific context of EIA Directive requirements of objectivity and as to conflicts of interest. To this one may add reference to the right to good administration – including impartial decision-making – found in Art 41 CFREU. However I do not see that in this respect EU Law differs from or adds to Irish law as to objective bias, nor did the Sweetman Applicants suggest how it might do so.

124. The recent judgment of the Supreme Court in Kelly326 reviewed the cases on bias.327 Though grounded in the unusual circumstances of that case and though all bias cases turn on their particular facts, the judgments in that and the other cases cited here seem to me authority for the following inexhaustive list of principles:

i. The purpose of the law of bias is the preservation of confidence in the objectivity, integrity, independence and impartiality of public administration. Bias is about trust in decision-making public institutions. That is the prism through which the question of objective bias is viewed.

ii. Judicial review of the validity of decisions “must be approached with a realistic understanding of the imperfections of any process, but if an outcome is invalid, the court must say so.”328 iii. Objective bias329 consists in the appearance as opposed to the actuality of bias. Actual bias distorts objective, independent and impartial decision-making. Objective bias consists in the reasonable suspicion of the presence of such distortion.

iv. Unsurprisingly, the presence of objective bias is objectively discerned: it is not the perception of the parties that matters. The better view seems to be that their evidence of that perception is inadmissible as evidence of objective bias.

323 Joyce Kemper v An Bord Pleanála [2020] IEHC 601 (High Court (Judicial Review), Allen J, 24 November 2020).

324 Cork Harbour Alliance For A Safe Environment v An Bord Pleanála [2021] IEHC 203 §116.

325 Joined Cases C-341/06 P and C-342/06 P Chronopost SA and La Poste v Union française de l’express (UFEX) and Others [2008] All ER (D) 41 (Jul).

326 Kelly v The Minister for Agriculture, Fisheries and Food, et al [2021] 2 IR 624.

327 I have also considered Usk and District Residents Assoc. Ltd. v An Bord Pleanála [2010] 4 I.R. 113, which was cited to me.

328 O’Donnell CJ §3.

329 Also known as perceived bias and apparent bias.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

v. Broadly, the test of objective bias is whether a hypothetical reasonable observer/bystander, aware of all relevant facts, would have a reasonable apprehension of a risk that the decision-maker will not hear and decide the matter fairly, independently and impartially – keeping an open mind as to the outcome until the proper point of its decision. It is not necessary to show a real likelihood of bias – a real possibility, a reasonable suspicion, suffices.330 But it must be reasonable and realistic rather than fanciful or vague.

vi. The hypothetical reasonable observer is an independent person, with knowledge of the relevant facts. That person is fair, but robust and not over-sensitive, over-scrupulous or precious. He or she is neither unduly cynical nor prone to conspiracy theories, is neither naïve and complacent nor unduly suspicious and is aware that if, by voicing mere suspicions, disappointed parties could establish objective bias, the result would be a near-impossible imposition of a very heavy price in decisions set aside and outcomes delayed. In short, (s)he comes well-equipped with a broad sense of justice and common sense.

vii. O’Donnell J in Kelly made the perceptive and important observation that the test of objective bias is easier stated than applied as, almost by definition, the views of reasonable people can differ on a given set of facts and as to, inter alia, political decision-making and the workings of public administration in a small country such as Ireland with a large public administration. At some risk of over-elaboration, to this observation I would respectfully add the following further observations:

▪ That reasonable people can properly – indeed will often – differ in their views of the significance of and inferences to be drawn from a given set of facts is a commonplace of the law. Most notably, it is the bedrock of the law of irrationality of public law decisions – to the extent that any reasonable basis for a decision will suffice (Weston331). If any reasonable basis is shown, the

decision will be deemed rational even if the Court is satisfied that the case against the decision was much stronger than the case for it (O’Keeffe332) and would itself have made a different (and, inferentially, also reasonable) decision. Indeed, a decision may be deemed rational if there were any materials before the decision-maker capable of supporting it (O’Keeffe). Accordingly, it has been suggested that the threshold to show irrationality “is extremely high and is almost never met in practice” (St. Audoen’s333) and requires “something overwhelming”.334 In other words, there may be multiple reasonable views of the significance of a given set of facts.

▪ This implies that, from a given set of facts, some may reasonably infer bias where others would not.

330 The phrases “reasonable apprehension” and “reasonable suspicion” are interchangeable.

331 Weston v An Bord Pleanála [2010] IEHC 255.

332 O'Keeffe v An Bord Pleanála [1993] 1 IR 39.

333 The Board of Management of St. Audoen’s National School v An Bord Pleanála [2021] IEHC 453.

334 Finlay CJ in O'Keeffe v An Bord Pleanála [1993] 1 I.R. 39 citing Henchy J in The State (Keegan) v Stardust Compensation Tribunal [1986] I.R. 642 in turn

citing Lord Greene M.R. in Associated Provincial Picture Houses Limited v Wednesbury Corporation [1948] 1 K.B. 223. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

▪ The cases also establish that judicial decisions on bias are highly fact-sensitive/fact-specific and require a “particularly careful exercise of the faculty of judgment”335.

▪ This is why the exercise of the faculty of judgment must be informed by a “realistic understanding of the imperfections of any process.”

viii. The mere fact of a decision within the decision-making process – even a perverse decision336 – contrary to the interest of the party alleging bias or of legal or other errors, does not show bias.

ix. To establish objective bias it is necessary to show the existence of something external to and predating the impugned decision.337 There are nuances to the issue of externality to which I will turn presently, but it is clear that bias external to and predating the impugned decision may first become apparent within the process.

x. Bias – actual or objective – of one member of a multi-member decision-maker will invalidate its impugned decision.338

Bias – External to the Impugned Process

125. Generally, bias must be demonstrated by factors outside of the impugned decision-making process. Reliance, in impugning a decision, on factors internal to the process falls to be considered on principles of fairness and constitutional justice: presumably that is because unless the factors internal to the process demonstrate unfairness or breach of the principles of constitutional justice they could not demonstrate bias and if they do demonstrate unfairness or breach of the principles of constitutional justice, the decision will be struck down on that account and recourse to the concept of bias is unnecessary.

126. McKechnie J reviewed the caselaw in this regard in Nurendale.339 He observed that “vitiating factors, which arise for the first time during the course of a hearing, are not without a remedy: they are dealt with by fair procedures and natural justice, but not by means of bias.” He noted that, in Orange #2, Murphy J required demonstration “that the adjudicator is affected by some factor external to the subject matter of his

decision” and Barron J stated that bias will always predate the actual or contemplated decision and that:

Kelly v The Minister for Agriculture, Fisheries and Food, et al [2021] 2 IR 624. Dunne J citing Bula Ltd v Tara Mines Ltd (No 6) [2000] 4 IR 412, at §441;

see also O'Neill v Beaumont Hospital Board [1990] ILRM 419, Dublin Wellwoman Centre Ltd v Ireland [1995] 1 ILRM 408 and O'Callaghan v Mahon [2007]

IESC 17, [2008] 2 IR 514).

336 Kelly v The Minister for Agriculture, Fisheries and Food, et al [2021] 2 IR 624. Dunne J §§109, 135, 142, 143 (§143 adopting the views of Costello J cited

in §§109, 135, 142).

337 Orange Ltd v Director of Telecoms (No. 2) [2000] 4 IR 159.

338 Reid v Industrial Development Agency [2015] IESC 82, [2015] 4 I.R. 494.

339 Nurendale Ltd t/a Panda Waste Services v Dublin City Council [2013] 3 IR 417, §158 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“Bias does not come into existence in the course of a hearing. It may become apparent in the course of a hearing and in that way alert a party to the possibility of bias and so enable such party to establish facts which show that the attitude adopted by the decision maker in the course of the hearing was one which might have been expected having regard to those facts.”

In other words, if during the hearing a party comes to suspect bias, it may be prompted to investigate the possibility of bias but one can still only show bias by relying on facts external to the process.

127. However while the externality requirement is general, it seems that exceptionally it may be possible to establish objective bias based on statements made during a hearing – Erdogan.340 Also, bias by way of prejudgment seems to be an exception to the externality requirement – see below.

128. In Kelly,341 O’Donnell CJ cautioned, obiter, against over-reading Orange #2 as to the requirement of externality.342 He considered that bias must be external to the correctness of the impugned decision but that this reasoning does not require that bias be external to the process leading to the decision. However, Dunne J, giving the majority decision343 in a passage appreciably relevant in the present case, said that:

“…. in considering the question of actual bias, decisions made or taken in the course of an investigation or proceedings cannot, of themselves, give rise to an inference of bias. …. the conduct of the decisionmaker during the process cannot establish bias as bias must arise from a factor outside the process impugned.” 344

And as I have said, Dunne J, as to actual bias, also rejected the view that “evidence of bias could be found in any alleged perversity of the decision-maker's decision.” 345 I cannot see that such a view as to actual bias could in logic be disapplied to objective bias.

Bias – Prejudgment – Internal to the Impugned Process?

Prejudgment Generally

129. IFI submits that prejudgment is a particular form of bias as to which particular considerations arise. It alleges objective bias in the form of prejudgment in that the objective observer would infer that “the result

340 Erdogan v Workplace Relations Commission [2021] IEHC 348 – though no such bias was found in that case and so the remark is obiter.

341 Kelly v The Minister for Agriculture, Fisheries and Food, et al [2021] 2 IR 624.

342 §§12 & 13. This his view is obiter is apparent from the first sentence of §14.

343 O’Donnell CJ concurred and what is canvassed here seems to be an island of possible disagreement.

344 My emphases.

345 Kelly v The Minister for Agriculture, Fisheries and Food, et al [2021] 2 IR 624. Dunne J §§109, 135, 142, 143 (§143 adopting the views of Costello J cited

in §§109, 135, 142). SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

of the appeal was a foregone conclusion based on the repeated delays in the determination of the appeal and the fact that at every turn MOWI was afforded ample opportunity to address any issues that arose.”

130. It may assist to first sketch some of the elements of prejudgment:

i. Prejudgment may be either actual/subjective or objective. Only the latter is at issue in this case – as IFI were at pains to emphasise.

ii. Prejudgment consists in either an actual/subjective premature decision by the decision-maker as to the outcome of the judicial or quasi-judicial process in question or in circumstances of objective bias such that a reasonable person in the circumstances would have a reasonable apprehension that, by reason of premature decision, the applicants would not have a fair hearing from an impartial

decision-maker.346 Prejudgment bias has been described pithily as consisting in “a prematurely closed mind” – Stubbs.347

iii. Prematurity arises at a stage at which further evidence or argument is to be expected.348 iv. A preconceived idea, strong view or provisional view on a matter – even as to what action should be taken or as to what the result ought to be – is not prejudgment, provided the decisionmaker has not resolved beforehand to come to a particular conclusion and remains willing to hear the submissions of the parties.349 There is English authority350 that “predisposition is not predetermination” and distinguishing,

“…… between a legitimate predisposition towards a particular outcome .. and an illegitimate predetermination of the outcome (for example, because of a decision already reached or a determination to reach a particular decision). The former is consistent with a preparedness to consider and weigh relevant factors in reaching the final decision; the latter involves a mind that is closed to the consideration and weighing of relevant factors.”

One need not apply the entirety of those English decisions in Ireland to observe that the foregoing extracts seem consistent with the Irish cases. As I have said, McKechnie J in Nurendale351 observed that “ …an administrative body may have some outcome in mind. That does not amount to prejudgment ..”. It seems to me usefully described as a “provisional view” – the proviso being a mind

346 This formulation applies to prejudgment the test of objective bias set down in Bula Ltd. v Tara Mines Ltd. (No. 6) [2000] 4 I.R. 412.

347 R v Stubbs; R v Davis; R v Evans [2018] UKPC 30 [2019] 1 All ER 581.

348 Murphy v DPP [2021] 2 I.L.R.M. 377.

McGrath v The Trustees of Maynooth College [1979] I.L.R.M. 166, Supreme Court. O’Higgins CJ said: “A preconceived view as to what action should

be taken or as to what the result ought to be does not invalidate administrative action of the kind in question provided the body concerned is willing to

hear what may be said by the party charged and affords to him a full and real opportunity of making his defence. The essential requirement that the

decision be a bona fide and honest one is not destroyed merely because prior to its making the body in question or individual members thereof already

had strong views as to what should be done.” In Nurendale, McKechnie J cited this passage in short form apparently with approval.

350 R (Lewis) v Redcar and Cleveland Borough Council [2009] 1 WLR 83 citing Condron v National Assembly for Wales [2007] LGR 87 and R (Island Farm

Development Ltd) v Bridgend County Borough Council [2007] LGR 60.

351 Nurendale Ltd t/a Panda Waste Services v Dublin City Council [2013] 3 IR 417, §158 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

open to change. McKechnie J cited McGrath352 and O’Neill353 to the effect that while “a preconceived idea or strong view on a matter did not invalidate a decision made by an administrative body, the body must still have been willing to hear the submissions of the parties and should not have resolved to come to a particular conclusion beforehand. This does not impose a requirement that the decision maker should have had no prior knowledge of the matter pre-hearing.” What is required, rather, is an “open mind”. The same view was taken in Lohan354 in the particular context of written submissions made prior to an oral hearing:

“The purpose of providing written submissions in advance of a hearing is to allow the decision maker to consider the submissions in advance. It is always open to a decision maker to form a preliminary view of a matter which may or may not change in light of oral arguments. This is precisely what occurred in this case: the Tribunal formed a preliminary view, and it was not persuaded by the oral submissions to the contrary position. I see no want of fair procedures or impermissible prejudgment on the facts in this case. On the contrary, the evidence is that the draft judgment was amended to reflect the oral submissions. The fact that they did not result in a different outcome does not lead to the conclusion that there was impermissible prejudgment or a failure to consider the arguments advanced by the appellant in his oral submissions.”

v. As decision-making in matters such as this is often properly, desirably and indeed inevitably iterative, it is to be expected that, over time and the progress of the process and as facts, information, arguments, views, and advice accumulate and are considered, a substantive decision may come progressively into view. That one “arrives” at a decision implies a journey to it. Decisions do not erupt, unheralded complete and perfectly formed, at the very end of a deliberative process. It would be very bad for the quality of public administration if they did.

vi. What is required is an open, as opposed to a closed, mind. That is, a mind open, to the very end of the process, to making a decision other than in accordance with any preconceived idea or even an idea developed iteratively in the process.

vii. It must be said that not all the decisions are entirely one way. Allen J in Kemper cites Denham J in Bula,355 citing in turn the High Court of Australia in Vakauta,356 to the effect that a lay observer likely would not appreciate the ability of a trial judge to ensure that preconceived views do not cause the actual decision to be tainted by prejudgment or bias. However in both Kemper and Bula, the objective bias allegation failed. In Joyce Kemper, Allen J observes that “The authorities on bias are, … necessarily highly fact specific.”

352 McGrath v Trustees of Maynooth College [1979] ILRM 166.

353 O'Neill v Beaumont Hospital Board [1990] ILRM 419.

354 Lohan v Solicitors Disciplinary Tribunal [2023] IECA 18.

355 Bula Ltd. v Tara Mines Ltd. (No. 6) [2000] 4 I.R. 412.

356 Vakuta v Kelly (1989) 167 C.L.R. 568.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Prejudgment – Internal to the Impugned Process? – Pattern of Erroneous Decisions.

131. IFI submits that objective prejudgment differs from other forms of objective bias in that it may arise internal to the process. It says that, while in Nurendale McKechnie J said that “vitiating factors, which arise for the first time during the course of a hearing, are not without a remedy: they are dealt with by fair procedures and natural justice, but not by means of bias”, it is difficult to fit prejudgment internal to a process into the usual categories of fair procedures. It is not, IFI submits, obviously an issue of nemo iudex in causa sua or an issue of audi alteram partem. The gravamen of this argument is that excluding the possibility of prejudgment internal to the process thus leaves the victim without a remedy. I respectfully disagree with that logic. Audi alteram partem expresses the duty to provide a fair hearing to both sides. “Audi” means listen or hear. It is difficult to see that prejudgment and a closed mind at a point when further evidence or argument is to be expected can be reconciled with the continuing obligation to hear and listen to both sides. However, the issue is governed by authority.

132. O’Callaghan et al,357 alleged objective prejudgment bias by a tribunal inquiring into alleged corruption in the planning process. They argued that bias could be inferred from the manner in which the tribunal addressed disclosure to them of documents in the Tribunal’s possession, such as witness statements made by a witness hostile to them, which statements were allegedly inconsistent with the statement of that witness disclosed to them. They cited seven examples from which, they said, a reasonable person would conclude that the respondents had decided that that particular witness was a credible witness and Mr O’Callaghan was not. The applicants obtained an order directing the release of such documents in previous judicial review proceedings – thereby establishing that certain decisions to not disclose such documents were incorrect in law.358

133. The Supreme Court held that bias could not be inferred from legal or other errors in the decisionmaking process or from a pattern of erroneous such decisions. It was necessary to show the existence of something external to the process. The question was whether there was a basis, external to the tribunal’s decisions, for a reasonable apprehension that the Tribunal had prejudged the credibility of the relevant witnesses.

134. Fennelly J reviewed the cases to a conclusion, expressly inclusive of prejudgment bias, that “objective bias cannot be inferred from the fact that a court or tribunal has made a number of erroneous decisions. Bias, if it exists, must flow from some element external to the decision-making process.”359 In the passages following360 he states that “the expression of strong though necessarily provisional views on the

357 O'Callaghan v Mahon [2007] IESC 17, [2008] 2 IR 514.

358 Fennelly J §87.

359 §70.

360 §§75 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

subject matter of the litigation, would not normally be considered to justify a finding of bias”361 but “a judge

may so behave that he steps outside his judicial role. If he does, it will be obvious. In my view, that is what is required, something quite outside the bounds of proper judicial behaviour to establish objective bias, based on judicial statements.” In other words, egregious judicial interventions during a trial could be evidence of prejudgment bias.

135. Fennelly J thereafter, in summarising the applicable principles, confirmed his view that “objective bias may not be inferred from legal or other errors made within the decision making process; it is necessary to show the existence of something external to that process.”362 Fennelly J’s next principle was that “objective bias may be established by showing that the decision maker has made statements which, if applied to the case at issue, would effectively decide it or which show prejudice, hostility or dislike towards one party or his witnesses”. It is unclear whether here Fennelly J was referring to statements external to the process or was allowing that statements internal to the process could establish prejudgment bias. Given his allowance that in egregious cases judicial interventions during a trial could be evidence of prejudgment bias I incline to think that he did so allow, but only in egregious cases.

136. He did not allow inference of bias merely on the basis of a series or pattern of erroneous decisions. Fennelly J said of the applicants that,

“…. they ask the court to infer from a detailed examination of the decisions and related behaviour of the tribunal that it has predetermined or prejudged the issue of credibility. If that exercise is different from the one condemned by this court in Orange363 …… I have to confess that the distinction is too fine from me. The applicants not only ask the court to look at the pattern of behaviour of the tribunal, but they have engaged, at the hearing in this court, in the most minute examination of individual pieces of evidence.”364

Denham J likewise held that a series of erroneous decisions internal to the process, in not disclosing the documents, is not a basis on which to find bias.365

137. AP v Judge McDonagh366 and Murphy v DPP367 are cited to me. O’Callaghan was not cited in McDonagh in which the “real issue of dispute” was as to externality of the matter allegedly showing bias. It appears to me to have been a case falling into Fennelly J’s category of egregious judicial comments at trial rather than the category of a series or pattern of erroneous decisions internal to the process. In Murphy, an

361 Citing the High Court of Australia: Vakauta v Kelly (1989) 167 C.L.R. 568.

362 §80.

363 Orange Ltd. v Director of Telecoms (No. 2) [2000] 4 I.R. 159.

364 §84.

365 p550 et seq.

366 P v Judge McDonagh [2009] IEHC 316.

367 Murphy v Director of Public Prosecutions, [2021] 2 I.L.R.M. 377 Supreme Court.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

accused person objected to his retrial before the judge who had conducted the trial in which the jury disagreed. O’Malley J observed that, in McDonagh, Clarke J had pointed to “a form of pre-judgement that arises, not from any pre-existing factor, but in circumstances where the decision maker indicates that he or she has reached a conclusion on a question in controversy at a time prior to it being proper to reach such a

conclusion. That will be the case where the decision maker uses language that might cause the reasonable observer to believe that he or she has excluded any realistic possibility of coming to any other conclusion, at a stage when further evidence or argument is to be expected. The observer will consider that a premature rush to judgement has occurred, although there may be no particular animus towards the party who stands to lose.” Clarke J had considered that “Most cases of bias involve an allegation that by reason of some factor external to the adjudicative process”. He appears to have included some types of pre-judgment in this view

but considered that there is “another form of prejudgment” in “a somewhat different category” – “where the

adjudicator indicates that the adjudicator has reached a conclusion on a question in controversy between the parties, at a time prior to it being proper for such adjudicator to reach such a decision”. Clarke J considered that the requirement of Barron J. in Orange that bias be external to the process could not,

“… be taken to prevent a finding of the form of pre-judgment (or premature judgment) to which I have referred (that is to say a pre-judgment stemming from an appearance being given of the adjudicator having made a decision at a time when further evidence or argument on the issue concerned remained to be presented), simply because such contention arises out of what happens at, rather than prior to, the hearing concerned.

The form of pre-judgment which I have sought to analyse can only happen at a hearing and arises from the adjudicator creating, in the minds of reasonable and informed people, an impression that a premature rush to judgment has occurred. Such a rush to judgment does not necessarily depend on the adjudicator having a particular animus towards one of the parties concerned, but rather stems from the fact that a reasonable apprehension has been created that the adjudicator has not considered all appropriate matters before reaching what appears to be a final view on the issue.

I am satisfied, therefore, that where the form of pre-judgment which is sought to be relied on in judicial review proceedings is of that type, it is not necessary to establish that it predated the adjudicative process.”

138. Shatter368 is an example of the application of McDonagh as to prejudgment internal to a process. It is an example also of the relevance of acquiescence. Meenan J held that an e-mail by the Data Protection Commissioner to Deputy Shatter, advising him of an intended investigation, displayed prejudgment bias in the sentence: “The Commissioner is satisfied the personal data of Deputy Wallace was processed by you in the incident complained of.” But Meenan J also held369 that Deputy Shatter’s failure to take steps to have the Commissioner recuse himself constituted acquiescence preventing his reliance on that prejudgment bias.

368 Shatter v Data Protection Commissioner [2017] IEHC 670 (High Court, Meenan J, 9 November 2017).

369 Citing Corrigan v Irish Land Commission [1977] I.R. 317.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

139. On the facts before Clarke J in McDonagh, prejudgment was discernible from explicit remarks by the judge. Assuming, just for the sake of argument, that prejudgment may be discernible by inference from the manner of conduct of a decision-making process, something quite egregious and a very considerable weight of evidence in its favour would be required. That must be especially so, if possible at all,

• when an inherently inquisitorial tribunal is accused, as here and in effect, of being excessively inquisitive and engaging in too much scrutiny of the application before it.

• given recent emphasis on the duties of decision-makers in environmental matters of “detailed scrutiny”,

“active and critical interrogation” “scrupulous rigour” and a “high standard of investigation” see cases

such as Weston,370 Balz,371 Atlantic Diamond,372 ETI373 and Jennings.374 The lay and reasonable observer, conscious of such duties, would also appreciate that such decision-making is an iterative process and that, as it proceeds, an eventual decision may come progressively, emergingly and increasingly into the view of the decision-maker before the final decision is made.

140. Seeking to reconcile the foregoing with the majority decision of Dunne J in Kelly, that “decisions made or taken in the course of an investigation or proceedings cannot, of themselves, give rise to an inference of bias”, and the similar decision in O’Callaghan, it seems to me that Dunne J had in mind the types of decision which ordinarily arise during the decision-making process and prior to the ultimate decision. She did not have in mind Fennelly J’s egregious judicial comments at trial category – of the kind at issue in McDonagh.

141. So, in my view, the authorities confirm that mere demonstration of a pattern or sequence in which decisions which ordinarily arise during the decision-making process have gone against the person alleging bias will not demonstrate pre-judgment bias, even where those decisions have been shown erroneous in law (as in O’Callaghan).

Bias – Advisors To Decision-Makers

142. The objective bias of an advisor to a decision-maker may infect the decision-maker’s decision – see Primary Health Investment Properties,375 Royal Brompton & Harefield376 and Sussex Justices.377 The point

370 Weston Ltd. v An Bord Pleanála [2010] IEHC 255.

Balz v An Bord Pleanála [2019] IESC 90.

372 Atlantic Diamond Ltd v An Bord Pleanála [2021] IEHC 322.

373 Environmental Trust Ireland v An Bord Pleanála [2022] IEHC 540.

374 Jennings & O’Connor v An Bord Pleanála & Colbeam [2023] IEHC 14.

375 R (Primary Health Investment Properties) v Secretary of State for Health [2009] EWHC 519 (Admin) [2009] All ER (D) 292 (Mar) §108.

376 Royal Brompton & Harefield NHS Found'n Trust v JCPCT & Anr [2012] EWCA Civ 472.

377 R v Sussex Justices, ex parte McCarthy [1924] 1 KB 256, 22 LGR 46, 88 JP 3 which led to the famous remark of Lord Hewart CJ about justice being “seen

to be done”. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

could have been argued but seems to have been assumed in Orange #2. However, whether that has occurred in a particular case will require fact-specific analysis – both as to the bias of the advisor and whether it has infected the decision.

143. The advice given by an advisor to a decision-maker – whether the advisor is retained by the decisionmaker itself or by a protagonist – must be expert, independent and impartial. Such advisors cannot see themselves as merely instruments in their client’s achieving a supposedly desired outcome of a statutory process. While it may be a demanding requirement in a context which is commercial from the advisor’s point of view, nonetheless, such experts must, to coin a cliché, speak truth to power. However, it must also be remembered that impartiality does not inhibit an advisor retained by the decision-maker itself from making a recommendation as to what the decision-maker should decide. That, indeed, is generally a main purpose of his/her retainer. And, while the advisor must keep an open mind to the end of the process, just as must the decision-maker, in his/her contribution to the iterative process of decision-making the advisor is entitled to make a provisional recommendation as to circumstances in which, on further inquiry, an application might be granted or refused. Further, the process of feedback from the decision-maker to the advisor – for example as to issues which concern the decision-maker or which it would like to see investigated – is at least generally as proper as it is necessary to effective decision-making.

144. It is necessary that bodies such as ALAB will, in retaining such advisors, brief them as to their task.378 That requirement may be particularly important where the expert is based in another jurisdiction and may be unfamiliar with the statutory, policy, procedural and factual contexts in which their advice is to be given. Indeed, it is incumbent on all experts participating in regulatory processes, to properly understand the legal and regulatory regime, including relevant policy and guidance, in which they are to operate. This may impose particular, but necessary and important, burdens on advisors normally practicing in other jurisdictions.

Bias – Allegations

145. Essentially, objective bias is alleged here on the following bases:

i. SWI and the Sweetman Applicants assert “structural bias” or “institutional bias”379 and conflicts of interest within ALAB as ▪ the statutory system of s.23 of the 1997 Act of appointment of members to ALAB representative of certain interests is constitutionally infirm as facilitating ALAB majorities in favour of the grant, as opposed to the refusal, of aquaculture licenses.

378 See for example the template expert report sent by ALAB to Dr Saunders in April 2016 – Exhibit MCC2 T53, Affidavit of Mona-Claire Costelloe p24.

379 These terms mean the same thing and I will use them interchangeably.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

▪ ALAB members determine appeals from decisions of the Minister who appointed them and may reappoint them.

▪ s.44 of the 1997 Act creates structural bias in that the Minister’s participation in and making of submissions in appeals results in his maintaining an ongoing supervision of the body that he appointed as it determines appeals against his decisions.

▪ The Minister is objectively (and/or subjectively) biased, the Government having adopted a nonstatutory policy “Harnessing our Ocean Wealth” in favour of aquaculture.

▪ Accordingly, s.23 and 44 of the 1997 Act are repugnant to Articles 34.1 and 40.3.1 of the Constitution and are invalid pursuant to Article 15.4.2 thereof.

▪ ss.23, 28, 40 and 44 of the 1997 Act are repugnant to/incompatible with Articles 6380 and 9a381 of the EIA Directive 2014 and Article 47 CFREU,382 s.3 of the European Convention on Human Rights Act 2003 and Article 6 ECHR383 in that the Applicants were deprived of a fair hearing before an independent and impartial tribunal. A declaration of incompatibility under s.3 of the European Convention on Human Rights Act 2003 is sought accordingly.

▪ The Impugned Decision and the EIA and AA done were vitiated by conflict of interest and lack of objectivity to the prejudice of the Applicants.

ii. Structural bias apart, the allegation of objective bias is epitomised in the assertion that, that the result of the appeal was a foregone conclusion and, having already decided to grant the licence before considering all the evidence before it and ALAB “bent over backwards” to “repeatedly groom the application to eliminate every uncertainty”384 and so grant the licence in collecting the material to justify its decision by repeatedly reverting to MOWI and others for further information, not least, it is alleged, to plug gaps in MOWI’s licence application.

iii. Generally, the pattern of decisions by ALAB in the Appeal and its delay in deciding it, combine to show prejudgment and rigidity of mind in seeking to grant the licence and a resultant determination to afford, unfairly to the other appellants, repeated and excessive opportunities to MOWI to cure defects in its licence application.

iv. Dr Saunders and Prof McIntyre were compromised by circumstances amounting to situations of objective bias – no actual bias is alleged.

v. ALAB’s interactions with Dr Saunders consisted of “massaging” 385 his reports to express views desired by ALAB. The allegation was not put precisely that way but that is what it amounted to.

380 Public Participation.

381 Article 9a requires of decisionmakers both objective decision-making and avoidance of conflict of interest.

Charter of Fundamental Rights of the European Union; Art. 47 establishes the right to effective remedy by way of “a fair and public hearing within a

reasonable time by an independent and impartial tribunal previously established by law.” SWP pleads Case C-873/19 Deutsche Umwelthilfe, §79 to the

effect that Art. 47 “is sufficient in itself and does not need to be made more specific by provisions of EU or national law in order to confer on individuals

a right which they may rely on as such.”

383 European Convention on Human Rights, Article 6.1. In the determination of his civil rights and obligations … everyone is entitled to a fair and public

hearing within a reasonable time by an independent and impartial tribunal established by law.

384 SWI Written Submissions §7.24.

385 Nurendale Ltd t/a Panda Waste Services v Dublin City Council [2013] 3 IR 417.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

vi. MOWI reverted repeatedly to MOWI to revise its NIS – while failing to publish the versions sent back for revision.

vii. ALAB’s relationship with the MI and the Department are problematic both generally and in respects specific to the Appeal.

146. As has been seen, in alleging bias, the Applicants variously cited articles of the Constitution, the CFREU386 and the ECHR387 as to such as the defence and vindication of personal rights, access to justice, constitutional justice and fair procedures as informing the law of bias. In my view, these apparently complex pleas say essentially the same thing by reference to laws saying essentially the same thing. In a nutshell, the question is one of bias. As the domestic law of bias is well developed substantively in vindication of such guarantees, I have generally not found it necessary to view the issues of bias specifically through those lenses.

Bias – Some pleading issues

147. A general point as to pleading bias should be made. The Sweetman Applicants plead that there are “numerous sources of bias and conflicts of interest in this process including”:

• Issues disclosed in ALAB’s February 2017 submission to an independent aquaculture licensing review group established by the Minister stating concerns as to o potential conflicts of interest in its relationship with the MI (who advise the Minister on aquaculture licences) o its shared office accommodation with and administrative support from the Agriculture Appeals Office of the DAFM.

• Dr Saunders’ having become technical advisor to ALAB “directly from his previous role in RPS and the Marine Institute”.

• Structural bias in the make-up of ALAB and recusals from consideration of the Appeal

• A conflict of interest arising from Professor Owen McIntyre, the Oral Hearing chair, being accidently overpaid over €18,000 and its repayment being resolved by ALAB, Prof McIntyre and the Department working together.

148. ALAB made it clear388 that it objects to any reliance by the Sweetman Applicants on the words “numerous” and “including” to extend their pleas. In my view, ALAB is entitled to that that position.

386 Charter of Fundamental Rights of the EU.

387 European Convention of Human Rights.

388 E.g. Affidavit of Francis Dowling sworn 28 June 2022.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Pleadings in judicial review must be particular and precise and that even objective bias is an “ugly allegation”389 emphasises the point.

149. There is an unpleaded submission that objective bias was discernible in ALAB’s seeking a presentation from RPS with a view to retaining it as an advisor, when RPS had already been retained by MOWI. However, as it is easily disposed of I will address it rather than let the ugly allegation lie.

150. So too IFI’s unpleaded submission of objective bias in that,

• having had advice from Dr Jervis Good/NPWS390 that diminution, by fish farm generated sea lice, of migratory salmonid numbers in the Dromagowlane/Trafrask River System would deprive the FwPM population there of larva hosts, ALAB sought out and preferred the advice of the Marine Institute that FwPM larva hosts in the Dromagowlane/Trafrask River are predominantly non-migratory brown trout which will (all agree) be unaffected by sea lice.

• the Marine Institute has demonstrated itself to be consistently supportive of salmon farming generally and this salmon farm in particular.

• the Marine Institute is primarily expert in matters marine – the sea and tidal waters – as opposed to IFI, whose primary expertise lies in freshwater fisheries and rivers.

• the manner in which ALAB managed its selection of sources of information and expert advice betokened bias.

151. I will say of this however that, even had it been pleaded, IFI would have had an uphill struggle given the Marine Institute

• is itself a statutory body391 of whose policies and objectives ALAB must keep itself informed.392

• is a prescribed consultee in Aquaculture Licence applications.393

• is identified as a body whose officers may be summoned by ALAB to oral hearings.394

• is identified as a body with which IFI “shall, as appropriate, co-operate and co-ordinate” in research and experimental work as to inland fisheries.395

• tendered its advice of 15 February 2019 by an expert – Dr Jackson, its director of fisheries. It was for ALAB and is not for the court, to choose between competing expert advice.

As to the “allegation”, if such it can be said to be, that the Marine Institute has demonstrated itself to be consistently supportive of salmon farming generally and this salmon farm in particular, at least as to the latter one might as well say the converse of IFI. Indeed, though it has no particular legal consequence, a

389 Shatter v Guerin [2019] IESC 9, [2021] 2 IR 415.

390 National Parks & Wildlife Service – part of the Department of Housing, Local Government & Heritage.

391 Established by the Marine Institute Act, 1991.

392 s.39 1997 Act.

393 Aquaculture (Licence Application) Regulations, 1998 (S.I. No. 236 of 1998) Art 10.

394 s.57 1997 Act.

395 s.7(7) Inland Fisheries Act 2010

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

striking feature of these proceedings is, what the IFI at least portrays as, the great schism between two statutory bodies – IFI and the Marine Institute – on the subject-matter of salmon farming.

152. The Sweetman Applicants make a similarly unpleaded submission as to the referral to the Marine Institute for comment of the report of bird expert Dr Gittings.396

153. All that said, I have thought it only fair to some against whom bias was alleged in open court to address the merits of some unpleaded allegations.

Bias – Structural

Structural Bias – the Statutory Architecture

154. S.23 of the 1997 Act prescribes a chairperson and 6 other members of ALAB. They are remunerated for their work at rates determined by the Minister with the consent of the Minister for Finance.397 Elected politicians may not be appointed.398 The Chairperson, who has a casting vote, is appointed by the Government, not the Minister.399 The Government is at large as to who to appoint as chairperson.400 As to the 6 other members, s.23 requires the Minister to prescribe at least 2 organisations representative of each of 6 listed classes of interest. Those classes are the promotion and business of aquaculture,401 the conservation, development and protection of wild fisheries,402 physical planning and development,403 the protection and preservation of the environment and amenities,404 the promotion of general economic development405 and community development.406 Those organisations are required407 to nominate and recommend408 candidates suitable for appointment to ALAB and the Minister must appoint a member from each class.409 ALAB members can be reappointed.410 Though not expressly stated in s.24, a purposive interpretation of the statutory architecture suggests that reappointed members must be regarded as drawn

396 “Bird Expert's Report: Briefing Note Bird impact assessment”, February 2018.

397 s.25 1997 Act.

398 s.23(1A) 1997 Act.

399 s.23(2) 1997 Act.

400 Members of the Oireachtas, the European Parliament and local authorities, are excluded – s.23(1A).

By the Prescribed Organisations for ALAB Regulations 2017 the Minister nominated IFA Aquaculture, Irish Salmon Growers Association Limited, Irish

Shellfish Association.

402 By the Prescribed Organisations for ALAB Regulations 2017 the Minister nominated Angling Council of Ireland Irish Fish Producers Organisation.

403 By the Prescribed Organisations for ALAB Regulations 2017 the Minister nominated Engineers Ireland, The Irish Planning Institute Ltd, Royal Institute

of Architects Ireland, The Royal Town Planning Institute, Irish Branch.

404 By the Prescribed Organisations for ALAB Regulations 2017 the Minister nominated An Taisce, Coast Watch Europe — Ireland Branch, The Irish

Environmental Network, The Irish Tourist Industry Confederation

405 By the Prescribed Organisations for ALAB Regulations 2017 the Minister nominated Irish Business and Employers Confederation Limited, Irish Congress

of Trade Unions, Irish Small and Medium Enterprises Association Limited, Small Firms Association, The Chambers of Commerce of Ireland Limited.

406 By the Prescribed Organisations for ALAB Regulations 2017 the Minister nominated, Muintir na Tire, Irish Rural Link.

407 s.56(6).

408 Each organisation must state the reasons why, in the opinion of the organisation, each nominee is suitable for appointment – s.23(6).

409 s.23(4) 1997 Act.

410 s.24(2) 1997 Act.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

from the class of the organisation by which they were originally nominated, so as to preserve the representation on ALAB of the interest represented by each class. More detailed provisions govern situations such as Ministerial rejection of all candidates nominated in a class, temporary appointments during temporary absences rendering ALAB inquorate and removal of members. At least generally, these provisions seek to preserve representation on ALAB of the interest represented by each class.

Structural Bias – SWI’s Case

155. SWI challenges the constitutionality of s.23 of the 1997 Act and the Prescribed Organisations for ALAB Regulations 2017411 – the statutory architecture for appointment of ALAB members – as constitutionally deficient for what SWI calls objective structural bias or institutional bias. Essentially, SWI says that the statutory architecture is such that the membership of ALAB is predominantly made up of people nominated by organisations that promote development. It pleads that “With 4 pro-development nominees and 2 pro-conservation nominees, this creates a structural bias in favour of development and against conservation.” It says the statutory scheme enables the Minister to pack ALAB with supporters of aquaculture.

156. SWI cites the Minister’s role as promoter of food production and the seafood industry, in disbursing funding for aquaculture development and calls in aid the Government’s adoption and publication of a nonstatutory policy in favour of aquaculture (“Harnessing our Ocean Wealth, 2012”) which sets out 3 goals: 1, “a thriving maritime economy, whereby Ireland harnesses the market opportunities to achieve economic recovery and socially inclusive, sustainable growth”; 2, “healthy ecosystems that provide monetary and nonmonetary goods and services (e.g. food, climate, health and well-being);” and 3, “to increase our engagement with the sea. Building on our rich maritime heritage, our goal is to strengthen our maritime identity and increase our awareness of the value (market and nonmarket), opportunities and social benefits of engaging with the sea.” SWI observes that though the three goals are stated to be of equal importance, the economic goal is goal 1, and the environmental goal, goal 2, is expressed in terms of goods and services, of a monetary and non-monetary nature. Goal 3 is also expressed in terms of value, market and non-market. SWI submits that “an ordinary reasonable person will read this as development first and sustainability second”.

157. I may as well address this last submission now to summarily reject it. Leaving aside whether the Government would be, as a matter of law and policy, entitled to prioritise development over sustainability, in my view, an ordinary reasonable person will read this policy very differently to SWI. I find SWI’s argument formalistic and unconvincing. It incorrectly purports to read a Government policy as if it were a statute – and even were it a statute it would not bear the meaning for which SWI contends. The Government is entitled as

411 The Fisheries (Amendment) Act 1997 (Prescribed Organisations for the Aquaculture Licences Appeals Board) Regulations 2017 (S.I. No. 588 of 2017).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

a matter of policy to generally favour aquaculture and is not, without evidence, to be disbelieved in its assertion that the three stated goals are of equal importance merely because of their order of enumeration in the policy document or on foot of a strained and artificial interpretation of goal 2. The policy clearly does favour aquaculture but also records that “Balancing the views on the development potential of the sector were concerns over the potential environmental interaction/impacts of aquaculture (e.g. disease, wild fish interactions); the need for adequate monitoring to ensure compliance with environmental regulations; and the need to consider the balance/interaction with other sectors.” Government policy generally favours many things on which views for and against are common and even controversial – data centres and some types of high density housing being two topical examples. Indeed, though the policy cited is non-statutory, it is common for statutory schemes (for example the Planning Acts) to require independent decisionmakers to have regard to government policy and s.61 of the 1997 Act contemplates ministerial policy directives as to aquaculture policy to that end in aquaculture licensing, including in appeals to ALAB. It is inherent in policy that it makes at least broad choices as between options and competing priorities. That is perhaps especially so when the choices are controversial – or, to put it another way, are “hard” choices. Such choices have to be made and that is Government’s job. The resulting policies do not represent bias in the Government or induce bias in quasi-judicial decisionmakers who have regard to them merely because they disappoint some who vehemently disagree with them. The existence of such policies and the regard to be had to them does not disable decision-makers on quasi-judicial tribunals either generally or because their members are appointed by Government or ministers of Government. Nor would much of the work of such tribunals be possible if it did.

158. Nor have I overlooked the possibility that, as informed by a broader view of relevant context, SWI’s interpretation of the policy could be upheld or that the terms of the policy could buttress other arguments for a finding of bias even if insufficient to stand as bias by itself. Neither possibility strikes me as plausible.

159. SWI asserts that in all the circumstances described above those of ALAB’s members who might be expected to articulate a contrary view, prioritising aquaculture’s alleged environmental detriments, will be outvoted on, or even excluded from membership of, ALAB. SWI says that, generally, appointees involved in the promotion and business of aquaculture, physical planning and development, the protection and the promotion of general economic development are likely, if only via “cognitive bias”, to carry that promotion into the exercise of their functions as ALAB members by favouring the grant of aquaculture licenses. More specifically, it says that by prescribing, in the class of those concerned with the protection and preservation of the environment and amenities, only the Angling Council of Ireland and the Irish Fish Producers Organisation, and in appointing only one ALAB member from that class, the Minister contrives a situation in which, to the exclusion of the Angling Council of Ireland, the Minister can appoint from this class an ALAB member likely to support aquaculture – and likely not to articulate concerns as to risks to wild salmon posed by aquaculture.

160. SWI also argues that structural bias lies in the fact that ALAB members determine appeals from decisions of the very Minister who appointed them, and may reappoint them, such that in deciding appeals SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

they will tend to favour the Minister’s view. It adds that, as a party to such appeals and entitled to make submissions therein, he/she maintains an ongoing supervision ALAB as it determines appeals against his

decision. This situation renders appeals neither independent nor impartial.

161. SWI invokes in these regards the following Articles of the Constitution:

• 40.3.1 & 34.1 – as to the personal rights of the citizen, as to the administration of justice in courts and by

judges and in public and as incorporating the principles of natural and constitutional justice including the right to an impartial and unbiased determination.

• 15.4.2 – as to the invalidation of laws repugnant to the Constitution.

162. In addition SWI invokes Art.47 CFREU which, it says is directly effective (citing Deutsche Umwelthilfe412) and reads, in part: “Everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal previously established by law.” SWI invokes Article 47 CFREU also for the right to an effective remedy. It invokes ss.3 and 5 of the European Convention on Human Rights Act 2003 and Article 6 of the EIA Directive and Article 6 of the Aarhus Convention as to rights of public participation to substantively the same effect in seeking a declaration of incompatibility ss.23(3)413 and (4)414, 28(4)415, and 44(2)416 of the 1997 Act.

163. SWI also challenges s.23 of the 1997 Act also as in breach of Article 9A of the EIA Directive which requires competent authorities perform EIA objectively and do not find themselves in a situation giving rise to a conflict of interest and asserts that the EIA done by ALAB was vitiated by conflict of interest and lack of objectivity. It makes similar points by reference to AA and WFD417 issues.

Structural Bias – the Sweetman Applicants’ Case

164. The Sweetman Applicants, relying on Article 9A of the EIA Directive 2014, echo SWI’s case in structural bias. They plead418 ALAB’s 2017 submission to the Independent Aquaculture Licensing Review Group as, in effect, admission of bias. They plead also the following recusals by ALAB members “for reasons of conflict of interest” from consideration of the Appeal:

412 Case C-873/19, Deutsche Umwelthilfe eV v Bundesrepublik Deutschland & Volkswagen AG, Judgment 8 November 2022 (Grand Chamber). §79 –

“Article 47 of the Charter is sufficient in itself and does not need to be made more specific by provisions of EU or national law in order to confer on

individuals a right which they may rely on as such.”

413 Minister to prescribe organisations in various classes to nominate candidates for ALAB membership.

414 Minister to appoint one member of each class to ALAB membership.

415 ALAB makes decisions by Majority vote.

416 Minister may make submissions in Appeal.

417 Citing the WFD Directive 2000/60 and Directive 2008/105/EC as to environmental quality standards in the field of water policy.

418 Grounds §E.36.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Francis O’Donnell: He had been CEO of the Irish Fish Producers Organisation. IFPO members include food suppliers to salmon farms. He recused himself on 9 October 2019 as he had accepted a new role with IFI, one of the objectors to the Salmon Farm, and he retired on 30 November 2019.

• Brendan Brice: He recused himself at the start of the appeal in October 2015, due to his prior directorship, as an engineer, of RPS. He retired from ALAB on 16 June 2019.

• Professor Owen McIntyre: He recused himself in September 2017 on presentation of his report of the oral hearing and he retired from ALAB on 16 June 2019. This left unoccupied the seat for a nominee organisations representative of persons concerned with the protection and preservation of the environment and amenities.

165. The Sweetman Applicants observe that, at times, vacancies and recusals reduced the ALAB members considering the Appeal to four, including the chairperson, instead of seven, which upset the balance of representation for which the 1997 Act provided and, he says, resulted a predominance on ALAB at critical stages of representatives of organisations concerned with the promotion of aquaculture and economic development. The Sweetman Applicants do not analyse this allegation in either temporal terms or in terms of the representations in question.

Structural Bias – Opposition

166. The State, ALAB and MOWI, in combination dispute the claim of structural bias and the unconstitutionality of s.23 of the 1997 Act. Mere traverses apart, they plead and aver as follows:

i. ALAB is an independent statutory body – not subject to the supervision of the Minister in the hearing of appeals or otherwise.

ii. ALAB is required by law to act fairly and without bias.

iii. The content of s.23 and the means of its implementation – it is not necessary to recite those here.

iv. ALAB’s members were appointed in accordance with law and s.23 and the Minister’s obligations thereunder such that there can be no reasonable apprehension of bias.

v. Estoppel by acquiescence as any defect in the form of structural bias was known since the commencement of the appeal in September 2015.

vi. Applicants’ characterisation of the organisations prescribed under s.23(3) as representative of those who promote development, aquaculture or conservation as the case may be is erroneous.

vii. Even if those characterisations are correct they do not imply bias.

viii. There is no basis in law or fact for any allegation, if made, that an ALAB member’s membership of a particular grouping determines or influences the outcome of his or her decision-making in respect of appeals.

ix. ALAB members are expected to, and in this case, did, consider an appeal fully on its merits and make a decision which is in the public interest, regardless of which prescribed organisation they belong to.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

x. Though entitled to, the Minister made no submissions to ALAB in the Appeal. The Minister did respond to a s.47 Request as to reports concerning a salmon escape in Bantry Bay on 1 February 2014. That request was withdrawn.

xi. The Marine Institute is an independent state agency established pursuant to the Marine Institute Act 1991. The same is true of the Agriculture Appeals Board and it has no function as to ALAB.

xii. If ALAB is not an independent and impartial tribunal, which is denied, the Applicant nevertheless retains an effective remedy in judicial review for any violation of its rights and freedoms.

167. Mr Dowling for ALAB419 asserts that:

i. ALAB’s submission to the Independent Aquaculture Licensing Review Group in February 2017 expresses ALAB’s concerns generally and not as to any specific appeal. It indicates that ALAB was alive to potential issues and how they should be and were being addressed.

ii. ALAB’s submission cited sharing an advisor with the Marine Institute as an example of a potential issue. It never happened. Nor has the Marine Institute acted as advisor to ALAB.

iii. ALAB did not share “office accommodation” with DAFM. It shares a building, within which it maintains its own office and systems.

iv. Currently all ALAB’s secretariat/staff420 work exclusively for ALAB. ALAB has had a full-time secretary since 2018. Prior to that ALAB did not share staff simultaneously with DAFM: ALAB’s secretary – who has no substantive role in the decision making process – was seconded to ALAB three days a week to work exclusively on ALAB matters and the other two days a week for the Agriculture Appeals office – which is administratively unrelated to aquaculture appeals or licensing.

Structural Bias – Discussion & Decision

168. The doctrine of judicial restraint requires that deciding the constitutionality of legislation is to be avoided unless necessary – unless the case cannot be decided on other grounds. That suggests that in this case I should not decide the constitutionality of s.23 of the 1997 Act if the impugned decisions are quashed on other grounds. Also, the State submits that certiorari will remedy any structural bias such that consideration of the constitutionality of s.23 is unnecessary and so SWI lacks standing on the issue.

169. However, as to an allegation of unconstitutionality specifically for structural bias, a quashed Aquaculture Licence decision is likely to be remitted to the allegedly structurally biased decision-making body for re-decision. In those circumstances it seems to me that SWI has standing to raise this constitutional issue, and it is necessary to decide it, whether or not the impugned Aquaculture Licence is quashed on other grounds.

419 Assistant Principal, Secretariat to ALAB. Affidavits sworn 28 June 2022.

420 An assistant principal, higher executive office (secretary), an executive officer and two clerical officers.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

170. Hogan et al421 refer to the phrase “structural bias” or “institutional bias” as “jargon” merely making

explicit a judicial policy discernible in the caselaw and referring to a situation in which, because of the statutory structure and functions of an institution, a possible appearance of bias is inevitably created. The authors say that the courts usually hold that, because of this inevitability, something like the necessity doctrine should be applied, to afford a substantial latitude for the decisions of such institutions.

Structural Bias is Objective Bias – Ardagh

171. An allegation of structural bias is a form of allegation of objective bias. The State cite Ardagh422 to the contrary but the full relevant quotation from the dissenting judgment of Keane CJ, on part of which the State relies, is as follows as to structural bias:

“It is important to distinguish this argument from an entirely separate argument which I will consider at a later stage and which related to what is alleged to be "objective bias" on the part of specified members of the committee in this case and, in particular, the Chairman and Deputies Howlin, Magennis and Shatter. It was said that comments they had made to the media and interventions by them during the hearings of the committee demonstrated … bias. ”

172. It seems to me that Keane CJ was not distinguishing structural from objective bias generally but was distinguishing structural bias from objective bias by reason of specific acts of specific committee members. Though they did not find it necessary to decide it, that is clearly how the unanimous the divisional High Court had understood the allegation – as of “structural inabilities in the conduct of such an inquiry by

parliamentarians by reason of objective bias deriving from their representative roles.”423 – as of “the alleged

inability of elected representatives to conduct adjudications of the type in suit because of perceived or structural bias arising particularly from their representative functions as elected parliamentarians.”424 And clearly, the touchstone of structural bias is the essentially same as that of objective bias generally: the perception of bias – the reasonable and realistic apprehension of reasonable people.

173. Turning to the substantive treatment of structural bias, Keane CJ saw “no reason why Dáil deputies and senators should not approach their participation in committees of this nature with minds unclouded by

421 Hogan, Morgan, Daly; Administrative Law in Ireland 5th Ed. 2019 §14-41 et seq.

422 Maguire v Ardagh [2002] 1 I.R. 385. The case concerned a challenge to the power of an Oireachtas sub-committee to investigate the “Abbeylara

incident” in which a man was shot dead by the gardaí. The investigation was declared ultra vires the Oireachtas as liable to result in adverse findings of

fact and conclusions (including a finding of unlawful killing) as to the personal culpability of an individual not a member of the Oireachtas so as to impugn

his or her good name.

423 p433.

424 p466.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

pressures from their constituents or, for that matter, from the political parties which they represent”. He saw no reason to deny them the benefit of the presumption established in cases such as the East Donegal case425 “that they will conduct their proceedings in accordance with the requirements of natural and constitutional justice.”

174. Murphy J, also dissenting, did not address structural bias in terms. But he validated the subcommittee’s power to conduct the proposed inquiry. He anticipated “immense practical difficulties in conducting such an inquiry in accordance with the requirements of natural justice and fair procedures. However, that defect is not inherent in the nature of the inquiry itself and it would be premature to express any opinion at this stage as to whether the difficulties which I foresee will be overcome.” This view seems incompatible with his having accepted any allegation of structural bias.

175. McGuinness J did not explicitly analyse the bias issue in terms of structural bias. But she did characterise respondent counsels’ submissions as to objective bias in essentially structural terms as follows:

“The contention of the applicants was that members of the sub-committee should not act simply because they have been elected. The respondents submitted that the mere fact that they had been elected and might again stand for election should not, on a reading of the case law, be sufficient to brand them as biased. To suggest that the members of the sub-committee would tailor their considerations on the Abbeylara incident in such a manner as to win favour with their constituents would be to suggest that no elected person could ever exercise independence while acting in his or her official capacity. This was an untenable contention.”

That said, McGuinness J expressed her concerns in essentially non-structural terms: “… it must in my view be unacceptable for members of an inquisitorial committee such as this to make public comments on the subject matter of the inquiry both in the run up to the inquiry and during its actual currency. ……. In at least some cases members expressed strongly held views which would undoubtedly give rise to a reasonable apprehension of bias.”

McGuinness J later appeared to discount the assertion of structural bias:

“The applicants say that given the nature of the work and life of members of the Oireachtas, this must mean that they are automatically incapable of carrying out an inquiry such as that on Abbeylara. I would not go so far. Members of the Oireachtas have been given highly important constitutional duties; they have been elected by their constituents to fulfil these duties. If, whether under statute or otherwise, they have been properly mandated to carry out an inquiry, I consider that they cannot be disabled from so doing by an automatic assumption of objective bias …”

425 East Donegal Co-operative Livestock Mart Ltd. v Attorney General [1970] I.R. 317. The Plaintiffs objected to the requirement, imposed by the Livestock

Marts Act, 1967, of a licence to sell livestock by auction. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

……..

“It is not, however, necessary for this court to make the general declaration sought by the applicants in their notice to vary. This would be to make an assumption of automatic objective bias arising from the very nature of the political and representational role of members of the Oireachtas. Apart from the undesirability in principle of making such an assumption, there was no evidence before the Divisional Court which would permit such a far-reaching finding.”

176. Geoghegan J took a view dramatically different to that of Keane CJ. He cited comments of a distinguished former Secretary of the Department of Finance, who considered that “It is difficult for

politicians to cease, even temporarily, to be politicians.” He saw an “inherent likelihood of structural bias or

at the very least the obvious difficulties in avoiding objective bias in any given case” and a “likely frequent perception of structural bias” and observed that “The joint committee, as was customary, was balanced in its membership according to the strength of the political parties. It would only be in rare circumstances that a body composed in that way would be perceived by reasonable members of the public as capable of independent arbitration.” No doubt the reasonable observer is reassured by the fact that s.23(1A) prohibits elected representatives from membership of ALAB.

177. The other judges in Ardagh, Murray, and Hardiman JJ, did not decide the issue of structural bias.

178. In the foregoing circumstances is somewhat difficult to say that Ardagh decided anything definitive about structural bias. But I do not think it is authority for the State’s assertion that structural bias is not a form of objective bias and I think that assertion incorrect. And it is difficult to avoid being struck by the reliance of Keane CJ on the East Donegal presumption “that they will conduct their proceedings in accordance with the requirements of natural and constitutional justice.”

Pigs Marketing Board, East Donegal, Galvin v DPP & Cahill v Sutton & Mohan v Ireland

179. Since the Pigs Marketing Board case426 it has been axiomatic “that a law passed by the Oireachtas, the elected representatives of the people, is presumed to be constitutional unless and until the contrary is clearly established.”427 The word “clearly” was not accidentally included in this passage. Also, and as Ní Raifeartaigh J noted in Galvin v DPP:428

426 Pigs Marketing Board v Donnelly (Dublin) Ltd [1939] IR 413, [1939] IR 413 (Digest).

427 Pigs Marketing Board v Donnelly [1939] I.R. 413. Cited, for example, by the Supreme Court In Damache v Director Of Public Prosecutions [2012] 2 IR

266; Re The Employment Equality Bill, 1996 – [1997] 2 IR 321; see also O’Doherty & Waters v Minister For Health And Others [2022] IESC 32 (O’Donnell

CJ).

428 Galvin v Director of Public Prosecutions [2020] IECA 319 (Court of Appeal (criminal), Collins J, Ní Raifeartaigh J, 19 November 2020).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“The parties must ensure that the court which is to hear the constitutional challenge has sufficient evidence before it to enable it to properly address the substantive constitutional arguments in a reasonably concrete evidential context. The court should not be asked to decide matters of constitutional importance in an evidential vacuum …”429

Ní Raifeartaigh J required that the evidence establish, beyond the requirements of standing, “a sufficiently concrete context” within which to decide the constitutionality of the impugned statute.

180. As I have said, SWI did not adduce evidence or make submissions to the effect that Minister had in fact operated the statutory architecture to pack ALAB with supporters of aquaculture or in breach of the presumption, established in the East Donegal case430 as an aspect of the presumption of constitutionality of legislation,

“…. that the Oireachtas intended that proceedings, procedures, discretions and adjudications which are permitted, provided for, or prescribed by an Act of the Oireachtas are to be conducted in accordance with the principles of constitutional justice. In such a case any departure from those principles would be restrained and corrected by the Courts.”

181. I have already mentioned that, in Ardagh, Keane CJ, and by inference Murphy J, would have applied this East Donegal presumption to the Oireachtas Sub-Committee as a basis for rejecting an allegation of objective bias in the particular circumstances of that case. Geoghegan J would not have done so and the other judges did not decide the issue. But none doubted the general correctness of the presumption. S.23 of the 1997 Act must be presumed to rest in part on that East Donegal presumption and its substantive content applies to ALAB.

182. Cahill v Sutton431 originally, and Mohan v Ireland432 more recently, are authority that a plaintiff challenging the validity of a statute must generally433 show that their interests had actually been or would be434 adversely affected by the operation of the statute. This is a threshold requirement designed to exclude those with no real connection to the statute. O’Donnell J, in Mohan, in an obiter from which there was no dissent in a court of five, said435 that:

429 Ní Raifeartaigh J.

East Donegal Co-Operative Livestock Mart Limited v The Attorney General, [1970] IR 317.

431 [1980] IR 269.

432 [2019] IESC 18, [2021] 1 IR 293.

433 The rule is one of practice and subject to exceptions.

434 “stood in real or imminent danger”.

435 And citing Henchy J in the High Court in The State (P. Woods) v Attorney General [1969] IR 385, to the effect that declaring the invalidity of legislation

“… unmakes what was put forth as a law by the legislature but, unlike the legislature, it cannot enact a law in its place. It is clear that if this power, which

may seem abrogative and quasi-legislative, were used indiscriminately it would tend to upset the structure of government.”

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“The invalidity of legislation is …. a very significant disruption of the legal order which operates in a blunt and essentially negative way. It simply removes a law, or an aspect of the law, can put nothing its place, and yet can throw into question transactions taken on foot of the provision. ……… permitting a challenge to the constitutional validity of a piece of legislation should not, therefore, be taken lightly and should only be taken when a person could show they were adversely affected in reality.”

And later:

“This normally requires a real case or controversy which the parties require (rather than simply desire)

to be resolved in order to establish and justify the court's exercise of jurisdiction, and the possibility of the invalidation of legislation. Accordingly, it is necessary to show adverse effect, or imminent adverse effect, upon the interests of a real plaintiff. This has the further benefit, as Henchy J observed in Cahill v Sutton,436 that:

“… normally the controversy will rest on facts which are referable primarily and specifically to the challenger, thus giving concreteness and first-hand reality to what must otherwise be an abstract or hypothetical legal argument.”

However, O’Donnell J notes that Irish law does not impose “a very strict requirement of establishing actual

harm before a legislative provision may be challenged” – “It is enough that the plaintiff is, or can plausibly

claim to be, affected or likely to be affected in a real way.” And effect on the Plaintiff’s interests – broadly

envisaged and as opposed to rights – suffices. However, it seems that a “real and significant effect” is required.

183. O’Donnell J also addressed the significance of a consideration of the evidence (and, one must infer, any absence of evidence) in the particular case. He said:

“There is no doubt that if, on hearing evidence, a court is satisfied that the impugned provisions had no effect upon a person, let alone on their interests or rights, that that would be fatal to the claim proper and, indeed, to the plaintiff's standing to bring the claim ….”437

184. Forde & Leonard 438 frame the requirement as being to show “a distinct loss due to the impugned measure” on the rationale that “the function of the courts is not to decide hypothetical questions of constitutional law. Requiring a distinct stake in the issue being raised ensures that scarce judicial resources are not wasted in resolving theoretical and moot points, and should guarantee that both sides of the argument are adequately prepared and presented.” They say that “requiring such impact ensures that the

436 [1980] IR 269, at §282.

437 Unless one of the exceptions to the primary rule of standing can be established.

438 Constitutional Law of Ireland Forde & Leonard, Bloomsbury Professional, Third edition, July 2013 §30.04.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

question at issue is decided against the background of actual events, which usually facilitates reaching the correct conclusion on the points in issue.” And “in the absence of circumstances showing loss inflicted or pending, cases tend to lack the force of urgency and reality.”

185. Of course, where the issue is objective bias, the Cahill/Mohan requirement does not require adverse real effect in the form of the operation of actual bias in the decision-making process. Objective bias suffices. But as applied to the present case, the Cahill/Mohan requirement seems to me to require at least a demonstration that the actual composition – as opposed to merely the machinery for appointment of – the decisionmaker is such as to demonstrate objective bias. If, as to an allegation of structural bias, there is no such demonstration, I cannot see how a Plaintiff can show actual adverse real effect – past or prospective – by reason of the operation against their interest of the allegedly unconstitutional statute. It seems to me that such a view is also consistent with the East Donegal439 presumption described above.

186. To strike down as repugnant to the Constitution the statutory scheme for making appointments to ALAB, I would require, and I do not have before me, evidence of unfair actual effect on the Applicants by the operation of that scheme both as a matter of standing and, assuming standing, to provide “a sufficiently concrete context” within which to decide the constitutionality of the impugned statute to the point where its repugnance to the Constitution has been “clearly established”. I use the word “actual” here in the attenuated sense of demonstration of objective bias (not actual bias) by reason of the actual constitution of ALAB. Clearly, the mere fact of loss of an appeal before ALAB cannot provide the necessary concrete context. I have been directed to no evidence that the Minister has, in the operation of the statutory scheme and in making appointments to ALAB, acted other than in accordance with constitutional norms. And any such fact, would have to be “clearly established” – Burke.440

187. In effect SWI asks me to presume, merely because (assuming for the sake of argument) the statutory architecture enables the Minister to pack ALAB and despite the absence of evidence that he did so, that the objective onlooker would have a reasonable apprehension or suspicion that the Minister had in fact packed ALAB such that ALAB might have been biased. I respectfully reject that invitation to reverse the presumption of constitutionality that public officials, such as the Minister, will operate statutory schemes in a manner conforming to constitutional norms. Had there been evidence that the Minister had packed ALAB the position might have been different – but there is none. To pick a posited example, I have no evidence that the Minister invariably or disproportionately often appoints a nominee of the Irish Fish Producers Organisation to the exclusion of nominees of the Angling Council of Ireland.

439 East Donegal Co-Operative Livestock Mart Limited & Others v The Attorney General [1970] IR 317. The Plaintiffs objected to the requirement imposed

by the Livestock Marts Act, 1967, of a licence to sell livestock by auction.

440 Burke v Minister for Education and Skills [2022] 1 I.L.R.M. 73, Charleton J.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

188. I respectfully reject the claim in structural bias by reason of the Applicants’ failure to make the required demonstration in evidence of actual441 effect on their interests. However, lest I am wrong in that dismissal, I will consider the issue further.

Structural Bias – “Balanced Tribunals”

189. It is clear from s.28(5) that the general qualification for appointment to ALAB is “the appropriate knowledge or experience in matters relevant to the functions of the Board”. An essential tension underlying allegations of structural bias as it relates to expert tribunals is that, on the one hand the appointment of experts (using that word in a broad sense) to such tribunals is seen as enhancing and even, where the tribunal’s subject-matter relates to arcane and/or technical issues, as essential to, good decision-making. On the other hand, what makes a person an expert – by way of qualifications and experience – may often be considered to have identified them in greater or lesser degree with a particular point of view on the tribunal’s subject-matter. Acquisition of expertise does not occur in a vacuum. The tension may be even greater where the legislature decides to draw the tribunal’s membership from, not merely experts, but from interest groups who may be expected, in broad terms, to nominate for membership persons generally holding a relevant point of view. The difficulty may be greater again where the area in which the tribunal will act is arcane or niche, such that the pool of knowledgeable persons is small. I have no reason to doubt ALAB’s 2017 submission to the Independent Aquaculture Licensing Review Group, which described conflict of interest as “an issue which regularly concerns the Board” and observed, “The fact is that there is a small pool of available people with the necessary expertise in the area of aquaculture.” Each such person will have inevitably formed at least some general views on the topics relevant to their decisions. Disqualification of potential appointees merely on that account would quickly depopulate, and so paralyse, many such tribunals. The chair and all ALAB members are engaged on a part-time basis442 and, as Hogan et al observe: such tribunals “ … as a matter of practicality ….. will often involve part-time and/or unpaid committees or boards composed of members who do not live in an ivory tower and must earn their living elsewhere as professionals or business people, with several linkages or associations. Thus, all kinds of linkages and associations occur and, arguably, are lawful.”443

190. Murphy J in Orange #2444 cited American sources as to judicial bias in terms which, mutatis mutandis, seem relevant here.

441 In the attenuated sense set out above.

442 See ALAB’s 2017 submission to the Independent Aquaculture Licensing Review Group and ALAB annual reports. That ALAB members serve part-time

is also readily apparent from the records of fees paid to ALAB members as disclosed in its annual reports. That this is so is also apparent from the facts relating

to the overpayment of fees Prof McIntyre, the recoupment of which was premised on the “one person one salary principle” as to public servants.

443 §14-92, in considering the doctrine of necessity.

444 Orange v The Director of Telecommunications Regulation and Meteor Mobile Communications Limited, (No. 2) [2000] 4 IR 159 at §242.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“… the proposition that justice should be blind and that the judge should be wholly untouched by extraneous considerations is not merely unattainable: it is undesirable. In Laird v. Tatum445 Rehnquist

J. (as he then was) considered a motion to disqualify himself, under the appropriate United States legislation, on the grounds of bias. The distinguished judge conceded that prior to his appointment, and in his capacity as an expert witness, he had appeared on behalf of the Justice Department before Senate hearings and expressed views material to the issues in the case pending before him. Having explained that most justices come to the bench no earlier than their middle years and that it was therefore probable that they would have expressed views in different forums on constitutional issues he went on to say at p. 835:-

‘Proof that a justice's mind at the time he joined the court was a complete tabula rasa in the area of constitutional adjudication would be evidence of lack of qualification, not lack of bias.’”

Murphy J cited “a particularly attractive passage” from In re JP Linahan446 in the following terms:-

"Democracy must, indeed, fail unless our courts try cases fairly, and there can be no fair trial before a judge lacking in impartiality and disinterestedness. If, however, 'bias' and 'partiality' be defined to mean the total absence of preconceptions in the mind of the judge, then no one has ever had a fair trial and no one ever will. The human mind, even at infancy, is no blank piece of paper. We are born with predispositions; and the process of education, formal and informal, creates attitudes in all men which affect them in judging situations, attitudes which precede reasoning in particular instances and which, therefore, by definition are prejudices."

191. Choice of such a system of populating a tribunal from interest groups as was adopted in the case of ALAB is, at least generally, a legitimate political decision (though not the only possible) in providing for

decision-making in areas of controversy. It involves choices as to what interests should be represented on such tribunals and in what respective degrees. The interests in question may be multifarious and range widely as to their points of view. It is often – perhaps even usually – not a simple matter of identifying interests “for and against”. The primary purpose may legitimately be that varied voices and experience are heard rather than that a particular or fine balance be achieved in the consideration of each and every

decision the body in question must make. In my view, the legislature in adopting such a system and the government in operating it, have a large margin of appreciation. The courts should only very reluctantly interfere with such decisions.

192. The law recognises that such tensions cannot be absolutely de-risked. The law answers that reality by imposing on members of quasi-judicial tribunals duties of impartiality, independence and openmindedness. Those duties are Constitutional in origin in that they are directed at ensuring constitutional

445 (1972) 409 U.S. 824.

446 [1943] 138 F. (2d) 650

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

justice in proceedings before such tribunals. The presumption of constitutionality extends to the manner in which public officials operate statutory processes – East Donegal. In judicial review it is an element of the presumption of validity of impugned decisions. It extends to the decisions made by ministers in effecting the appointment process. That is, it is presumed that the minister has not “packed” the tribunal and that he/she will appoint persons not merely knowledgeable but of “character and standing”447 fit to exercise quasijudicial office. Notably, in the present case no allegation is made that the Minister in fact “packed” ALAB – the challenge is on the more rarefied basis that the reasonable observer would be troubled that the statutory architecture allows him to do so if he/she wished. Nor has the fitness for office of any ALAB member been impugned.

193. The presumption also extends to the decisions made by such appointees in the exercise of their quasi-judicial functions. As to tribunal members and as Hogan et al observe, their powers “must be exercised

reasonably, objectively and judicially”.448 The word “judicially” necessarily implies objectivity, independence,

impartiality and reasoning. All this makes it clear that members of tribunals such as ALAB are representatives of their nominating bodies only in the sense that those bodies consider them to be persons of integrity and to have the expertise and experience characteristic of the interests and fields of operation of those bodies. And, of course, those bodies only nominate candidates for appointment: the decision to appoint rests with the Minister who must similarly satisfy himself or herself of those qualities.

194. Appointees are not delegates of the bodies which nominated them. They are not put in place to “do a job”, as it were, to protect the interests of, or represented by, those bodies. They cannot take instruction, or even countenance lobbying, from those bodies as to the performance of their role as ALAB members. When they attend ALAB they must, as it were, leave their loyalties at the door and have the strength of character required to do so. Indeed, Hogan et al449 identify ALAB as “significant example” of a quasi-judicial tribunal “which has a high level of institutional independence”.

195. Hogan et al450 also consider what they call “Balanced Tribunals”, saying “There is a distinction between a situation in which members are appointed for their particular expertise and where they are appointed as members of interest groups, in order to constitute a “balanced” tribunal. …. The most sophisticated attempt in this direction involves An Bord Pleanála.” There follows a description of the system of nomination and appointment to that Board which is very similar to that as to ALAB. The authors continue:

“Given the state of orthodox law, such “balanced” tribunals may not be without their dangers.

Certainly, such tribunals are designed on the basis that, where there are divergent interests, it might

447 In re the Solicitors Act 1954 [1960] IR 239.

448 Hogan, Morgan & Daly, Administrative Law in Ireland, 5th ed’n §6.33.

449 §6.37.

450 §6.43.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

seem fair that these interests should have their representatives on the tribunal. We have seen many examples of such a tribunal in Irish law. …….

Nevertheless, the conventional view is that constitutional justice requires that each individual member of a tribunal must be impartial, rather than the leaning of one member to one side being regarded as cancelling out the leaning of another member in the opposite direction. There appears to be no case in which this precise point has been taken. If this proposition were argued successfully, then many Irish tribunals would be cut down. But it is probable that a judge would recoil from this “appalling vista” and would rely heavily on the doctrine of necessity 103 to avoid reaching such a result.”

196. It seems fair to say that this is case in which the point has been taken. However, and with appreciable diffidence given the standing of the authors, I respectfully suggest that recourse to the doctrine of necessity is unlikely to be the solution and “appalling vista” reasoning, perhaps for historical reasons but they illustrate good legal reasons, has little allure – not least here, where the challenge is to the constitutionality of the structure itself. Necessity seems absent as the Oireachtas could have chosen another structure. And its invocation implies that but for the necessity bias would be found.

197. I hope not naively, I prefer to rely on the “conventional view” that each member of a tribunal must be independent and impartial, rather than the leaning of one member to one side being regarded as cancelling out the leaning of another member to the other. And in the present case there is ample evidence in most, if not all the ALAB minutes, and indeed in ALAB’s 2007 Submission to the Review Group, which described conflict of interest as “an issue which regularly concerns the Board”, that its members are conscious of those duties.

198. In other words, in basing their structural bias case, as in substance they do though they did not use the phrase, on a “balance of prejudice” requirement in the composition of ALAB, the Applicants start from a flawed premise. The correct premise is of independence and impartiality of each and all ALAB members.

199. SWI adduced no expert or other evidence in support of the assertion in submissions of operative unconscious “cognitive bias” in this case – for which evidence the cited Encyclopaedia Britannica definition of “confirmation bias” is no substitute. This argument was made at an entirely abstract level. All decisionmakers must attempt to guard against unconscious biases. It is generally accepted that few are without at least some. That is likely to be particularly so of tribunals composed of experts and/or nominees of interest groups. As pre-conceived views of a given case are not prejudgment as long as the decisionmaker’s mind stays open (Nurendale), a fortiori pre-conceived views of a general sort are not prejudgment as long as the decision-maker’s mind stays open.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

200. The Supreme Court in Radio Limerick One451 did not use the term “structural bias” and the factual allegation was not of structural bias. But it did make observations relevant to that issue as it relates to quasijudicial bodies as opposed to courts, to the appointment to such bodies of persons with special knowledge of the relevant subject-matter and to the obligation on and expectation of tribunal members that they will “resist and rise” above their backgrounds and behave independently and impartially. The Court considered that “there are bound to be instances in which an administrative tribunal charged with quasi-judicial duties may lack the appearance of strict impartiality expected from a court administering justice.”

201. The Court cited In re the Solicitors Act 1954452 As relevant here, Kingsmill Moore J, for the Court, said:

“It is true that in a hearing before the Committee a solicitor will not have the protections he would receive in a Court of justice. Complainant, tribunal and the person who conducts the complaint are inextricably interconnected. Moreover the circumstances are such as to make it difficult for the tribunal to be impartial. In many cases the person against whom a complaint is made will be a solicitor with whom members of the tribunal have had professional dealings which may have predisposed them in his favour or against him. All of the members are liable to contribute yearly to a compensation fund established under the Act to relieve or mitigate losses sustained in consequence of dishonesty of solicitors and the amount of such contribution may be increased if found necessary … so that there might be a tendency to bear hardly on a solicitor charged with dishonesty. Although the character and standing of the members is such that they can be expected to resist and rise superior to any influences which might affect their impartiality, and it is not suggested that they do not so do, the tribunal is not constituted in a manner best calculated to provide the security against bias and partiality which a Court of justice affords. In the opinion of the Court these considerations, though advanced by the appellants, are not in point. If the Committee are not administering justice the Constitution imposes no restrictions on the composition of the body.”

202. The Court in Radio Limerick One cited the maxim, nemo judex in sua causa as, in effect, expressing part of the rule as to bias and continued:

“… the application of that fundamental principle of natural justice may differ, depending on whether the body concerned is a court engaged in the administration of justice under the Constitution or an

451 Radio Limerick One Ltd. v Independent Radio and Television Commission, [1997] 2 IR 291. The IRTC had terminated the applicants broadcasting

contract for breach of statutory obligations on grounds, the most serious of which were: the “blacking” of a news item, outside broadcasts which the

respondent believed were a form of illicit advertising; the failure to provide tapes of programmes broadcast by the applicant, and, alleged refusal to co-

operate with a request to inspect. Bias was alleged on the basis that one IRTC member was the reporter whose report had been “blacked” by the applicant

and had issued unfair dismissal proceedings against the applicant. The allegation failed on the facts as that member had recused herself from the

impugned decision.

452 [1960] IR 239. The former Supreme Court held that the powers of the Disciplinary Committee of the Law Society were not “limited” within the meaning

of Article 37 of the Constitution and, accordingly, the exercise of such powers in striking off the appellant solicitors was an unconstitutional administration

of justice. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

administrative body, not so engaged, but exercising powers which can be regarded as quasi-judicial in nature.

The Court then cited part of the excerpt from In re the Solicitors Act 1954 set out above, and continued:

“In the present case, the statutory provisions already referred to clearly envisage that the membership of the commission may consist, in part at least, of persons engaged in the media who may be assumed to have a special knowledge of the matters with which the commission has to deal.

The fact that its membership is so composed may mean that, in specific instances, when it comes to deal with matters as crucial as the entering into of a contract or its suspension or termination, its decisions, although undoubtedly quasi-judicial in nature and necessitating the observance of natural justice and fair procedures, may also not have the appearance of impartiality which would be required of a court of justice. That, of itself, would not vitiate its conclusions, provided it reached them in good faith and having given the persons affected the protection of natural justice and fair procedures.”

There is a further consideration applicable to bodies of this nature which is relevant to the present case. Because of the factors to which I have already referred, a body such as the commission may not, in given circumstances, present the appearance of strict impartiality required of a court administering justice. That, however, does not relieve the commission of the obligation to take every step reasonably open to it to ensure that its conclusions are reached in a manner, not merely free from bias, but also of the apprehension of bias in the minds of reasonable people. But where, as here, a body is obliged to carry out certain statutory functions and no issue arises as to the constitutionality of the relevant provisions, a court cannot by the strict application of the legal principles already referred to prevent the body from exercising those functions, where all practical steps have been taken by it to free itself, not merely from actual bias but the apprehension of bias in the minds of reasonable people.453

203. Part at least of the logic underlying the foregoing passages from Radio Limerick One454 and In re the

Solicitors Act 1954455 seems to be that if, in the legislative discretion of the Oireachtas, it is thought desirable

that a body such as ALAB be composed of persons with relevant expertise and/or experience and/or of persons who may be considered in greater or lesser degree to represent different, and even broadly opposing, legitimate interests relevant to the subject matter on which they will come to make decisions, it follows that the Oireachtas has thought it proper, as a price of appointing such persons, to accept that they

Citing O'Neill v Beaumont Hospital [1990] ILRM 419.

454 Radio Limerick One Ltd. v Independent Radio and Television Commission [1997] 2 IR 291. The IRTC had terminated the applicants broadcasting contract

for breach of statutory obligations on grounds, the most serious of which were: the “blacking” of a news item, outside broadcasts which the respondent

believed were a form of illicit advertising; the failure to provide tapes of programmes broadcast by the applicant, and, alleged refusal to co-operate with

a request to inspect. Bias was alleged on the basis that one IRTC member was the reporter whose report had been “blacked” by the applicant and had

issued unfair dismissal proceedings against the applicant. The allegation failed on the facts as that member had recused herself from the impugned

decision.

455 [1960] IR 239.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

may, in greater or lesser degree come to their membership of the body in question with, as it were, “baggage”. Often, that baggage is the price of their expertise. But they must leave that baggage outside the door when they come to making their quasi-judicial decisions. Hence it is expected that the Minister will, in the exercise of his/her functions in accordance with statute and Constitution norms of fair procedures, prescribe only reputable bodies to nominate persons for appointment to ALAB and will appoint to ALAB from such nominees only persons of high character and integrity who can be relied upon to leave their baggage at the door.

204. The Sweetman Applicants cite North Wall Quay Property456 as having followed Radio Limerick as to structural bias. In that case the Dublin Docklands Authority (“DDA”) had statutory functions as both a development agency and also as a form of planning authority issuing to developers certificates that their proposed developments would conform to an applicable planning scheme. Finlay-Geoghegan J cited Radio

Limerick for the DDA’s “obligation to take practical steps to free itself in taking a decision on an application

from a s. 25 certificate, not merely from actual bias, but the apprehension of bias in the minds of reasonable people”. Prior to the relevant application for a certificate, the DDA executive had made an explicitly confidential agreement with the developer to recommend to the DDA board that the certificate be granted (in circumstances in which the non-executive board was considerably reliant on, and could be expected usually to take the advice of, the executive) and that if the certificate was granted, the developer would transfer certain lands to the DDA. The facts in the case seem to have been compelling in a sense not similar to the circumstances of the present case. While it is not irrelevant, I do not see that North Wall Quay Property adds to the Applicants’ case in structural bias anything not found in Radio Limerick.

205. As to the manner of the Minister’s performance of those functions of prescription of nominating bodies and appointment of their nominees to ALAB, it is an implication of the presumption of constitutionality of legislation recognised by the Supreme Court in McDonald v Bord na gCon457 that, as recognised in the East Donegal case, the Oireachtas is presumed to intend that all proceedings, procedures, discretions and adjudications permitted or prescribed by a statute are to be conducted in accordance with the principles of constitutional justice and are not to be conducted a manner amounting to a breach of a right guaranteed by the Constitution. It is also presumed, until the contrary is proved, that public officials have carried that intention of the Oireachtas into effect. Ordinarily that principle applies to proceedings, procedures, discretions and adjudications in which requirements of constitutional justice are directly engaged – such as in the Minister’s decision of MOWI’s licensing applications. It seems to me to inevitably flow from that principle that, in exercising his/her powers of appointment to the body constituted to hear appeals from those very decisions, the Minister must refrain from and is presumed to have refrained from,

456 Northwall Quay Property Holding Co. Ltd. v Dublin Docklands Development Authority [2008] IEHC 305 §99.

457 [1965] IR 217. This case stablished the “double construction rule” as deriving from the presumption of constitutionality of legislation. The Court said

“One practical effect of this presumption is that if in respect of any provision or provisions of the Act two or more constructions are reasonably open,

one of which is constitutional and the other or others are unconstitutional, it must be presumed that the Oireachtas intended only the constitutional

construction and a Court called upon to adjudicate upon the constitutionality of the statutory provision should uphold the constitutional construction. It

is only when there is no construction reasonably open which is not repugnant to the Constitution that the provision should be held to be repugnant.”

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

choosing such appointees so as to constitute ALAB to deny to the appellant the constitutional justice which (s)he had afforded the appellant at first instance. That presumption is rebuttable, but only by evidence, of which there is none here.

206. As to requirements of independence, impartiality and open-mindedness imposed on tribunal members, and though decided in a different constitutional context, Lewis458 provides an illustration. It related to alleged objective political bias in local councillors’ making planning decisions.459 It was observed that “Councillors will inevitably be bound to have views on and may well have expressed them about issues of public interest locally. Such may, as here, have been raised as election issues. It would be quite impossible for decisions to be made by the elected members whom the law requires to make them if their observations could disqualify them because it might appear that they had formed a view in advance. …… The reality is that councillors must be trusted to abide by the rules which the law lays down, namely that, whatever their views, they must approach their decision-making with an open mind in the sense that they must have regard to all material considerations and be prepared to change their views if persuaded that they should …” It should be said that the English court regarded the impugned planning decision as not being quasi-judicial and suggested a stronger application of the rules to quasi-judicial decisions. It should also be said that the

decision might not have survived the scrutiny of Geoghegan J in Ardagh that it is difficult for politicians to cease, even temporarily to be politicians. We need not pursue the issue as politicians are excluded from ALAB. But there is at least an appreciable analogy between councillors being elected for their views, inter alia on planning, and experts and those representing interests being appointed to ALAB and that they are to be trusted to do their job properly.

207. Hogan et al460 cite Greenstar461 in which a disappointed waste collector asserted bias inherent in structures operated under Dublin City Council’s Waste Management Plan. It provided that a single operator would collect waste from Dublin households and, under the legislation and at the Council’s choice, would be either the Council itself or a private operator. Greenstar asserted that the position of the Council, as operator and competitor in, and regulator of, the market, “almost automatically” implied structural bias such that its decisions were objectively biased. McKechnie J disagreed: “… given the nature of the decision in question and the statutory functions of the local authority in relation thereto .. Those features alone could not give rise to objective bias” and “ultimately” the dual role of the respondents did not create bias. However, the situation increased the likelihood of objective bias such that McKechnie J considered precautions were “to be encouraged” and the “statutory requirement to consult with those affected” was emphasised. He warned that a decision-maker may not bulldoze through a decision in the face of trenchant, legitimate opposition, regardless of the relevant evidence before it. The case seems to me to imply a high

458 R (Lewis) v Redcar and Cleveland Borough Council [2009] 1 WLR 83 citing R (Island Farm Development Ltd) v Bridgend County Borough Council [2007]

LGR 60.

459 As the English planning system provides.

460 Hogan, Morgan, Daly; Administrative Law in Ireland 5th Ed. 2019 §14-41 et seq.

461 Greenstar Ltd v Dublin City Council [2009] IEHC 589, [2013] 3 I.R. 510.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

bar for demonstrating that a statutory scheme creates structural bias and such a bar seems to me consistent with the presumption of constitutionality of both statutes and the acts of public officials.

208. While not necessarily dispositive of the question of structural bias, it is at least relevant that ALAB, unlike the City Council as to waste collection in Greenstar, has no interest in operating aquaculture businesses itself. While those who do are represented on ALAB, they comprise one only of the six classes of appointee.

209. Nurendale462 was linked and similar to Greenstar but, as Hogan et al observe, on the facts the

decision was quashed, not for structural bias, but because the Council, in the person of the Assistant City Manager (and even though he was not a decisionmaker) had made egregious remarks that he would do everything in his power to stop the applicant collecting household waste in the Dublin region. McKechnie J inevitably described these as showing “a rigidity of mind, so that from the start there could have been no other outcome”.

210. Nor does the present case appear to me to approach the situation in Damache463 – cited in this context by Hogan et al – in which the Supreme Court struck down as unconstitutional s.29(1) of the Offences against the State Act 1939, which authorised a Garda superintendent who was part of the investigating team to grant a search warrant. The superintendent was not an “independent person”. At issue was the inviolability of the dwelling. Bias was not explicitly at issue in the case, nor was the impugned action external to the impugned decision. But that the crucial issue was the independence of the decisionmaker explains the citation of the case by Hogan et al in the context of bias.

211. SWI in written submissions cite Reid464 as citing Hogan et al465 citing in turn a Canadian case, Québec

Inc. v Québec,466 on structural bias. In fact Reid contains no such citation. But it is clear that the reference is

in fact to DPP v Behan467 in which Reid is cited just before the court turns to the appellants’ invocation of “structural bias” in the issue of search warrants and cites the Canadian case. SWI cite Québec Inc, in turn citing Lippé,468 for propositions that “there may also exist a reasonable apprehension of bias on an institutional or structural level” and “if the system is structured in such a way as to create a reasonable apprehension of bias on an institutional level, the requirement of impartiality is not met.”. However, the Supreme Court in Behan, and Hogan et al, cite a different excerpt from Québec Inc – to the effect that

462 Nurendale Ltd t/a Panda Waste Services v Dublin City Council [2013] 3 IR 417.

463 Damache v DPP [2012] 2 I.R. 266.

464 Reid v IDA [2015] 4 I.R. 494 at 528.

465 Hogan, Morgan, Daly; Administrative Law in Ireland 5th Ed. 2019 §14-45.

466 Québec Inc. v Quebec (Régie des permis d’alcool) [1996] 3 SCR 919.

467 [2022] IESC 23.

468 R. v Lippé, 1990 CanLII 18 (SCC), [1991] 2 S.C.R. 114

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

demonstration of structural bias,469 “presupposes that a well-informed person, viewing the matter realistically and practically – and having thought the matter through – would have a reasonable apprehension of bias in a substantial number of cases.470 In this regard, all factors must be considered, but the guarantees provided for in the legislation to counter the prejudicial effects of certain institutional characteristics must be given special attention.” Hogan et al explain this passage as reflecting that the court considering structural bias must be aware of the particular dimension that a number of cases are potentially affected; so the focus should not be exclusively on the person affected in the situation before the court.

Hogan et al suggest that this is an “articulate version of the direction in which Irish law is travelling, through

its emphasis on the necessity doctrine, where structural bias is in issue.” That may be the direction of travel but the Canadian case was cited by the appellant invoking the doctrine of structural bias. O’Malley J in Behan did not approve or disapprove of the Canadian case. She invoked necessity as permitting circumstances which “might well not satisfy an observer who fears that the decision may be influenced by institutional

bias”. But it was the quite particular necessity of urgency in a specific instance requiring the issue of a search

warrant by a senior Garda officer instead or by a court. And even then the majority disapproved of the specific course taken by the Garda in seeking the warrant from a senior officer who was not, they found, independent of the investigation in question as required by statute. For my part, I am happy to accept the requirement that the “well-informed person” views the matter “realistically and practically” and thinks the matter through. I would leave to another case and argument a view on the proposition that the reasonable apprehension must be of bias “in a substantial number of cases” if by that is meant that structural bias would not be found in a case in which there was a reasonable apprehension of bias in that case.

212. All that said, the statutory scheme at issue in this case provides considerable safeguards to ensure the representation of a wide range of knowledgeable interests – on broadly both sides and neither of the pro-aquaculture and environmental arguments. By that observation I do not suggest acceptance of the proaquaculture/pro-environment dichotomy which the Applicants presume. In my view, the precise decisions as to such representation – as to the statutory designation of classes of interest to be represented, as to the identification of nominating organisations to represent such classes and as to the ultimate appointment of members to ALAB from amongst nominees of prescribed organisations representing those classes such classes are matters of political and administrative judgment in which the court should interfere only in very clear cases of unfairness or bias. Any other view would offend the Constitutional principle of separation of powers.

213. Nor, at least in the absence of evidence,471 should the court embark on a detailed examination of the ostensible “balance of power” on ALAB deriving from such appointments. I say “ostensible” as the court is not equipped to make assumptions in that regard. For example, it is by no means obvious that an appointee from the “community development” class will tend to one view or another. He or she may tend to favour the commercial and employment benefits of aquaculture or tend to favour like priorities relating to angling

469 What the Canadian Courts call “institutional” bias.

470 Emphasis in original judgment in Québec Inc. v Quebec.

471 And I am not to be taken as suggesting that the presence of evidence would necessarily alter the position – that is for another case.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

tourism, tourism generally and/or may prioritise his or her view of what contributes to the quality of life of local residents.472 Indeed (s)he may tend to take different views depending on the location at issue.

214. Nor should I make, in the absence of any evidence, and whether for the purpose of considering allegations of unfairness, prejudgment or structural bias, potentially facile assumptions as to the likelihood that ALAB members of particular classes473 will betray their undoubted duties of independence, impartiality and fairness in favour of more or less doctrinaire advancement of their respective partisan interests. Until the contrary is proved, appointees must be considered, as to their character, judgment, impartiality, worthy of the public trust vested in them by their appointment. It is, of course, possible that evidence requiring a different view will be adduced in another case, but not so here.

215. Though the argument in Kemper was not framed in structural bias, the facts are instructive on the argument made as to the backgrounds of ALAB members. In Kemper, one member of An Bord Pleanála had, before his appointment to the Board, been the head of planning for a developer which had an interest in the solution to the deficiency of wastewater treatment capacity in the Dublin area to be addressed by the development at issue. In rejecting the allegation, Allen J agreed with the Board that the logic of a finding of bias on that account “would disqualify anyone and everyone who might previously have worked for a developer, whether as an employee or as a construction professional”. Another Board member had worked in the Department of Housing, Planning and Local Government which was responsible for the National Planning Framework from which, ultimately, the proposal for the wastewater treatment plant arose. Allen J, observing that decisions as to bias are “necessarily highly fact specific”, commented:- “In the mix are the duration of the previous professional relationship, the time that has elapsed since it ended, the nature and extent of the previous relationship, the nature and extent of the previous work, and, most of all, I think, the connection between the previous work and the assignment under consideration.” He found no “rational or cogent link” between that board member’s previous “limited involvement as a civil servant” in the Department and the decision made by the Board. It is important to state that framing the allegation as one of “structural bias” is not a basis for evading the requirement of fact-specificity. As stated, SWI did not adduce any evidence that the Minister had in fact operated the statutory architecture to pack ALAB with supporters of aquaculture or in support of its submissions that appointees of particular classes or nominees of particular organisations were likely to favour aquaculture over other, in particular environmental, interests. SWI tendered no evidence that particular associations of particular members of ALAB were of concern. The Sweetman Applicants tendered some such evidence, as to ALAB members John Evans and Micheál Ó Cinnéide but, as I will explain below these are not examples of bias.

472 For example, taking a view that salmon farms degrade beautiful vistas.

473 In the sense of class of appointee as described above.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

216. I should add that I did not find the Sweetman Applicants’ citation of McLoughlin474 much assisted. It recognises the category of structural bias but the facts were very different to the present facts.

Bias – Depletion of the Board

217. The Sweetman Applicants also complain that the balance on ALAB of supposedly competing interests, supposedly intended by the 1997 Act, was upset in the present case at crucial points in the process by combinations of intervals between resignations and appointments of members of particular classes of appointee and by recusals of ALAB members for conflict of interest. Though, in a sense, not a structural bias point, as it is premised on the Sweetman Applicants’ view of the structure mandated by the Act and on their premise of a requirement of a structural balance of sectional interests, it is convenient to address it here. The Sweetman Applicants say that at such junctures there were only four ALAB members engaged in the Appeal. This assertion was not analysed in detail as to the crucial points in the process concerned, their duration or as to the interests thereby left represented and unrepresented in ALAB’s consideration of the appeal or as to any decisions specifically made during the periods in question. However, in general terms and assuming the premises just stated, I accept that any such effect would have been exacerbated by the patterns of retirement and replacement of ALAB members over the time it took to determine the Appeal and by the recusals from consideration of the Appeal of Mr Brice and Mr O’Donnell and of Prof McIntyre after he had reported to ALAB on the oral hearing.

218. In my view, this submission is erroneous first, and for reasons I have set out, in its premise which views ALAB members as partisans rather than as independent and impartial. That apart, while there is a logic to the complaint, it lacks practicality. Resignations and intervals between resignations and appointments are inevitable. So too are recusals, especially in a field in which the “small pool of available people with the necessary expertise” will often have acquired that expertise on one “side of the fence” or the other. Upholding the Sweetman Applicants’ complaint in this regard would come at the expense of sclerosis in

decision-making if, every time a resignation or other gap on ALAB occurred or a member missed a meeting, consideration of appeals had to be suspended or the question be interrogated and fine judgments made (perhaps having adjourned to get legal advice, by the time of receipt of which a different combination of members would be in attendance) on whether the gap was such as to unbalance ALAB to such a degree as to require a suspension. In this regard, the doctrine of necessity does seem to me to apply. Appeals must be decided and only ALAB can decide them. Even if one accepted, which I do not, the Sweetman Applicants’ requirement, in effect, of a “balance of prejudice” on ALAB in dealing with each appeal, the observation of

474 McLoughlin v The Labour Court [2022] IEHC 283. The issue related to the respective statutory functions of the Workplace Relations Commission, the

Labour Court and the and The Minister for Enterprise Trade and Employment. Mr McLoughlin, a labour inspector and civil servant employed by the

Minister and attached to the WRC, complained that the Labour Court was in a position of objective bias in considering his complaint against his employer

the Minister and the WRC. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Supreme Court in Radio Limerick One475 would apply to the effect that “there are bound to be instances in which an administrative tribunal charged with quasi-judicial duties may lack the appearance of strict impartiality expected from a court administering justice.” Also applicable in rejecting the Sweetman Applicants’ argument is the observation of O’Donnell CJ in Kelly476 as to a bias issue that, while “if an outcome is invalid, the court must say so”, judicial review of the validity of decisions “must be approached with a realistic understanding of the imperfections of any process”. So too, is the requirement set in the Québec case that the issue be considered realistically and practically.

The Quorum & Voting

219. However the simpler answer to this point and, more generally, to the “balance of prejudice” premise of the structural bias allegation that ALAB members are assumed to advance the interests of their nominating bodies, is to note that s.29 of the 1997 Act explicitly contemplates, without distinction as between classes of appointee to ALAB, that “The quorum for a meeting of the Board shall be four.” S.28 provides that ALAB make its decisions “by a majority of votes of the members present”. So, the scheme of the Act as to ALAB’s decision-making, is clearly not based on representation of particular interests when making individual decisions. Rather, it relies on ALAB members’ duties of independence and impartiality. Ss.

28 and 29 are consistent only with an expectation that ALAB members will, indeed, leave their baggage at the door as, otherwise, those sections would indeed have the distorting effect which the Sweetman Applicants allege.

Structural Bias – Article 9A of the EIA Directive 2014

SWI also pleads that as ALAB’s Impugned Decision was in part made as a competent authority for the purposes of EIA, which it purported to perform, its decision must be quashed on the basis that s.23 of the 1997 Act as to the composition of ALAB and appointment to ALAB is “in breach” of Art. 9A of the EIA Directive 2014.477 In substance this is a non-transposition argument. Art. 9A requires that competent authorities perform their duties “in an objective manner and do not find themselves in a situation giving rise to a conflict of interest.” While Art. 9A is undoubtedly valuable, I cannot conceive it as other than declaratory of pre-existing law. Leaving aside the question whether the 2014 Directive applies to the Appeal, it has not

475 Radio Limerick One Ltd. v Independent Radio and Television Commission, [1997] 2 IR 291. The IRTC had terminated the applicants broadcasting

contract for breach of statutory obligations on grounds, the most serious of which were: the “blacking” of a news item, outside broadcasts which the

respondent believed were a form of illicit advertising; the failure to provide tapes of programmes broadcast by the applicant, and, alleged refusal to co-

operate with a request to inspect. Bias was alleged on the basis that one IRTC member was the reporter whose report had been “blacked” by the applicant

and had issued unfair dismissal proceedings against the applicant. The allegation failed on the facts as that member had recused herself from the

impugned decision.

476 Kelly v The Minister for Agriculture, Fisheries and Food, et al [2021] 2 IR 624 §3. I have reversed the order in which O’Donnell J cited these two

principles, but I think legitimately.

477 Directive 2011/92/EU of 13 December 2011 on the assessment of the effects of certain public and private projects on the environment as amended

by Directive 2014/52/EU of 16 April 2014. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

been shown that the obligations of Art. 9A exceed those imposed by the domestic law of bias and conflict of interest as they apply to ALAB’s decision-making in appeals generally, including in the performance of is functions as a competent authority in EIA. Accordingly, as the Applicants’ reliance on that domestic law fails, so too does its reliance on Article 9A.

220. SWI’s plea that there is no provision in the Aquaculture (Licence Application) Regulations 1998478 requiring the ALAB to perform its duties objectively and not to find themselves in a position of conflict of interest is misconceived. First because such duties are imposed by the law generally. And second, given the Art. 7 of the Aquaculture Appeals (Environmental Impact Assessment) Regulations 2012479 as follows:

“7. (1) Where a person has a duty under these regulations, he or she shall perform that duty in an objective manner so as not to find themselves in a situation giving rise to a conflict of interest.

(2) Where a person is party to, or has a material interest in, an appeal under these Regulations and that person may be involved in determining the appeal, then the Board shall make appropriate administrative arrangements to ensure there is a functional separation between that person’s duties and the relevant appeal.”

Structural Bias – Decision

221. For the reasons set out above, I reject the allegations of structural bias. With that rejection, the challenge to the constitutionality of s.23 of the 1997 Act also falls.

Bias – ALAB’s relationship with the Marine Institute, the Aquaculture Appeals Office and the DAFM

222. This allegation is not quite of structural bias but is analogous to it. The Sweetman Applicants assert objective bias in that all ALAB’s staff are seconded to it by DAFM.480 ALAB has no fixed assets, and shares office accommodation in Portlaoise and administrative support with the DAFM Agriculture Appeals Office. They assert that ALAB’s submission of February 2017 to an independent group established by the Minister to review aquaculture licensing sets out ALAB’s own concerns as to potential conflicts of interest in its relationship with the Marine Institute (which advises the Minister on aquaculture licence applications) and with DAFM. The Sweetman Applicants plead that “the potential for conflicts of interest” identified by ALAB in its submission “in terms of its day-to-day interactions with the Department” …. “is analogous to the hypothetical example of say the Dublin City Council planning office sharing advisors, administrative staff, office accommodation and a canteen with An Bord Pleanála.”

478 S.I. No. 236 of 1998.

479 as inserted by the Aquaculture Appeals (Environmental Impact Assessment) (Amendment) Regulations 2019.

480 Department of Agriculture Food and Marine.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

223. The State pleads and avers that The Agriculture Appeals Office has no role or function as to ALAB. The two are completely separate and do not functionally interact with one another. They share a building and some administrative support services but each has its own office, computers and filing systems, to which the other has no access.

224. ALAB’s submission of February 2017 recorded that it shared a part-time secretary (a seconded civil servant of the Minister481) with the DAFM Agriculture Appeals Office, both offices were in the same building) and its only other staff resource is a clerical officer. Unsurprisingly, ALAB complained of inadequacy of resources. ALAB observed that “it may be considered legitimate to question whether the mere fact of sharing a building provides the appropriate perception of independence to appellants and applicants.” But it

emphasised that it is “acutely conscious of this and is very careful” and “absolutely satisfied that all necessary

steps are taken to ensure the confidentiality of the files, papers and processes of ALAB from those of DAFM, and takes stringent measures to endure the ALAS material is absolutely secure.” It also recorded its consciousness of its need to maintain its independence in the context of its funding and oversight by DAFM (for which the 1997 Act in effect provides via ss.36 to 38).

225. The allegation that ALAB’s submission of February 2017 recorded conflicts of interest in its relationship with the Marine Institute is theoretically correct but greatly overblown. In fact, in emphasising

that it needed a “permanent technical advisor available to it on an exclusive basis” (in the event, Dr O’Toole),

it illustrated the point by postulating a hypothetical and merely illustrative proposal that “ALAB could share a Technical Advisor with say, the Marine Institute (Ml), this would In ALAB's view give rise to a conflict of interest for the Technical Advisor as the Ml is a reference body for the Minister for initial aquaculture licence applications.” There is no suggestion such an eventuality occurred.

226. The Sweetman Applicants jump from this supposed potential conflict of interest in ALAB’s relationship with the Marine Institute to allegedly “numerous examples of the exact conflicts of interest identified by ALAB itself as being problematic in this case including the manner in which the Marine Institute

involved.” It refers “for example” to the referral of Dr Gitting’s February 2017 report to the Marine Institute

for comment. It differed from his views on bird impacts. This criticism has no force as an allegation of objective bias. It ignores the facts that the Marine Institute is not merely a statutory body482 with a general statutory function to “undertake, to co-ordinate, to promote and to assist in marine research and development and to provide such services related to marine research and development, that in the opinion of the Institute will promote economic development and create employment and protect the marine environment”483 and whose functions include advisory functions as to marine matters including aquaculture,

481 See above s.32 of the 1997 Act.

482 Established by the Marine Institute Act, 1991.

483 Marine Institute Act, 1991.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

it is a prescribed consultee in aquaculture licensing applications484 and, for that matter, a body of whose policies and objectives ALAB is obliged to keep itself informed.485

227. The Sweetman Applicants plead that at the ALAB meeting of 15 May 2020, it was recorded that ALAB members John Evans and Micheál Ó Cinnéide “agreed to work together in reviewing the draft Technical Advisor's report with a view to identifying any issues that may need to be clarified”.486 The Sweetman Applicants plead that “No issue was raised then about the potential biases that could ensue from two former Marine Institute directors reviewing a report prepared by a former Marine Institute scientist when it was clear even then that the adopted position of the Marine Institute in relation to risks to wild salmon from sea lice is opposite to the position of Inland Fisheries Ireland and much of the international scientific community.”

Of this submission, I observe as follows:

• Holding a position, assuming the Marine Institute does, at odds with that of the Sweetman Applicants is not evidence of bias any more than is IFI’s opposition to salmon farming evidence of bias. That “much of the international scientific community” holds a view contrary to that of the Marine Institute, if true, is not evidence of bias. Nor does it exclude a view that “much of the international scientific community”, even if a different cohort, may broadly share the position of the Marine Institute while recognising risks.487

• ALAB’s canvassing the view of the Marine Institute, a statutory consultee, is not evidence of bias.

• The description of Dr Saunders is incorrect, or at least incomplete. He was never a Marine Institute employee. His CV records that he is a self-employed “marine ecologist and independent environmental consultant” whose clients included the Marine Institute. For that client and operating as an Associate to RPS and reporting to the Marine Institute, he had been Programme Manager of its Marine Strategy Framework Directive488 Project. That directive aims at good environmental status of marine resources and ecosystems.

• As to Mr Evans and Mr Ó Cinnéide, the Sweetman Applicants’ submission seeking to quash a presumptively valid statutory decision that “At the very least there was scope for a type of group think that could bias the decision” seems forlorn.

• John Evans was employed by the Marine Institute between 2012 and 2014 and in 2017. That he brought this to ALAB’s attention is minuted under the heading “Conflicts of Interest/Section 31 Declaration” on 19 March 2020. Micheál Ó Cinnéide was employed by the Marine Institute period which ended in 2008. That

484 Aquaculture (Licence Application) Regulations, 1998 Art 34.

485 1997 Act s.39.

486 Dr Saunders was still the Technical Advisor at that point.

487 See e.g. the exhibited 2016 Report of the Workshop to address the NASCO request for advice on possible effects of salmonid aquaculture on wild

Atlantic salmon populations in the North Atlantic (WKCULEF) which lists the Marine Institute as a participant.

488 Directive 2008/56/EC establishing a framework for community action in the field of marine environmental policy (Marine Strategy Framework

Directive). SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

he brought this to ALAB’s attention is likewise minuted on 19 March 2020. Also minuted on 19 March 2020 is ALAB’s consideration of this information and conclusion that neither Mr Evans nor Mr Ó Cinnéide was conflicted.

228. In my view, the reasonable observer would be reassured by these minutes that all three of Mr Evans, Mr Ó Cinnéide and ALAB generally were alive to their duties as to matters such as bias and had addressed them. There is no suggestion that either had had, when working with the Marine Institute, anything to do with the subject applications. And if mere past association with one of the parties in a decision-making

process is insufficient to imply a ‘cogent and rational link’ and objective bias (Kemper489) a fortiori that must

be so where the alleged link is not with one of the parties but with a statutory consultee in the process. As far as it goes, I see this allegation as falling well short of engendering the reasonable and realistic apprehension of bias required by the cases. Nor do I see it as contributing, in any wider context of other relevant facts, to such an apprehension.

229. The Sweetman Applicants’ complaint on this issue continues by placing this supposed basis of bias, consisting in the contemplated liaison between Mr Evans, Mr Ó Cinnéide and Dr Saunders, in the context of the Sweetman Applicants’ assertion that “it was clear even then that the adopted position of the Marine Institute in relation to risks to wild salmon from sea lice is opposite to the position of Inland Fisheries Ireland and much of the international scientific community. At the very least there was scope for a type of group think that could bias the decision.” Of course, the Sweetman Applicants and others are entitled to their, no doubt reputable and in some cases expert, views as “to risks to wild salmon from sea lice”. And one might expect ALAB to pay particular attention to the views of IFI. But so too is the Marine Institute entitled to its views. As I have said, it is an expert body established by Act of the Oireachtas and a prescribed consultee in aquaculture licensing applications. Indeed the last sentence of this part of the submission, as to “scope for a type of group think”, demonstrates its paucity. Without accepting that there was such a scope, I observe that

decision-makers are expected to recognise and overcome any dangers of group think. It is expected that they are appointed for their independence as well as their competence expertise and their appreciation of their duties of fairness and to avoid bias and of the expectations of them reflected in the East Donegal

decision. In this case, as it happens, ALAB, Mr Evans and Mr Ó Cinnéide are minuted as having demonstrated their appreciation of their duties.

230. The risk of group think is endemic in any collegiate decision-making body and its members are expected to combat it. It is not, per se and without very considerably more, a basis for a finding of objective bias. Indeed, as it is internal to the decision-making process, it is not properly a basis for a finding of objective bias at all.

489 Kemper v An Bord Pleanála [2020] IEHC 601 §53.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

231. As specifically concerns Mr Evans and Mr Ó Cinnéide, I would respectfully recall that as O’Donnell J said in Shatter490 an assertion of even objective bias “is an ugly allegation and the label can cling once

applied.” It “should not be lightly made” against “a person performing a public function”. Charleton J in the

same case described it as a “really serious” allegation which “should be thought through before it is made”.

As the matter was not argued in those terms, I will not make any finding in those terms, but will confine myself to rejecting the allegation unequivocally. Mr Evans and Mr Ó Cinnéide are entitled to that. I must add, of course, that the tenor of these remarks is not intended to dilute the force of my rejection of allegations of objective bias as they concern others against whom they are made.

Bias – Specific Allegations

232. I now turn to specific allegations of objective bias, I will consider them in roughly chronological order.

Bias – RPS & ALAB meeting 24 November 2015 – Decision

233. IFI cites,491 as exhibiting objective bias in ALAB, the minute of its meeting of Tuesday 24 November 2015. The minute includes the following:

“As had been agreed at the last Board meeting, the expert panel of Technical Advisors used by ALAB were requested to examine the grounds of appeal and give an initial outline for the Board as to how they would approach the evaluation of the appeal grounds. The Board had requested that the core issues of the appeal be identified and consolidated in a single report. The technical Advisors were required to submit a proposal and submission, a brief scoping study and were invited to make a presentation to the Board.

The Board noted that when contacted, Xxxxxx492 disclosed they were conflicted and therefore were not in a position to evaluate.

Reports were provided by both Yyyyyy493 and RPS. The reports presented by Yyyyyy and RPS summarised the issues each had identified and the grounds of the appeals, and suggested a proposed methodology for progressing the appeal. They identified the core issues for the Board and where information would be required.”

490 Shatter v Guerin [2019] IESC 9; [2021] 2 IR 415, §30.

491 IFI Written Submissions §4.59.

492 Name of technical advisor redacted by me.

493 Name of technical advisor redacted by me.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

234. The minute is notable first in that it reflects ALAB’s contemporaneous proper (and, it should be said typical494) awareness and concern as to conflict of interest issues – both internal to ALAB and as recognised by its Technical Advisor Xxxxxx.495 Before coming to the second notable element of the minute, some further context is required.

235. The exhibited extract of the ALAB minute of its meeting of 20 October 2015 records its decision to bespeak from each of its panel of technical advisors a proposal and submission, a brief scoping study and a presentation to the Board as to the Appeal. Notably, they were to be sent only the Appeal documents themselves. While the minute records that each panel member would be paid a modest fee496 for its submission, it is clear that this was essentially a scoping and procurement process with a view to identifying a technical advisor to the Board. It did not bespeak substantive advice on the merits of the Appeal.

236. RPS is not mentioned in the minute of 20 October 2015 but ALAB’s members must be presumed to have known the identity of its panel members and to have envisaged that a proposal would be sought from RPS. ALAB had not yet received from MOWI the RPS Water Quality Modelling Report 2015 which RPS prepared for MOWI and which loomed so large in the Appeal. MOWI submitted the RPS Water Quality Modelling Report 2015 to ALAB under cover of letter dated 20 November 2015. That was a Friday. We do not know if it was posted that day or only later. The copy exhibited is not date-stamped received by ALAB. In my view, it is not apparent that by the following Tuesday ALAB had the RPS Water Quality Modelling Report 2015 such that ALAB should have noted that RPS would have a conflict of interest in acting as its technical advisor.

237. However, the 2011 EIS records that RPS had a contributed a wave climate analysis and a modelling study of the deposition and dispersal of cage discharges. That said, the Appeals had come in only in early to mid-October and the meeting was held 4 days after the appeal deadline and was clearly a very initial consideration of the Appeals. It is not entirely clear, but the minute of that meeting discloses no possession of any documents other than the notices of appeal and envisages a request under s.43 of the 1997 Act to the Minister for documents, including the EIS. So, at this point, ALAB as a probability, did not have the EIS either.

238. As RPS is not before the court and as the entire circumstances are not necessarily known to me, it would be inappropriate of me to say more of RPS in this context than that, RPS having been retained by MOWI, it is at least puzzling that it decided to submit such a presentation to ALAB.

494 In that similar considerations are evident in the minutes of its other meetings.

495 Name of technical advisor redacted by me.

496 €2,000 + VAT.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

239. What seems to me important for present purposes however is that:

• RPS had been sent only the Appeal documents – the content of which was already known to MOWI.

• The RPS presentation to ALAB was in the nature of a scoping and procurement presentation with a view to its retainer as technical advisor to ALAB and was not likely to have included substantive advice on the merits of the Appeal.

• It is not apparent that by its meeting of 24 November 2015 ALAB had the RPS Water Quality Modelling Report 2015 or the 2011 EIS such that ALAB should have noted that RPS would have a conflict of interest in acting as its technical advisor.

• Even if, somewhere in the documents ALAB had at that point, RPS were mentioned, a reasonable observer would infer that at this initial meeting 4 days after the appeals deadline, ALAB simply didn’t advert to any RPS involvement.

• RPS was not in fact retained as technical advisor to ALAB and its direct contact with ALAB as to the Appeal ceased with that scoping and procurement presentation.

• The subsequent, lengthy, technical and detailed course of the Appeal was such that it is highly unlikely, in any realistic and practical view, that any content of the RPS presentation had any effect on the processing, consideration or determination of the appeal.

Accordingly, I find that the circumstances disclosed on this issue do not amount to objective bias in ALAB.

240. As a public body whose decision is impugned in judicial review, ALAB must, as a matter of candour, to put all its cards face up on the table – e.g. Sherwin, CHASE and O’Lone.497 No doubt any commercially sensitive content in the RPS presentation to ALAB could have been redacted. In this context and had IFI pleaded this issue, its complaint at trial that ALAB had not exhibited it or addressed the issues it raised, would have been particularly understandable. However, for the reasons given, the complaint of objective bias in this respect cannot be upheld.

241. All that said, ALAB objects to IFI’s submission in this regard as not pleaded. IFI relies on its Core Ground 5 which alleges that ALAB’s conduct of the appeal evinced objective bias by the large number of time extensions granted, ALAB’s directing the production of further documents and bespeaking additional reports, thereby collecting the material to justify its pre-determined decision to grant the Aquaculture Licence. The particulars of this allegation do not add substance to it. There is no reference to the events described above and they do not, even generally, fit the description in the Core Ground.

497 E.g. Sherwin v An Bord Pleanála [2023] IEHC 26 (High Court (Judicial Review), Humphreys J, 27 January 2023), Cork Harbour Alliance for a Safe

Environment v An Bord Pleanála & ors, [2019] IEHC 85, O’Lone v An Bord Pleanála [2023] IEHC 136 (High Court (Judicial Review), Humphreys J, 21 March

2023) all citing R. v Lancashire County Council ex parte Huddleston [1986] 2 All E.R. 941. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

242. As it was necessary to describe the allegation to decide the pleading point, I have stated the view I would have taken had it been pleaded. But the fact remains that the allegation was not pleaded and is rejected on that account.

Bias – Dr Saunders’ other Role & Declaration of Interests

Introduction

243. The Sweetman Applicants plead and aver, as an example of objective bias, that ALAB’s Technical Advisor, Dr Graham Saunders, joined that role directly from his previous role in RPS and the Marine Institute. He was appointed Technical Advisor to ALAB on 7 April 2016. In July 2016, Dr Saunders finished his role as Independent Associate to the RPS Group reporting to the Department of Environment and Local Government and the Marine Institute while managing the team undertaking Ireland’s delivery of the Marine Strategy Framework Directive498 (“MFSD”). The assertions by the Sweetman Applicants are that:

• Dr Saunders’ status as “an Associate to RPS” for about 5 years prior to and to July 2016 – a few months after his appointment as ALAB’s Technical Advisor – in circumstances in which RPS had since 2015 acted as consultant to MOWI as to their Aquaculture and Foreshore Licenses Applications, was such as to justify an inference of objective bias on his part, which, in turn, infected ALAB’s decision.

• Alternatively, the foregoing circumstances, together with Dr Saunders’ failure to disclose those circumstances in his Declaration of Interests to ALAB, were such as to justify an inference of objective bias on his part, which, in turn, infected ALAB’s decision.

244. S.33 of the 1997 Act required Dr Saunders, on his retainer as Technical Advisor to ALAB, to declare specified interests.499 As a matter of fact, and as is apparent on the CV which accompanied his February 2016 tender to ALAB to provide services as their Technical Advisor for the Appeal, Dr Saunders had been retained by the State (via the Marine Institute – which he described as “Key Clients”) from 2011 as Programme Manager of Ireland’s response to the MFSD. In his CV he describes himself as “Operating as an Associate to RPS and reporting to the Marine Institute and the client Steering Group, involving considerable collaboration

498 Directive 2008/56/EE of the European Parliament and of the Council of 17 June 2008 establishing a framework for community action in the field of

marine environmental policy (Marine Strategy Framework Directive).

499 S.33(3) This section applies to the following interests:

(a) any estate or interest which a person to whom this section applies has in any land, or in any process, development or operation, associated

with aquaculture or fish processing or the manufacture, sale or distribution of products used in aquaculture or fish processing;

(b) any business or dealing in or developing land, or any process, development or operation associated with aquaculture or fish processing or the

manufacture, sale or distribution of products used in aquaculture or fish processing, in which such a person is engaged or employed and any such business

carried on by a company or other body of which he or she, or any nominee of his or hers, is a member;

(c) any profession, business or occupation in which such a person is engaged, whether on his or her own behalf or otherwise, and which relates

to dealing in or developing land or to any process, development or operation associated with aquaculture or fish processing or the manufacture, sale or

distribution of products used in aquaculture or fish processing. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

with State Agencies and academic institutions.” His CV records his authorship of 3 documents in that capacity.500

245. The CV also describes Dr Saunders’ expertise as a “Marine Ecology and Independent Environmental Consultancy” with over thirty years' experience “supporting government and industry in the UK, Ireland and worldwide”. His stated experience includes “2009 – Present – Independent Environmental Consultant”. His exhibited Linked-In profile describes him in this capacity as “Operating as an independent Associate to RPS Group but reporting to the Department of Environment and Local Government and the Marine Institute”. More generally, it clearly indicates that he is a “consultant” and an independent contractor trading as “Graham Saunders Marine Ecology” with contracts in various countries. Incidentally, it also records that his BSc was in Applied Marine Biology and his PhD is in Marine Ecology & Toxicology.

246. Dr Saunders’ CV was, erroneously, exhibited by ALAB in June 2022501 as his declaration of interests. Clearly it was not that. S.33 of the 1997 Act requires declarations in the “prescribed form”, which the CV was not. This error was not noticed until during trial, at which point ALAB exhibited Dr Saunders’ actual declaration of interests, dated 13 April 2016. It recorded his appointment as ALAB’s Technical Advisor on 7 April 2016. It disclosed no relevant interests. In July 2016, Dr Saunders exited his role as MFSD Project Programme Manager after about 5 years involvement. The Sweetman Applicants complain of the overlap of roles, which lasted a few months in 2016.

247. There is no evidence or argument that Dr Saunders’ was ever an RPS employee or that his role as MFSD Project Programme Manager, or as “an Independent Associate to RPS” in that regard, involved him in any way in MOWI’s Aquaculture and Foreshore Licenses Applications which had been in train since 2011 or in the RPS studies and reports for MOWI. Indeed, that is MOWI’s evidence and it was not disputed. It says that “Mr Saunders was at no time directly employed by RPS nor did Mr Saunders collaborate with RPS on

(the) water quality studies.”502

Issues

248. The following potential issues appear to arise from the foregoing:

500 Department of Environment Community and Local Government (2015). Marine Strategy Framework Directive. Article 11 Monitoring Programmes

Report. (G. Saunders contributor and primary editor); Haynes, T., Bell, J., Saunders, G., Irving, R., Williams J., and Bell, G., (2014), Marine Strategy

Framework Directive Shallow Sublittoral Rock Indicators for Fragile Sponges and Anthozoan Assemblages, JNCC Report 524, ISSN 0963 809.; Department

of Environment Community and Local Government (2013). Ireland’s Marine Strategy Framework Directive Article 19 Report. Initial Assessment, GES and

Targets and Indicators. (G. Saunders contributor and primary editor).

501 Exhibit ‘FD1’ Tab 1 to the Affidavit of Francis Dowling, sworn 28 June 2022.

502 Affidavit of Catherine McManus 6 July 2022, §35.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

i. Does Dr Saunders’ association with RPS indicate objective bias?

ii. Did Dr Saunders breach his duty as to declaration of interests?

iii. Does either or both together of the foregoing indicate objective bias in Dr Saunders?

iv. Does objective bias of an advisor to a decision-maker infect the decision-maker’s decision?

Consideration & Decision

249. In my view, the facts disclosed on Dr Saunders’ CV – his acting as “an Associate to RPS” and as MFSD Project Programme Manager – do not of themselves imply objective bias on the part of Dr Saunders. There is no suggestion that the role bore on anything relevant to MOWI’s licensing applications. The assertion of bias is based entirely on the mere fact of association between RPS and Dr Saunders.

250. The relevant question is whether the reasonable observer would see a “cogent and rational link between the association and its capacity to influence the decision to be made in the particular case …” – CHASE.503 As CHASE records, in the case of judges, that a judge when previously a barrister worked for or

against a litigant will not indicate objective bias. It is true that factors such as the judicial oath and the “cab

rank” principle applicable to barristers are not present here. But I think the reasonable observer would understand that consultant professionals, of the nature of their work, act for one client today and another tomorrow and that, in doing so, project-specific teams including such external associates are formed and come to a natural end in time. Commercial contacts and associations of various kinds naturally arise, wax and wane with circumstance. None of these considerations blind the reasonable observer to the risk of bias but (s)he would look at the matter broadly and realistically.

251. All cases are fact-specific and analogies are to be gingerly drawn. But I find Kemper helpful. In that case, the Board member who led the appeal process had, prior to his appointment to the Board, been the head of planning for a developer which had an interest in the solution to the lack of capacity to deal with wastewater in the Greater Dublin area which lack was to be addressed by the development at issue before the Board – the Greater Dublin Drainage Project. Allen J stressed the highly fact-specific enquiry required when considering an allegation of objective bias. He could not see how that member’s previous employment with a developer (which was not involved in the development at issue and whose land valuation would not be enhanced or diminished by the location of the development) could give rise to a reasonable apprehension by a reasonable observer of bias. Allen J. agreed with the Board that “the logic of such a proposition would disqualify anyone and everyone who might previously have worked for a developer, whether as an employee or as a construction professional”.

503 §104 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

252. Allen J noted that Ms Joyce Kemper had focused on the closeness in time of the previous professional associations of the two Board members concerned to their assignment as Board members to deal with the application at issue. In that regard, Allen J observed:- “Proximity in time may be a factor but it cannot be determinative. Each case is to be decided on its own facts. If there is no link between a previous engagement and the administrative decision, the proximity in time may count for little. If there is a substantial link, a long lapse of time may count for very little.”504 Allen J commented: “In the mix are the duration of the previous professional relationship, the time that has elapsed since it ended, the nature and extent of the previous relationship, the nature and extent of the previous work, and, most of all, I think, the connection between the previous work and the assignment under consideration.”505 In this case there is no evidence of any such connection.

253. Dr Saunders’ position is, if anything, less concerning than that which arose in Kemper. He is not a member of ALAB (though I accept that advisors’ bias can infect decision-making) and he was never an employee of RPS. However, I do not rest my view on those distinctions alone. In my view, the logic and analogy of Kemper assists in my concluding, as I do, that this issue of association with RPS does not justify a finding of objective bias as I do not see the required cogent and rational link between that association and the proposed Salmon Farm.

254. On the distinct issue of statutory disclosure, s.33 of the 1997 Act requires disclosure of specified interests footnoted above. All the interests in question are associated, in various ways, with aquaculture or fish processing. There was no evidence before me, and no allegation or submission was made to me, that Dr Saunders’ association with RPS related to aquaculture or fish processing. I was not told anything of the substantive content of the MFSD project. Those asserting bias bore the onus of proof in that regard. Even assuming the MFSD project bore in some way on aquaculture or fish processing, there is no evidence that any resultant involvement in that project was such that he acquired an interest in aquaculture or fish processing of any of the three kinds identified in s.33(3) of the 1997 Act.

255. In hindsight, Dr Saunders might consider that, of an abundance of caution, he might wisely have declared his involvement in the MFSD project as an independent associate of RPS. But that observation is no basis for finding him in breach, in his absence and by way of implication of a possible criminal offence,506 of his statutory obligation of disclosure. Were he involved in these proceedings, he might well have replied that he had drawn his association with RPS prominently to ALAB’s attention when he sent them his CV. Of course, that would not appear in the register of interests required by s.33(8) of the 1997 Act. But the net point is that he did not, in my view, breach his statutory obligation.

504 §95.

505 §92.

506 s.33(11) of the 1997 Act.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

256. I should note that there are some factors here which tend to a finding of objective bias:

• Dr Saunders’ prior association with RPS was lengthy and there is the timing overlap in that Dr Saunders’ appointment to ALAB was in April and his departure from the MFSD project was in July.

• Despite my view that he was not obliged in law to do so, it is perhaps regrettable that Dr Saunders did not, of an abundance of caution, disclose in his prior association with RPS in his declaration of interests.

• ALAB needlessly failed to exhibit his declaration of interests until during the trial.

257. However,

• while in hindsight that the CV was erroneously exhibited is obvious, it was not noticed by anyone – Applicants, Respondents or Notice Party – until at trial. I have no reason to believe the error was anything but genuine.

• while his declaration of interests in April 2016 did not disclose Dr Saunders’ prior association with RPS, the

CV which accompanied his February 2016 tender to ALAB had done so.

• it is clear that Dr Saunders’ association with RPS was by some form of contractual hiring in or retainer by RPS of Dr Saunders as an independent consultant for, presumably, his particular specialist expertise. There is no suggestion that it was full-time or an employment relationship or anything akin to it.

• on the most important issue identified by Allen J in Kemper, there is no suggestion of “connection between the previous work and the assignment under consideration”.

• here there is a further, and to my mind, significant consideration. Though bias of an advisor can infect the

decision-maker depending on circumstance, it does diminish the prospect of bias that Both Dr Saunders and RPS are consultants to their respective clients not principals. They are neither developers on the one side nor decision-makers on the other.

258. In the end, in my fact-specific view as to the issues of Dr Saunders’ association with RPS and his alleged failure to disclose it, the allegation of objective bias in Dr Saunders’ is not made out. It follows that no such bias in Dr Saunders can be attributed to ALAB or can have infected its processes.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Bias – ALAB/Saunders Interaction as to Saunders Report 2016/2017

Introduction & Principles

259. Dr Saunders was retained as Technical Advisor to ALAB specifically and only for the Appeal. He was not an ALAB employee – as is the case with Dr O’Toole. As I have said, it is necessary that bodies such as ALAB will, in retaining such advisors, adequately brief them early as to their task. The correspondence I outline below suggests inadequate initial briefing of Dr Saunders may have been an issue in the present case. Nor, it seems to me, is there anything inherently wrong with liaison, while the expert is considering the case with a view to reporting on it, between the expert and the statutory body retaining him/her. Indeed, it will often be very desirable in various respects. For example, the expert statutory body may have particular concerns it wishes investigated. However, the statutory body must not attempt to procure from the expert a report which supports a preconceived decision in the statutory body. An expert’s report must be the product of his/her independent and open mind. To that end in controversial matters, and if only to avoid unwarranted suspicions of ‘uisce faoi thalamh’, prudence if not the law, suggests that careful records of such interactions – including oral interactions – should be kept. Such practices ensure to public confidence in process.

260. The Sweetman Applicants cite Nurendale for a proposition that where the occupier of a senior position within the decision-making body, though not a member of it, makes representations or comments on the outcome of a decision yet to be taken, they may suffice to give a perception of prejudgment of the subsequent decision. As has been noted earlier, in Nurendale, the Court found bias external to the waste management plan variation process in the prior statement of the Assistant Dublin City Manager in charge of waste management services507 that he would do everything in his power to stop the applicant collecting household waste in the Dublin region. McKechnie J also said the following:

“[175] I would further note that in the course of the hearing a number of draft reports, prepared by Dr. Francis O'Toole and RPS, were handed up to the court. The drafts of the former contained comments written by the respondents indicating which parts of earlier drafts were acceptable to them, and either deleting or rewording those parts which would not have supported their position.

There were also email references to meetings with the authors of these reports as well as notes of some meetings ……… which would indicate that the findings of the reports were a foregone

conclusion. Whether or not the city managers were aware of this fact is, in my opinion, immaterial:

Mr. Twomey508 certainly was. Such massaging of reports, which were later, in their edited versions, released publicly, is a strong indicator, to me, of unacceptable influence in a process, supposedly carried out in the public interest, and further elucidates a high level of prejudgment in the decision to vary the WMP.

507 This is a brief description of his relevant responsibilities which are more fully described in the Nurendale judgment.

508 The Assistant Dublin City Manager in charge of waste management services.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

261. These principles are clear, important and must be adopted by the statutory body and its expert advisors. “Massaging of reports”, is unacceptable in a process carried out in the public interest. That said, “Massaging of reports” is not a term of art and whether it has occurred in a given case is a matter of fact and may be a matter of degree. In Nurendale the bias was, on the facts, egregious – both as to the background of the statement by the Assistant Dublin City Manager and, it seems, as to the “paper trail” of redrafting the report. And whatever considerations of prudence may suggest as to record-keeping, courts should not adopt a suspicious approach to interactions between a statutory body and its expert advisors – much less lightly conclude massaging of reports or legal error in their interactions. Courts must be careful not to stymie legitimate and productive interaction between advisor and decision-maker in an iterative process in which, as I have said, the eventual outcome may progressively come into view. Indeed, it is properly part of the job of advisors to recommend a course of action – a particular decision – to the decision-maker. And we should be reassured rather than alarmed if the statutory body interrogates and tests those recommendations in light of the possibility of not following them. Nor should the terms of such interactions be interpreted as if statutes or as if drafted by lawyers. Advisors and decision-maker must be afforded the legitimate room or margin required to facilitate a high standard of decision-making. In these respects, the presumption of validity applies and the onus of proof of error remains on the applicant for judicial review.

Specific Allegations to December 2016 & Analysis Thereof

262. IFI and SWI have criticised, in particular and as “management of expertise”, a series of interactions between Dr Saunders and the ALAB secretariat in the autumn of 2016509 which preceded Dr Saunders’ interim report of 31 December 2016. As stated, Dr Saunders was retained in mid-2016. From April 2016 he read into the papers and identified issues arising. He made a site visit in early August 2016. By e-mail of 22 August 2016 he sent his draft report to the ALAB secretariat. I have not seen it. His e-mail envisaged “a subsequent draft, subject to ALAB comment and approval”. His draft recommendation was that the licence should be “temporarily withheld” pending satisfactory resolution of issues incompletely addressed by MOWI.

Clearly this implied, as his overall view, that in the event of satisfactory resolution of those issues, the licence

might be granted. What follows must, in my view, be considered in that particular light.

263. By reply of 23 August 2016, the ALAB secretariat observed that, by the 1997 Act, the option of temporarily withholding a licence was not open. The secretariat cited s.40(4) of the 1997 Act510 and asked Dr Saunders to amend his draft report to “include the appropriate recommendation”. The secretariat identified

509 Exhibit MCC2 T53 to the Affidavit of Mona-Claire Costelloe sworn 15 February 2023 for IFI.

510 (4) Where an appeal is brought under this section and is not withdrawn, the Board shall, subject to subsection (5), determine the appeal by —

(a) confirming the decision or action of the Minister,

(b) determining the application for the licence as if the application had been made to the Board in the first instance, or

(c) in relation to the revocation or amendment of a licence, substituting its decision on the matter for that of the Minister.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

options511 but it was non-directional as to the substantive recommendation. That was unobjectionable as to bias.

264. Dr Saunders replied in turn on 23 August 2016 expressing, it seems to me, two categories of view. The first, and it seems to me the most important, was his overall and clear view as to what, in general terms, should happen. He said that “Overall, almost all of the evidence indicates that the licence should be granted.” This overall view was entirely consistent with the view expressed in his e-mail of 22 August 2016. I have seen no evidence that this overall view was procured or affected by any improper interaction with ALAB. I have no reason to infer that it was anything but the expression of his expert and independent view. His overall view was followed perfectly properly, by his observation that “There are, however, a few issues which haven't been clarified or addressed, or have just been incompletely addressed, but they could turn out to be nonissues with the provision of a small amount of additional information.”

265. The second category of view he expressed in his e-mail of 23 August 2016 related to the statutory/technical options open to him to recommend given his overall view. He said that his “instinct” in light of s.40(4) of the 1997 Act was to amend his draft report to read: “It is recommended that the site licence should be refused” ... on the basis of the identified remaining or unresolved issues. He asked “Is this correct? Or should it be: “It is recommended that the site licence should be revoked?”. Clearly, the latter inquiry derived from ALAB’s misconceived suggestion to him that s.40(4)(c) might apply. And the former contemplation of refusal can only have proceeded from an incomplete knowledge of the statutory options open to ALAB in terms of further information-gathering – as it clearly would not carry into effect his clearly expressed overall view of the previous day that “almost all of the evidence indicates that the licence should

be granted” if “a few issues” were satisfactorily clarified – perhaps by “the provision of a small amount of

additional information.” Unsurprisingly, he expressed unease as to his understanding of the options open to him under s.40(4) of the 1997 Act and anxiety to discuss the matter with ALAB. Dr Saunders seems to have been unaware that it was open to him to issue an interim, as opposed to a final, report and open to ALAB to defer a decision and seek such information. He may have been inadvertently misled to this impression by his having been told by the ALAB secretariat that the licence could not be “temporarily withheld” pending provision of such information. That was correct in that no formal order determining the Appeal could have been made in those terms. But it was incorrect in that it was open to the interpretation (and seems to have been so understood by Dr Saunders) that the Appeal had to be decided only on the information then to hand. In my view, it would have been understandable if a distinction had not been apparent to him between, on the one hand, deferral of a decision pending further information-gathering and investigation and, on the other hand determining the Appeal by temporarily withholding the licence.

511 This secretariat’s reply of 23 August 2016 did confuse the issue as one option it identified as open – s.40(4)(c) in relation to the revocation or

amendment of a licence, substituting its decision on the matter for that of the Minister – did not in fact arise, as revocation or amendment of a licence

was not in question. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

266. ALAB’s secretariat replied within minutes to Dr Saunders’ e-mail of 23 August 2016, indicating that it would contact him on 28 August 2016. It is not apparent that ALAB did so. It seems not as, by e-mail at 10:59 on 26 August 2016, Dr Saunders enclosed an updated draft report of which he said that he had “amended sections512 the recommendation to a refusal of the licence, although, as previously indicated, I am unclear as to whether this should really be a revocation. I have also re-ordered section 10 so that the arguments for and against the granting of the licence are in a more logical sequence.” This appears to reflect his incomplete understanding of the options open to ALAB under the 1997 Act – specifically of further informationgathering and investigation before determination of the Appeal. That he had not, before updating his draft report, awaited the liaison with ALAB contemplated in ALAB’s e-mail of 23 August 2016 is suggested both by his text I have just quoted and his conclusion “I am available now to discuss if you wish?”

267. ALAB’s secretariat replied to Dr Saunders by e-mail of 31 August 2016 – at this point drawing to his attention ss.46 and 47 of the 1997 Act as allowing ALAB to request “submissions or observations in relation to any matter which has arisen in relation to the appeal” and/or “any document, particulars or other information is or are necessary for the purpose of enabling it to determine an appeal”. The e-mail asked that Dr Saunders revisit his report to “see if you can identify what additional submissions, observations or documents the Board might request, which would enable you and the Board obtain the additional information needed, or clarify any inconsistencies.”

268. ALAB’s secretariat by that e-mail of 31 August 2016,

• also noted Dr Saunders’ recommendation of an oral hearing. o I observe that his recommendation of an oral hearing must be inconsistent with his recommendation of a refusal of the licence and seems supportive of a view that his recommendation of refusal was based on a mistaken understanding of the statutory scheme – to the effect that he had no option but to recommend refusal. It is also at least to some degree consistent with his initial view that the licence should be “temporarily withheld”.

• sensibly suggested that Dr Saunders could defer a view as to an oral hearing pending receipt of any additional information. o I observe that deferral of a view as to an oral hearing is also consistent with ALAB’s general practice – at least at that time. Its 2017 submission to the Independent Aquaculture Licensing Review Group records: “Because of ALAB's limited resources it is quite a challenge for ALAB to conduct an oral hearing and it puts an enormous strain on the ALAB resources and personnel.”

269. It seems clear to me that:

512 Sic.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• ALAB’s e-mail of 31 August 2016 was entirely consonant with Dr Saunders’ overall view of 23 August 2016 that “almost all of the evidence indicates that the licence should be granted” if “a few issues” were satisfactorily clarified – perhaps by “the provision of a small amount of additional information.”

• Dr Saunders’ draft recommendation on 26 August 2016 of refusal of the licence was inconsistent with his overall view and resulted from his lack of appreciation of the statutory scheme and, particularly, the procedural possibilities of getting that additional information.

270. Entirely unsurprisingly, it seems to me, Dr Saunders responded on 1 September 2016 with an updated draft report recommending, in essence, that the additional information be bespoken – including by way of a request for a Supplementary EIS which ALAB’s secretariat had suggested.513 Understandably, Dr Saunders deferred finalising his report until after the additional information had come to hand and indeed until after a decision had been made whether to hold an oral hearing.

271. IFI describe Dr Saunders’ draft recommendation that further information be sought as a “volte face” from his recommendation that the licence be refused. The accusation is superficially correct but falls away quickly on close examination of the interactions described above. It seems to me that they:

• are primarily characterised by Dr Saunders’ clear overall view that “almost all of the evidence indicates that the licence should be granted” if “a few issues” were satisfactorily clarified – perhaps by “the provision of a small amount of additional information.”

• proceed from failures of communication as between Dr Saunders and ALAB as to the statutory options open to ALAB – including as to seeking further information. Whether this resulted from inadequate initial briefing of Dr Saunders or from failure by him to appreciate briefings given is at this point irrelevant – save to emphasise the importance of such briefings in avoiding the confusion apparent in the e-mails I have just described.

• ultimately reflect the practical working out of procedural recommendations by Dr Saunders, as to bespeaking further information, which are perfectly consistent with his clear overall view as described above. ALAB merely assisted Dr Saunders to the recommendations which logically followed from that clear overall view and which he had failed to make due to his misunderstanding of the available statutory options.

272. It seems that to this point Dr Saunders had envisaged that he would issue only a final report – which, as I have said, he understandably deferred finalising. However, on 21 December 2016 ALAB asked him for an interim report to flag issues requiring further exploration (likely at the oral hearing) before he could make a recommendation as to the outcome of the Appeal. Of itself, this was an unobjectionable request. Indeed it

513 I infer in an oral communication as I have not found reference to it in the e-mails.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

seems to have been very sensible as providing an informed basis on which to embark on further informationgathering and investigation before determination of the Appeal.

273. So, when closely examined, the ALAB/Saunders interactions preceding his interim report of December 2016 are entirely unobjectionable. They do not reflect any prejudgment of, or premature decision as to, the ultimate outcome of the process, any bias external to the process or any unfairness internal to the process.

S.47 Requests & Replies & Decision to hold an Oral Hearing – October to December 2016

274. On 6 October 2016, ALAB issued s.47 requests to MOWI, IFI, NPWS and the Marine Institute. They differed as between recipients but together related to issues including whether and in what degree the Dromagowlane/Trafrask River system supports salmonids and FwPM, sea lice treatments (in particular, EmBz /Slice and alternative treatments when it could not be used), cage design detail and assessment of revised cage design.514 The s.47 replies were in by mid-December 2016.

275. Meanwhile, ALAB met on 22 November 2016. Its lengthy minute suggests detailed and systematic consideration of the matters identified. It also reflects good governance in recording the advice of its advisor to ALAB and ALAB’s views on that advice. Dr Saunders presented to the meeting in light of the s.47 replies by then to hand515 – including what he considered to be queries arising therefrom. Broadly, his recorded advice at that time made its way into his interim report of December 2016.

276. Issues regarded by Dr Saunders and the Board at the meeting of 22 November 2016 as

• having been satisfactorily resolved in favour of the grant of the licence included, o the suitability of the cage and mooring system for the Site – subject to inclusion in the licence of a condition regarding DAFM approval of the system having regard to its 2016 “Protocol for Structural Design of Marine Fish Farms”. o the effect on the benthos of the addition of 4 cages to the proposal. Essentially, ALAB accepted that as the MAB of 2,800 tonnes was not to change and would be spread more widely over more cages, any effect would be more diffuse than modelled. Whether one might agree or not, this was clearly a judgment on the merits of the proposal within ALAB’s competence.

• not having been satisfactorily resolved included,

514 This is not a complete list of the information sought.

515 All but the Marine Institute as to whether there were commercially harvestable Dublin Bay Prawns on the Site.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o certain elements of MOWI’s Pest Management Plan and use of EmBz to control sea lice – though in general the plan was considered robust and there was at present no sea lice problem in Bantry Bay.

Subject to the clarification required, the issue was considered resolved. ALAB “agreed to consider this aspect further as the appeal progresses.”

o issues as to the presence of salmonids and freshwater pearl mussels in the Dromagowlane/Trafrask River system.

277. That ALAB had not prejudged the general issue of grant or refusal of the licence is illustrated in the phrase minuted as to the Pest Management Plan issue and conditions to be imposed “should the Board decide to grant a licence following consideration of all issues.” That ALAB had not prejudged the general issue is also apparent in its decision, taken at that meeting, to hold an oral hearing as to the following issues:

• Risks to salmonids in the Dromagowlane/Trafrask River system;

• Associated impact on the FwPM;

• The robustness of MOWI’s Integrated Pest Management/Single Bay Management Plan dated 26.10.2016.

The Saunders Interim Report of December 2016 & Specific Allegations as to early 2017

278. The Saunders Interim Report is dated 31 December 2016. Inter alia, he considered the then-recent s.47 replies. He recommended an oral hearing “due to the remaining or unresolved uncertainties” and “unaddressed issues”. In fact, as recorded above, in consultation with Dr Saunders, ALAB had already at its meeting of 22 November 2016, decided on an oral hearing and the Interim Report in reality, if anachronistically, merely reflected that decision. While the anachronism is unhelpful it is not of legal consequence. Dr Saunders recommended screening out EIA. He considered as to AA that the Salmon Farm was highly unlikely to have any deleterious effect on the qualifying features of any Natura 2000 Site. As to substantive issues, he expressed various concerns to be considered in the ongoing process.

279. IFI has expressed concern that Saunders/ALAB e-mails in January 2017516 reflected editorial intervention by ALAB in the terms of the Saunders Interim Report. Dr Saunders had enclosed the Interim Report dated 31 December 2016 with changes tracked from an earlier draft. His covering e-mail of 2 January 2017 said: “Once I get confirmation that the Board are happy with my revision I will accept my changes (and incorporate any suggestions from you and the Board) and issue a "clean" version.” This phraseology is at least unfortunate as suggesting that the final terms of his interim report were subject to ALAB approval and as suggesting a willingness to, willy nilly, incorporate ALAB’s suggestions in his report. An e-mail inviting a general response to a draft would not have been objectionable. While an e-mail framed as Dr Saunders’ was could contribute, in combination with other evidence, to a finding of unfairness internal to a process or a want of independence in an expert, there is a danger of being too precious and suspicious as to such

516 Pages 710-712 of Exhibit MCC2.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

interactions. As I have said already, they should not be approached from a standpoint of suspicion or read as if statutes or as if drafted by lawyers.

280. By e-mail dated 10 January 2017, ALAB took up the invitation to suggest changes to the report. ALAB did so in terms not readily comprehensible as I do not have the draft report – I have only the resulting published report – and so, though they incite a degree of concern, I cannot properly appreciate the import of the changes suggested and made.

281. However, it is striking that, in the event and following that e-mail traffic:

• Dr Saunders in his Interim Report considered that the “proximity of the proposed Shot Head salmon farm to the Trafrask embayment entrance might arguably constitute an enhanced sea lice risk, in which a significant infestation event may substantially affect the viability of the river’s salmonid population. The key issue is therefore whether the Applicant’s recently submitted Integrated Pest Management Plan is sufficient to mitigate any future fish farm-derived impact on salmonid populations within the Dromagowlane/Trafrask River.” He was “not fully convinced that the presence of the farm presents no lice infection risk to any salmonid populations associated with the (Trafrask) river. … the .. Integrated Pest Management Plan … will serve to maintain lice levels at manageable levels, but a concern remains in respect of the risk to these populations and the implications for the interdependency with freshwater pearl mussel.”517 As to that Pest Management Plan he said: “Overall, the present situation suggests that lice issues remain a low-level risk. We do, however, remain to be convinced in respect of the Strategy’s ability to safeguard the wild salmonid populations (and by association the freshwater pearl mussel population) in the Dromagowlane/Trafrask River system.” 518

• Dr Saunders’ conclusion in his Interim Report was to recommend “that final licence refusal or approval should be determined subsequent to further consideration of the outstanding technical issues in respect of the potential risk to the salmonid and freshwater pearl mussel populations in the Dromagowlane/Trafrask River system.”

• Two of Dr Saunders’ important conclusions, that EIA and AA were unnecessary, were not followed.

282. It is true that Dr Saunders’ Interim Report, both generally and in various specific respects, favoured the MOWI application. It is even the case that, as to the outstanding matters, his recommendation of deferral of a decision implied at least the possibility of a grant of the licence. But it cannot be said that the Interim Report bears the fingerprints of improper influence by ALAB or suggests prejudgment or other bias by either Dr Saunders or ALAB. Not merely were outstanding and objectively important issues identified

517 pp 96 & 95 – underlined in original.

518 pp 96, 95 & 98 – underlined in original.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

which had the potential to result in a refusal of the licence but ALAB had taken what, for it and at the time at least, was the rare, major and demanding step of directing an oral hearing. It must be remembered that an oral hearing provides a public forum for, inter alia, the expression (including by experts), and scrutiny of licence applications and objections to licences. As Clarke CJ said in Klohn519 as to public oral hearings in court, while they are not always needed, “It is almost inevitably the case that issues become much clearer as a result of an oral hearing. ……. Where disagreement remains it is almost always the case that the differing positions of the parties emerge with greater clarity after oral debate and questioning from the Court.” Of course, failing to hold an oral hearing may prompt judicial review. Nonetheless, if ALAB was biased in favour of MOWI’s application, deciding on an oral hearing and the public scrutiny it inevitably involves was a very odd way of carrying its bias into effect. Much more plausibly, the decision was consonant with the outstanding and objectively significant concerns identified in its minute of its meeting of 22 November 2016 at which oral hearing was decided upon and with those concerns as later expressed in the Interim Saunders Report of December 2016.

283. Given the presumption of validity and the onus of proof in judicial review, I cannot infer, on the information available and as a matter of probability, that ALAB’s suggestion to Dr Saunders of changes in his draft interim report and the resulting changes in his interim report reflect unfairness, bias or were improper.

284. However I should express one concern. As stated, Dr Saunders’ revised interim report was sent to ALAB on 11 January 2017. But, at ALAB’s request,520 it was backdated to 31 December 2016. To be clear, while I would not rule out the possibility of exceptions to the rule for good reason (though none occur to

me) such requests must not be made by commissioning statutory authorities and, if made, must not be complied with by the reporting independent experts. Reports must be accurately dated. To inaccurately date reports, especially at the request of a commissioning statutory authority, only invites suspicion – warranted or not – of what McKechnie J in Nurendale termed “massage” of expert reports. However, it would be disproportionate, at least generally, to quash an impugned decision on that account alone and in the absence of evidence of what I might call substantive massage or that relevant parties were, by such misdating, materially misled. There is no such evidence here.

285. For the reasons set out above, I reject the challenge to the impugned Aquaculture Licence as it is based on any allegation of impropriety in the period to January 2017 as to the terms of and redrafting of the Saunders report which became the Saunders Interim Report dated 31 December 2016.

519 Klohn v An Bord Pleanála [2017] IESC 11.

520 11 January 2017 at 10:04.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Late 2016/early 2017 – Candour & Pleadings Issues

286. Perhaps in error, I have decided the “late 2016/early 2017” issue on its merits. Not least I am conscious that it would be difficult to ignore them in the context of the “overview” which it is suggested I should take of the process and to which I will come in due course. But it is important to note that, in this regard, contending accusations of lack of candour against ALAB and lack of pleadings against IFI were made.

287. Though leave to seek judicial review had been sought earlier, directions had required that it be sought on notice. Ultimately, IFI got leave on 5 December 2022 on foot of an amended Statement of Grounds dated 2 December 2022.521 IFI had made a Freedom of Information (“FoI”) request in March 2022 for “All records held and/or under the control of ALAB, in relation to the appeals”. ALAB later that month, indicated that it intended to refuse the request as unreasonably burdensome – explaining in some detail why that was so. But ALAB, by that letter, suggested a revised FoI request and suspended the timeline for a final FoI decision pending response by IFI. By affidavit of 30 June 2022522 in the IFI proceedings, ALAB scheduled a 34-page tabulated chronology of events since the date of the EIS of May 2011 and exhibited in electronic form over 5,000 pages of “documents from ALAB’s file on the application”. The account in the affidavit of their organisation and the system of folders used went some way to making them navigable. Though this was not an FoI reply and it did not purport to exhibit every document in ALAB’s possession it, in substance and inevitably, went a considerable way towards one. IFI sent a reformulated FoI request on 6 September 2022. ALAB’s FoI decision issued on 22 October 2022. It granted access to all categories sought – partial in the case of some, including Category 5 – records as to ALAB’s engagement of consultants and/or advisers in the Appeal. It enclosed an index of the documents to be released and, as required by s.26 of the FoI Act 2014, stayed the release of the Category 5 records (only) pending expiry of time limits for review or appeal of the decision by consultees to the FoI application. The process was completed on 24 November 2022 with delivery by ALAB to IFI of the Saunders/ALAB correspondence of late 2016/early 2017 relevant to this issue. It did not include the draft Saunders reports but made their existence apparent. IFI’s motion for judicial review, on foot of its leave, issued on 13 December 2022. IFI did not, then or thereafter or at trial, seek,

• to amend its Statement of Grounds to reflect the content of that Saunders/ALAB correspondence, or

• discovery or directions in the proceedings requiring production of the draft Saunders reports.

288. There was some inter partes correspondence in early 2023 in the IFI Proceedings seeking the draft Saunders reports of 2016/2017523 – in the context, inter alia, of the FoI process and of ALAB’s duty of candour in judicial review. On 15 February 2023, IFI exhibited524 documents which ALAB had provided in the FoI process – including the Saunders/ALAB correspondence of late 2016/early 2017 – and which IFI

521 It had been amended a number of times from that originally proffered. It had last been amended on foot of an order of 21 November 2022 granting

liberty to amend.

522 Sworn by Francis Dowling – Exhibit FD1.

523 Between 2 February 2023 and 10 February 2023 – in proceedings commenced in May 2022.

524 Exhibit MCC2 T53 to the Affidavit of Mona-Claire Costelloe sworn 15 February 2023 for IFI.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

considered relevant additions to the documents already exhibited. That affidavit did not complain of the non-provision of the draft Saunders reports of 2016/2017 or seek discovery thereof. At trial, IFI asserted that by failing to exhibit the draft reports ALAB was in breach of its duty of candour.

289. ALAB replied that no duty of candour had arisen as to those draft Saunders reports of 2016/2017 as the complaint that it had “massaged” the process to finalisation of Dr Saunders’ Interim Report in late 2016/early 2017 was not pleaded and was first made in IFI’s written submissions of 15 February 2023. It says that its duty of candour arises only as to the case pleaded against it and that IFI had not, on obtaining the email traffic described above, sought to amend its pleadings.

290. IFI in turn reply that this complaint was pleaded in general terms in their Ground 5 and that the issues raised are of inference from the evidence and are not new grounds. They say the Saunders/ALAB correspondence of late 2016/early 2017 only surfaced long after pleadings had closed and that, by later letter agreeing to the exhibition of the documents thus disclosed ALAB, waived any pleading point. As I have noted, IFI exhibited the documents on 15 February 2023.525 IFI, by letter dated 2 February 2023, had alleged that ALAB had breached its duty of candour by failing to exhibit documents, including that Saunders/ALAB correspondence. ALAB by letter dated 10 February 2023, rejected any suggestion that the directions timetable be delayed, disputed the allegation of lack of candour and included the following:

“Our Client considers that all documents held by it and relevant to the challenge to its Determination have been exhibited in the three sets of proceedings.

We do not consider that it is necessary to exhibit the additional documents released further to the FOI request which have already been exhibited or are not relevant to the proceedings. However, in the interests of progressing matters, and subject to the Court, we have no objection to you filing your supplemental affidavit and including reference to same in your legal submissions.”

It is clear that ALAB maintained its position that the documents in question were “not relevant to the proceedings”. That is the same as saying not relevant on the pleadings. I do not see this letter as ALAB waiving any pleading point. It simply agrees that the documents can be exhibited and submissions made on them.

291. As to the duty of candour in judicial review, Clarke CJ in RAS Medical526 cited Huddleston527 as authority that public authorities should conduct public law litigation “with all cards face upwards on the

525 By Affidavit of Mona-Claire Costelloe, solicitor, sworn 15 February 2023.

526 RAS Medical Limited v Royal College of Surgeons in Ireland [2019] IR 63.

527 R v Lancashire County Council Ex p. Huddleston [1986] 2 All E.R. 941.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

table”. Humphreys J in Shao528 cited Treasury Holdings,529 McEvoy530 and Fordham531 for the rule that, “A defendant public authority and its lawyers owe a vital duty to make full and fair disclosure of relevant

material” and Citizens UK532 for respondents’ “duty of candour and cooperation with the court” to “assist the

court with full and accurate explanations of all the facts relevant to the issues which the court must decide” because “the underlying principle is that public authorities are not engaged in ordinary litigation, trying to defend their own private interests. Rather, they are engaged in a common enterprise with the court to fulfil the public interest in upholding the rule of law.”

292. In the UK, in Hook533 the rule was said by Denning MR to extend to “documents/ information which will assist the claimant's case and/or give rise to additional (and otherwise unknown) grounds of challenge” as “.. the applicant knows very little of what has happened behind the scenes. He only knows that a decision has been taken which is adverse to him, and he complains of it.” Scarman LJ said: “the applicant is in a state of only partial knowledge as to the matters affecting his rights. He knows what has happened to him and what has been done in his presence, but he knows nothing else. At that stage he does the best that he can with his advisers in putting his material into his statement of case, but, of course, if that material is enough to persuade the Divisional Court to grant leave, then there comes a duty upon the respondent to file evidence, if need be, and in filing evidence to fill up the gaps in the knowledge of the court, bearing in mind that the court

is exercising a supervisory jurisdiction”. It seems that the UK state, in the form of the treasury solicitor, also

takes that view.534

293. However, it has been suggested535 that “this generous view may represent aspiration, not actuality” and has been noted that in Huddleston,536 perhaps the origin case on candour in judicial review, Parker J had put the matter more cautiously:

''I would not wish it to be thought that once an applicant has obtained leave he is entitled to demand from the authority a detailed account of every step in the process of reaching the challenged decision in the hope that something will be revealed which will enable him to advance some argument which has not previously occurred to him.''

And in Citizens UK537 the respondents’ duty was said to be to explain all facts “relevant to the issues which the court must decide”. Those issues, ordinarily, are those identified in the pleadings. It seems very arguable that imposing on respondents a duty of candour and disclosure beyond issues arising on the pleadings was,

528 Shao v Minister for Justice (No. 2) [2020] IEHC 68.

529 Treasury Holdings v NAMA [2012] IEHC 66, Finlay Geoghegan J, §§126 and 127.

McEvoy v Garda Síochána Ombudsman Commission [2015] IEHC 203, McDermott J.

531 Judicial Review Handbook (5th ed.) §10.4.

532 R. (Citizens U.K.) v Secretary of State for the Home Department [2018] EWCA 1812, Singh L.J., §106.

533 R v Barnsley Metropolitan Borough Council ex p. Hook [1976] 1 WLR 1052.

534 Guidance on Discharging the Duty of Candour and Disclosure in Judicial Review Proceedings Treasury Solicitor's Department January 2010, citing Hook.

535 Supperstone, Goudie and Walker on Judicial Review/Chapter 11.

536 R v Lancashire County Council, ex parte Huddleston [1986] 2 All ER 941 CA.

537 R(Citizens U.K.) v Secretary of State for the Home Department [2018] EWCA 1812, Singh L.J., §106.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

in Hook, based on a view of the status and importance of pleadings in judicial review not consonant with present Irish law on that issue. To do so would seem to go beyond present Irish authority on the issue, though I am not clear the question has been decided. Further, it may be arguable that since Hook was decided in 1976 the landscape of public administration has greatly changed in terms of duties of publication of decisions and the materials on which they are based and the scope of the requirement to give reasons. Accordingly, the view that “the applicant knows very little of what has happened behind the scenes” is, arguably, no longer typically correct.

294. On the arguments made (the account above goes beyond them) I am not prepared to impose such duty of candour on ALAB to exhibit or disclose documents irrelevant to the issues joined in the pleadings. I leave the issue open to further argument in another case.

295. That conclusion renders it necessary to resolve the pleading issue. IFI relies on its pleaded Core Ground 5, which reads:

“The manner in which the appeal process was conducted by the first Respondent, including the large number of extensions and the directing by the first Respondent of the production of further documentation, or the bespeaking of additional reports by the first Respondent, would have been likely to have created the impression in an objective observer that the first Respondent had determined to grant the appeal, and was engaged in collecting the material to justify this decision, such as to give an appearance of objective bias and a lack of independence, by reason of which the determination should be set aside.”

Optimistic pleading of the word “including” does not allow expansion on grounds actually pleaded where, as in judicial review, pleading rules require precision and particularity. Here, the plea is of objective bias evinced by the large number of time extensions granted, by directing the production of further documents, and by bespeaking additional reports. Even these pleas are insufficiently particular – though that is not a criticism of them as core grounds. So one turns to the particulars in the Statement of Grounds.

296. The IFI Statement of Grounds of 2 December 2022 includes a paragraph538 pleading that,

• ALAB was obliged “in pursuance of its duties to ensure transparency and public participation to publish all documents relating to the progress of the appeals”.

• IFI relies on all those documents “it has been made aware of” and reserves the right to refer to such documents “as should have been so published” and which are provided to the Court in the proceedings.

• IFI reserves the right to apply, if “necessary”, for discovery of any materials which were before ALAB in the

appeal. (As stated, it did not do so as to the draft Interim Saunders reports).

538 Parts 2&3: Particulars of Grounds & Factual Grounds, §12.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Of this I comment that:

• As stated, IFI was aware of, but did not seek discovery of, the draft Saunders Interim Reports.

• Beyond vague invocation of “duties to ensure transparency and public participation”, the Statement of Grounds did not plead any legal basis for the assertion of a duty “to publish all documents relating to the progress of the appeals”. While ALAB has certain statutory obligations of publication it was not alleged that it had breached them and no authority for such a “blanket” duty of publication of “all documents”

was opened to me.

• Nor, as I have said, did the grounds specify, o The Saunders/ALAB correspondence of late 2016/early 2017 (of which IFI were in possession) o The draft Interim Saunders reports (of the existence of which IFI were on notice from the Saunders/ALAB correspondence in its possession).

I am conscious that IFI only got the Saunders/ALAB correspondence of late 2016/early 2017, a few days before delivering its Statement of Grounds of 2 December 2022. But I do not accept that it lacked time to make any necessary applications between then and the trial which started in mid-April 2023 or that any such consideration could justify my deeming issues to have been pleaded which had not been pleaded.

Conclusion

297. Accordingly I find that ALAB was not in breach of any duty of candour and that its objection on pleading grounds to IFI’s reliance on the Saunders/ALAB correspondence of late 2016/early 2017 and any IFI reliance on the existence or non-publication of the draft Saunders reports is well-founded. If I am wrong in those conclusions, and as will have been seen, IFI’s complaints fail on their merits. And in any event, they cannot constitute bias as they are internal to the Appeal process.

Bias – Chair of Oral Hearing – 2017

298. The Sweetman Applicants assert objective – not actual – bias, on the basis of the circumstances in which Professor McIntyre chaired ALAB’s oral hearing in the Appeal – or, more accurately given the sequence of events, those in which he reported on the oral hearing. I will assume, without finding, that those events can be properly regarded as external to the licensing process for purposes of analysis of the issue of bias.

299. Professor McIntyre was a member of ALAB from May 2013. From then to December 2016 he was paid the ordinary fees payable for the ordinary part-time duties of an ALAB member. ALAB’s annual report 2016 (dated 1 December 2017) disclosed,539 that that in October 2017 it came to ALAB’s attention that

539 Pursuant to the 2016 Code of Practice for the Governance of State Bodies.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

between May 2013 and December 2016 fees totalling €18,231 were paid “under” the one person one salary principle, to an ALAB member who is a public servant. It is clear, and was no doubt clear to anyone reading the report, that the phrase cited should have read “contrary” to the one person one salary principle. Francis Dowling of the ALAB secretariat, explains540 that Professor McIntyre was the ALAB member in question and had been paid in error, not having been advised that, for purposes of the principle he was classified as a public servant, such that he would not be entitled to be paid such fees. He says that the issue first came to the attention of both ALAB and Prof McIntyre in October 2017. As the complaint is of objective bias, that explanation is not disputed.

300. The 2016 Annual Report stated that ALAB worked with the DAFM and the ALAB member concerned, resolved the issue satisfactorily and that all fees paid in error were recouped in full. Mr Dowling records that, on discovery of the error in October 2017, Professor McIntyre agreed, without objection or delay, to refund the fees and had refunded them in full by the end of March 2018. ALAB agreed that repayment could be made by way of an offset against other fees due to Professor McIntyre for his work in chairing the Shot Head Oral Hearing. Again, I do not understand any of this to be in dispute.

301. At ALAB’s meeting of 16 November 2016, Professor McIntyre had been asked by the ALAB Chairperson, and had agreed, to chair the oral hearing at a fee payable at the same rate as for a previous oral hearing in a different appeal. At this stage, Professor McIntyre was still unaware that he had been mistakenly overpaid ALAB fees and properly expected to be paid additional fees for chairing the oral hearing. Clearly, in and about his appointment as chair of the oral hearing, neither ALAB nor Professor McIntyre could have been in any way influenced by the fee repayment issue of which they were, as yet, ignorant. No doubt that remained his mindset throughout the oral hearing in February and September 2017 as only in October 2017, was the overpayment issue discovered. In November 2017 he reported to ALAB on the oral hearing.

302. Professor McIntyre’s report recommended that an aquaculture licence issue. That of itself cannot imply bias. But, far from being unalloyed advocacy of the grant of a licence, that recommendation was highly conditional. It recommended:

• that a supplemental EIS and supplemental EIA address o risk of sea lice infestation of wild salmonids, any resulting implications for FwPMs. o potential impact of salmon farm waste on water quality, having particular regard to the Water Framework Directive.

• investigation of the need for a supplemental AA screening as to possible adverse effects on otters, seals and wild birds.

540 Affidavit 28 June 2022.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• that ALAB make every effort to consider the potential impacts of large-scale farmed salmon escapes.

Professor McIntyre went on to recommend that licence conditions address certain matters “If, on the basis of such further information, the Board should decide to grant an aquaculture licence”.541

303. It is clear that Professor McIntyre’s report generally favoured the grant of an aquaculture licence. Of itself, that is unimpugnable as bias. However, he had clearly taken considerable account of issues raised in the oral hearing in opposition to a grant and he had recommended actions which both reflected appreciable conditionality in his general view and clearly implied that a grant was not assured. Not least, he had reinvigorated not merely the EIA process but an AA process which, depending on its outcome, had the potential to require refusal of the licence application as a matter of law. That these were significant and demanding recommendations is apparent in the ensuing investigative and reporting processes: it was not until June 2021 that ALAB was able to make its determination.

304. It is also to be noted that Professor McIntyre’s recommendation of efforts to consider the potential impacts of large-scale farmed salmon escapes must be read in light of the considerable controversy at the oral hearing as to the status of the DAFMs’ report and associated documents on the large-scale escape of 230,000 salmon542 from a farm in Bantry Bay in 2014. I address this controversy in more detail elsewhere.

305. Suffice it to say here that Professor McIntyre’s recommendation was made against the background of the potential significance of such escapes for the prospects of the grant of a licence and continuing DAFM resistance to the release of the report and associated documents despite:

• its appearing at the oral hearing that, at least, some of the associated documents had been released to others months earlier – in March 2017 – on foot of a Freedom of Information request.

• the fact that the DAFM had, by letter to ALAB of 28 July 2017, described the report as “at an advanced state of completion”.

• ALAB’s having, a fortnight before the resumed oral hearing in September 2017, withdrawn its s.47 requirement that the DAFM release the report to it. It is not apparent that, in resisting the s.47 request, the DAFM had made ALAB aware of the Freedom of Information release in March 2017.

306. Many objectors had raised the issue of escapes and Professor McIntyre was no doubt aware that Dr Saunders’ Interim Technical Advisor’s report of 31 December 2016 had included the following:

541 Emphases added.

542 Figure taken from Saunders’ Interim Technical Advisor’s report of 31 December 2016 p63.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“… the EIA543 undervalues the knowledge to be gained from this escape,544 in particular with respect to the genetic and sea lice impacts it may have had on the wild salmon population of the Bantry Bay catchment. ……. Contrary to the statement in the EIA, the escape of these fish is not in question. Some 230,000 fish were absent from the cages following the storm.” 545

“Escaped farmed salmon can impact on wild salmonid populations via resource competition in the riverine environment and through interbreeding and subsequent genetic dilution of native traits, resulting in reduced fitness.”546

“…….. the effect of lice on the wild Bantry Bay salmonid population should also be considered alongside the potential impacts from the 2014 salmon escape in Bantry Bay ……. because of the known relationship between escapee fish and the augmentation or enhancement of natural infections in wild stocks.”547

“The impacts of escaped farmed salmon can be significant where they achieve proportionally large numbers in comparison to the population size of wild conspecifics548. Farmed escapee fish can compete with wild salmon for resources, may breed with wild counterparts resulting in reduced genetic fitness, and constitute a disease and parasite transfer risk. The prevention of escapes and approaches to reduce impacts are of fundamental importance to both the interests of the aquaculture industry and the conservation of declining wild salmonid stocks.”549

“Despite the requirement for statutory reporting, the number of escapes since 2004 does not appear to have been made publicly available.” 550

Dr Saunders was “… unable to evaluate the risk of fish escapes from the Shot Head site.”551

This is by no means a full account of Dr Saunders’ consideration of the escape issue – though it does convey a reasonable impression of his view. My purpose is to illustrate that further investigation and resolution of that issue had appreciable potential to tend against the grant of an aquaculture licence.

307. Further, Professor McIntyre can only have been conscious that the risks posed by farmed salmon escape are implicitly but clearly recognised by the Oireachtas in that the 1997 Act,

543 Presumably a reference to the Minister’s EIA.

The 2014 escape.

545 §9.2.

546 §6.5.5.

547 §9.1.

548 Sic – presumably it should read “conspecies”.

549 §9.2.

550 §9.2.

551 §9.2.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• specifies the possibility of licence conditions imposing measures for preventing fish escapes.552

• allows the Minister to authorise the licensee or “any person” to take specified action to recapture such stock.553

• constitutes the recapture of escaped farmed salmon a discretionary statutory function of IFI.

308. Professor McIntyre must also have been conscious that, despite this statutory recognition of the importance of recapturing escapees, ALAB’s technical advisor, Dr Saunders, had advised that “In reality, any attempt to recover escapee fish is likely to result in a very low level of success. It has been previously reported (Thorstad et al. 2008) that less than 3% of escaped salmon have been recaptured through organised fishing after large escape episodes.” 554

309. It is clear, and would be clear to any reasonable observer, that, far from seeking to leverage his report as to the oral hearing to evade repayment of the overpaid fees by producing a report favourable to ALAB’s supposed desires to grant a licence and clearing the way for the grant of a licence, Professor McIntyre produced what might well (dubiously assuming ALAB’s supposed determination to grant the licence) have been considered a highly inconvenient report – as, even on that dubious view and given what ensued, it proved to be. His report envisaged significant, further and robust investigation of highly controversial issues before a licence might be granted.

310. Also, and far from leveraging his report to evade repayment of the overpaid fees he, without objection or delay, set off against them the fees he had expected to be paid from the oral hearing. I should add that in any event I have little doubt that he, ALAB and an objective observer would have considered, in light of the ordinary requirements of governance and reporting, any hope (if, which I do not think, he can have harboured any) of evading repayment of the fees to be properly forlorn.

311. It is convenient here to address a related complaint made by IFI. Professor McIntyre’s report is dated

8 November 2017. The following day, 9 November 2017, he presented it in draft to a meeting of ALAB. The minute of the meeting records its purpose as “for Prof McIntyre to present his final draft of the Shot Head Oral Hearing report to the Board for final signoff subject to some editing and clarifications.” IFI chooses to read this as redolent of what McKechnie J in Nurendale termed “massage” of reports. The words cited are loose as to who is to do the sign-off, editing and clarification. But it is wrong to read such minutes as if they were statutes. The minute describes Professor McIntyre as submitting his report pursuant to s.59 of the 1997 Act which requires written report to ALAB and the words cited must be considered in that context. Any such report must be the independent and exclusive product of the rapporteur’s mind – especially if the rapporteur is, as is often the case, to make a recommendation as to the anticipated decision. Clearly, care is

552 s.7.

553 s.77.

554 Interim report 23 December 2016.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

required in any oral interaction between the chair of an oral hearing and the deciding body on whose behalf the oral hearing is conducted. Certainly, as was done here, any such discussion should be formal and carefully minuted. The decision-maker should never seek to procure that the report make a recommendation other than that on which the rapporteur independently decides. If it disagrees, the

decision-maker’s proper response is not to try to change the recommendation but, if appropriate, to reject it in whole or in part and give reasons for doing so. However, once these considerations are determinedly borne in mind, I do not think discussion is forbidden between a decision-maker and its rapporteur of the latter’s draft report. For example, it may assist the rapporteur in clarifying the draft, remedying omissions or addressing issues brought to his/her attention in such a discussion on which the content of the report may properly bear and assist the decision-maker. It may well thereby improve the quality of decision-making.

312. The reality must also be borne in mind that, where the rapporteur is also a member of the decisionmaker and will participate in the decision, that rapporteur will inevitably and properly be drawn, at least after the report is presented, into detailed discussion of its content. Professor McIntyre remained a member of ALAB until June 2019. But he recused himself (I am not convinced necessarily – but I need not decide that question) from involvement in the MOWI application after submitting his report on the oral hearing. An objective observer would see this as further reassurance of the absence of bias by reason of Professor McIntyre’s position as to recoupment of monies mistakenly paid to him (if I needed reassurance, which I do not).

313. Nurendale was cited on this issue of interaction between decision-makers and rapporteurs as to drafts of their reports. Nurendale was egregious on its facts. The “massaging” of the reports was apparent in drafts which “contained comments written by the respondents indicating which parts of earlier drafts were acceptable to them, and either deleting or rewording those parts which would not have supported their position. There were also email references to meetings with the authors of these reports as well as notes of some meetings ….which would indicate that the findings of the reports were a foregone conclusion”. It must particularly be remembered that this occurred in Nurendale against a backdrop of a strikingly egregious statement by an assistant city manager that he would do everything in his power to ensure that Nurendale would not collect domestic waste in the Dublin region.

314. In considerable contrast with the egregious facts of Nurendale, the remainder of the minute of the ALAB meeting of 9 November 2017, at which Professor McIntyre’s report was discussed, seem to me unobjectionable. They record in effect Professor McIntyre’s oral account of and clarification of his report. There is no hint of improper influence on the finalisation of his report. If anything, the minute concentrates on the difficulties with the Shot Head proposal rather than its virtues.

315. In any event, I reject the allegation that the content of the minute of 9 November 2017 disclosed bias as it was not pleaded.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

316. In short, I see no reason to conclude that a fair-minded, objective and fully-informed observer would have had any suspicion of bias on Professor McIntyre’s part or of any improper interference by ALAB in the terms of his report. Inasmuch as O’Donnell J in Shatter555 has described allegations of even objective bias as

“ugly”, the allegation as to Professor McIntyre seems to me to fit that description. As O’Donnell J said, “An

allegation of bias against a decision-maker or other person performing a public function is one which should

not be lightly made. Bias is an ugly allegation, and the label can cling once applied.” I reject the complaints of

objective bias as they relate to Professor McIntyre.

Bias & Unfairness – Interactions as to AA/NIS

Introduction & Pleading Point

317. IFI submits that events relating to the iteration and finalising of MOWI’s NIS556 of July 2020 betoken objective bias in ALAB. ALAB responds that these allegations were not pleaded.

318. IFI relies on its pleaded Core Ground 5 alleging objective bias – which I have set out above. It does cite “the directing by the first Respondent of the production of further documentation, or the bespeaking of additional reports by the first Respondent”. As I observed earlier, these pleas required particularisation. Those events relating to the iteration and finalising of MOWI’s NIS were apparent to IFI on the same timescale as those relating to its submissions as to ALAB’s interactions with Dr Saunders in “late 2016/early 2017”. They are not particularised in IFI’s Statement of Grounds and fall to be treated, on the pleadings point, as were ALAB’s interactions with Dr Saunders. The O’Toole reports were put up on the ALAB website on 1 June 2021 and the latest iteration of IFI’s Statement of Grounds is dated December 2022. Yet the particulars set out in that Statement of Grounds do not complain of this issue. I reject these complaints as not pleaded. In any event, those issues cannot constitute bias as they are clearly internal to the Appeal process.

319. The Sweetman Applicants on affidavit,557 described these interactions as to, and iterations of, the NIS as ALAB’s “process of correcting the homework of the developer, behind the backs of the public” and as tending to confirm “the direction of the process having been fixed prior to June 2019”. However, the point is not pleaded in those proceedings – either as to bias or unfairness.

555 Shatter v Guerin [2019] IESC 9; [2021] 2 IR 415.

556 Natura Impact Statement within the meaning of the Habitats Directive.

557 Affidavit of Noel Carr Sworn 27 September 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

320. However, and unusually, I propose to consider these issues nonetheless and as I considered the issue as to ALAB’s interactions with Dr Saunders in “late 2016/early 2017”. I do so as I am conscious that it would

be difficult to ignore them in the context of the “overview” which it is suggested I should take of alleged bias

in the process - to which I will come in due course.

AA - Public Participation

321. The Habitats Directive does not explicitly require public participation in AA. Article 6(3) requires it only if it is ‘considered appropriate’. The Irish Habitats Regulations 2011, Article 42(13) are framed accordingly. However, the CJEU in LZ #2,558 interpreting the Directive in light of the Aarhus Convention, has held that the public is entitled to participate “effectively” in AA. The 2011 Regulations were not updated to provide for this interpretation until 21 June 2021559 which required publication of the NIS and “any other information of documentation relevant to the application in the public authority’s possession” such that “where additional information is published” the public shall have 30 days thereafter to make submissions. Of course this amendment postdated the impugned determination. But it is at least illustrative of the scope of requirement imposed almost 5 years earlier in LZ #2. It seems to me that an interpretation of the unamended Article 42(13) in conformity with LZ #2 would not be contra legem in considering that in substance it amounted to a position that public participation was to be ‘considered appropriate’ in all cases of AA. In fairness to ALAB it accepted and applied the general proposition of public participation in the Appeals as they encompassed AA.

AA – the Facts & Comment thereon

322. It will assist, in understanding what follows, to understand the iteration of consideration of the proposed Salmon Farm for purposes of the Habitats Directive. The EPA advised MOWI560 by letter of 3 September 2009 and by reference to Directive 92/43/EEC (Habitats), Directive 79/409/EEC (Wild Birds) and the “Natural Habitats Regulations”,561 that AA562 may be required and that a “reasoned response” should be provided if it was considered that AA was not required. So, in formal terms, MOWI was from the first, on notice of the need for AA Screening and/or AA.

323. Here is not the place for an historical account of the transposition of the Habitats Directive or the 1997 Natural Habitats Regulations or their relationship to aquaculture appeals. But it is fair to say that by the

Case C-243/15, Lesoochranárske zoskupenie VLK, Judgment of 8 November 2016, §46-49. Also known as “Slovak Brown Bears #2”. The case related

to participation by NGOs but is cited for public participation in Commission Notice C(2021) 6913 final, 28.9.2021 – Assessment of plans and projects in

relation to Natura 2000 sites – Methodological guidance on Article 6(3) and (4) of the Habitats Directive 92/43/EEC §3.2.6. see also Commission Notice

2019/C 33/01 Managing Natura 2000 sites. The provisions of Article 6 of the Habitats Directive 92/43/EEC §4.2.7.

559 S.I. No. 293 of 2021 – European Union (Birds and Natural Habitats) (Amendment) Regulations 2021.

560 In fact Silver King Seafoods. It was later taken over by MHI, now MOWI. MOWI included this letter in Volume 2 of its 2011 EIS at p.42.

561 Clearly a reference to the European Communities (Natural Habitats) Regulations (S.I. No. 94 of 1997).

562 Appropriate Assessment, within the meaning of the Habitats Directive.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1997 Regulations an EIA was generally, and erroneously as a matter of EU Law, deemed to constitute an AA – albeit subject to the proviso that a proposed development could be permitted only after it had been ascertained that it would not adversely affect the integrity of a Natura 2000 Site. As the 2011 EIS is dated May, the licensing applications were made in June 2011 and the corrective Birds and Natural Habitats Regulations 2011563 were published and came into effect in September 2011, it is unsurprising that MOWI’s licensing application was not accompanied by a Natura Impact Statement (“NIS”). The 2011 EIS did mention the Habitats Directive, local European Sites and protected species but at no point and, despite exhibiting the EPA’s letter, did it address the need for or purport to perform an AA or address risk of adverse effect on the integrity of a Natura 2000 Site. However, and whether correctly or not in light of the corrective necessity of the 2011 Regulations, the absence of an NIS appears to have persisted until July 2020 in the ALAB appeal.

324. Save Bantry Bay in a submission to the Minister, and perhaps other objectors,564 had complained of the absence of an AA. Even so, the Minister’s 2015 EIA, in identifying the applicable legislation,565 did not identify the Habitats Directive or either of its 1997 and 2011 implementing regulations. It did identify relevant SACs566 and did briefly refer to the Habitats Directive and implementing regulations as applicable to seals567 and also in the context of the Control of Dangerous Substances in Aquaculture Regulations 2008.568 However it did not purport to perform an AA. Nor did the civil servants’ recommendations to the Minister which, by his agreement, became the record of his decision dated 5 September 2015, purport to perform an AA. Amongst the exhibits, a Marine Institute report to the Minister dated 27 January 2014569 records consideration of “Interactions with Natura 2000 sites and protected species” as to which the Marine Institute considered that the proposed fish farm “does not pose significant risk to the conservation features of the adjacent sites” and “there are no grounds for refusal based on environmental considerations.” But the report does not in this respect cite or invoke the Habitats Directive, the 1997 and 2011 implementing regulations, the concept of AA or the level of certainty required in AA and it does not amount to a record of an AA. It is, to put it at its mildest, surprising that as late as a decision of 2015, by which time the 2011 Regulations had reflected a proper appreciation of the great difference between EIA and AA, the Minster did not screen for AA or perform AA.570

325. In short, I have neither found nor been directed to any AA or AA Screening by the Minister as competent authority and, on the evidence before me, neither had been done before the matter came, by way of appeal, to ALAB. SWI’s appeal asserted that there had been no AA of cumulative effects, though it omits to make the more obvious point that no AA at all had been done. However, the appeal by Save Bantry

563 European Communities (Birds and Natural Habitats) Regulations 2011.

564 As available to me objections are anonymised which makes the position a little unclear, but the Save Bantry Bay objection may have been adopted by

and enclosed with other objections.

§3.

566 Special Area Of Conservation within the meaning of the Habitats Directive.

567 §15.1.

568 It cited the European Communities (Control of Dangerous Substances in Aquaculture) Regulations 2008 (S.I. No. 466 of 2008) as introduced to effect

the Dangerous Substances Directive (2006/11/EC), Habitats Directive (92/43/EEC) and Water Framework Directive (2000/60/EC).

569 MI reports dated 2 April 2014 and 14 January 2015 are also exhibited but are not here relevant.

570 This seems to be confirmed by the Minister’s representative at the Oral Hearing on 14 February 2017 at 12:42 on the misconceived basis that the Shot

Head Site was not in a European site. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Bay makes that point directly and as far as I can see correctly: “An Appropriate Assessment as required by the Habitats Directive has not been completed” and it asserts that to grant the licence “without conducting an Appropriate Assessment is inappropriate, and therefore the license should be revoked”.

326. The question of AA was addressed thereafter in the Interim Report dated 31 December 2016 of the Technical Advisor, Dr Saunders. It identified571 the relevant SACs572 and SPAs573 and their SCIs574 and generally deemed adverse effects unlikely. A brief paragraph headed “Screening for Appropriate Assessment”,575 though expressing the view that “all supporting scientific knowledge indicates that the proposed fish farm is highly unlikely to have any deleterious effect …. on the qualifying features of any of the designated sites”, can only be described as conclusionary and perfunctory, even when considered in the context of other content of the Interim Report to which it is cross-referenced.

327. Prof McIntyre, Chairman of ALAB’s oral hearing suggested in his report of November 2017 that gaps in the EIS might require “supplemental” AA screening. In fact, there had been no original AA screening to which further screening could have been supplemental. Appellants at the oral hearing had raised the necessity of, and absence of AA Screening, an NIS and of AA and observed that, even at that stage, ALAB had not turned its mind to whether an AA should have been required576 (including as to common seals and wild birds577). All this represents considerable and unexplained delay by ALAB in getting to grips, as to AA, with the Appeal made in October 2015.

328. In any event Prof McIntyre, in his report of November 2017 to ALAB, recommended578 desk-top studies, which might indicate the need for “supplemental” AA screening of potential impacts on:

i. the otter population of the Dromagowlane and Trafrask catchments, including the potential impact of declining wild salmon stocks, ii. common seal populations in the Glengarriff Harbour and Woodland SAC, and iii. wild birds in nearby SPAs.

329. Prof McIntyre’s recommendation resulted in the following lengthy and complex sequence of events. I list these events partly as substantively informative, but also as bearing on the allegations of delay and that an objective observer would reasonably suspect that ALAB was contriving to grant the licence despite the deficiencies in MOWI’s application.

§5.4.

572 Special Area Of Conservation within the meaning of the Habitats Directive.

573 Special Protection Area within the meaning of the Wild Birds Directive.

574 Species of Conservation Interest for which SACs and SPAs are designated as such.

575 §8.0.

576 Oral Hearing on 14 February 2017 at 16:00.

577 Oral Hearing on 20 September 2017 at 15:50 et seq.

578 Amongst other recommendations.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Date Event

24 Otter Impact Assessment (Saunders – desk-top study) November • Dr Saunders asserts that on the basis of overwhelming scientific evidence, it can only be 2017 concluded that a fish farm at Shot Head is highly unlikely to have any detrimental impact on the otter population in the Glengarriff Harbour and Woodland SAC and throughout the Bantry Bay catchment. Note:

• Given the terms of the oral hearing report, it is indeed surprising that Dr Saunders does not cite and explicitly apply the concepts of AA and AA Screening. He also cites the Habitats Regulations 1997 which had been long-since replaced by the Habitats Regulations 2011.

• That said, his opinion is in terms consistent with the level of scientific certainty required in AA Screening to justify screening out AA of risk to otters.

1 February Common Seal Impact Assessment (Coram – desk-top study) 2018 • This desk-top study was done by Alex Coram of St Andrews Marine Research in Scotland.

• He concludes on the basis of scientific evidence that the Salmon Farm is unlikely to negatively impact the conservation status of seals in the Glengarriff Harbour and Woodland SAC. Note: I will return to a more detailed consideration of this report in due course.

5 February Bird Impact Assessment (Gittings – desk-top study) 2018 Under the heading “AA requirements” Dr Gittings states:

• The EIS and EIA are inadequate, … further AA screening is required.

• The present briefing note largely contains the information required for this screening.

• AA may be required of the potential impact of mortalities579 on the Gannet SCI of the Bull and the Cow Rocks SPA.

Note

• While he could not rule out that the Site is within the likely core foraging range of the Storm Petrel of the Bull and the Cow Rocks SPA,580 I note Dr Gittings’ view that AA might be needed was limited to the Gannets of the Bull and the Cow Rocks SPA.

• I will return to the issue of AA as to birds in due course.

28 March Marine Institute AA Screening 2018 • This report was informed, inter alia, by the Gittings report.

• It concluded that AA was not required as to the Gannets of the Bull and the Cow Rocks SPA or as to any other birds.

579 By entanglement in cage nets.

580 The Site is around 45km from the Cow Rock. He cites an academic paper which gives a maximum foraging range of > 65 km but gives no information

on mean foraging ranges. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Date Event

16 NPWS Advice581 November • NPWS recommended, as a safer option, AA of the risk to gannets of the Bull & Cow Rocks 2018 SPA of mortality by entanglement in salmon farm netting.

• It had no comment on the Coram Seal report or the Saunders Otter report.

Note: This document also included advice, not reproduced here, for EIA purposes as to the FwPM.

11 ALAB decided on further AA Screening as to risk to gannets of the Bull & Cow Rocks SPA.

December 2018

6 March Marine Institute Screening Matrix for Aquaculture in Outer Bantry Bay 2019 • It briefly lists and maps all aquaculture licences and aquaculture licence applications in the bay.582 It lists the relevant Natura 2000 Sites and maps those nearby.583

• In the somewhat presumptively-entitled “Finding of no significance effect report” it states that “The plan is to licence the shellfish and fishfish culture activity in Bantry Bay”.

• “It is concluded that the culture of shellfish and finfish, as it is currently constituted and proposed, in Bantry Bay does not pose significant risk to the conservation features of the adjacent sites and as such does not require a full appropriate assessment. On the basis of the above it is considered that there will be no significant effects on the qualifying interests’ of the Natura 2000 sites.”

Note: I will return to this matrix below.

April 2019 AA Screening Report – Birds (Crowe – desk-top study)

• Dr Crowe considered the preceding materials and concluded that AA was required as to the o Fulmar – as an SCI of the Beara Peninsula SPA, Iveragh Peninsula SPA, Deenish Island and Scariff Island SPA o Gannet – as an SCI of The Bull & Cow Rocks SPA and Skelligs SPA, o Guillemot – as an SCI of the Iveragh Peninsula SPA.

April – June ALAB 2019 • Noted the conflicting advice it had received as to whether AA was required.

• Accepted the April 2019 advice of Dr Crowe.

• Decided accordingly on AA as to birds – it specified Species of Conservation Interest (“SCIs”)584 associated with identified Special Protection Areas (“SPAs”).585

581 In response to an ALAB s.47 request dated 3 October 2018 seeking its advice as responsible authority for Natura site designation and species protection.

Oysters (10) clams (1) abalone (1), sea urchins (1), mussels (8), and finfish (4).

Additionally, applications have been received for the following species – oysters (12), scallops (5), clams (1), sea urchins (2), mussels (13), kelp/seaweed

(1) and finfish (1).

583 Sheeps Head cSAC, the Glengarriff Harbour and Woodland SAC and two SPAs, Beara Peninsula SPA and Sheeps Head to Toe Head SPA.

584 Within the meaning of the Wild Birds Directive.

585 Special Protection Area within the meaning of the Wild Birds Directive. The specific SCIs and Natura sites of concern are Fulmar (Beara Peninsula SPA,

lveragh Peninsula SPA, Deenish Island and Scariff Island SPA), Gannet (The Bull and The Cow Rocks SPA and Skelligs SPA) and Guillemot (lveragh Peninsula

SPA). SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Date Event

• Notified MOWI that it had decided that an AA was required and required an NIS by 21 September 2019. That deadline was later extended to 31 October 2019.

18 October MOWI submitted an NIS to ALAB 2019 • It related to the 3 bird species and associated Natura Sites for which Dr Crowe had advised AA. Note: this NIS has never been published and is not before me.

330. ALAB had retained MERC586 – whose team included two marine ecologists and a professional ornithologist specialising in seabirds and coastal species – to draft a report intended to constitute the AA if approved by ALAB. This approach incorporated the necessary expert assessment of all relevant information – including independent expert assessment of the developer’s NIS – with ALAB’s status as competent authority.

331. MERC orally presented their draft report on MOWI’s NIS of 18 October 2019 to a meeting of ALAB on

31 January 2020. Dr Saunders was present. I have not seen MERC’s draft report, but the ALAB minute records MERC’s conclusion that there were no significant lacunae in the assessment, that risks to SCI species had been identified and appraised and that the Salmon Farm would not adversely impact on SCI species or SPA conservation objectives. ALAB considered the draft, expressed itself satisfied to rely on it and asked MERC to issue a final report, clarifying some points. I pause here to note that by this point ALAB had, properly, advanced significantly towards a decision that AA did not require refusal of a licence. As will be seen, however, they had not closed their mind in that regard. What followed must be considered in that light.

332. There follows the content of the minute of 31 January 2020 to which IFI objects:

“It was however noted that the NIS did require further elaboration and the Board was advised by MERC that the NIS should include a firm and explicit statement in relation to the potential impact of the project on the conservation objectives of connected SPA's, as determined by the NIS. MERC advised that while in the concluding statements it is implied the project will not have a significant adverse impact on the conservation objectives of any connected SPA based on the evidence presented, the NIS stops short of making a firm and explicit statement which is unequivocal, in relation to the potential impact of the project on the conservation objectives of connected SPA's, as determined by the NIS.

586 MERC Consultants Ltd.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

The Board agreed it would issue a section 47 Notice to the Applicant, with the draft text to be agreed between Graham Saunders and MERC.”

333. IFI is troubled by the ALAB/Saunders correspondence from 25 February, 2020587 as, in its view,

redolent of ALAB’s securing, by “management of expertise” and “managing the content of its “independent”

reports”, confirmation for a conclusion ALAB had prejudged. It cites an e-mail dated 25 February 2020 from Dr Saunders to ALAB advising that ALAB seek from MOWI a “full and revised NIS”, rather than a supplementary NIS, “Given that we only need them to strengthen their assertions of "no adverse impact" at selected locations in the document”. Out of context these words can be viewed as unfortunate. But they must be taken in context. Context includes the nature of the document in question – Murtagh588 – and these are words in an e-mail, not in a statute. Looseness of language in e-mails is inveterate – that does not excuse all such language, but e-mails are often not formal documents and this e-mail was not.

• First, one may observe that the words “strengthen their assertions” imply that the assertions of no adverse impact already exist in the NIS.

• Second, that view is consistent with MERC’s analysis, reflected in the ALAB meeting minute of 31 January 2020, to the effect that these assertions were implicit in the NIS, had been inferred from it by MERC and informed MERC’s own expert advice that there were no significant lacunae in the assessment, that risks to SCI species had been identified and appraised and that the Salmon Farm would not adversely impact on SCI species or conservation objectives for connected SPA sites. That view is supported by the fact that the e-mail of 25 February 2020 from Dr Saunders to ALAB is in response to a query by the ALAB secretariat earlier that day as to the terms of the intended letter to MOWI. The ALAB secretariat said, “What I need is confirmation that this letter is adequate to address the issue that was identified at the last board meeting.”589

• Third, as will be seen, the resultant s.47 Notice to MOWI did not make any request of them in the terms used in the e-mail of 25 February 2020 or in like terms.

334. That s.47 Notice to MOWI, dated 23 March 2020, cited its regulatory power to seek additional information for AA Screening purposes,590 cited also NPWS guidance as to the proper content of an NIS and thereafter read as follows:

587 pages 725-730 of Exhibit MCC2.

588 Desmond Murtagh Construction Ltd v Hannan [2014] IESC 52, McKechnie J, §61 – “The approach therefore is to have regard to the nature of the

document in question and to consider the words used, by reference to the context in which they are stated.”

589 Emphasis added.

590 Pursuant to Regulation 42(3)(b) of the European Communities (Birds and Natural Habitats) Regulations 2011, as amended , the Board may give notice

in writing to the applicant, directing him or her to furnish any additional information it considers necessary for the purposes of Regulation 42 ('Screening

for Appropriate Assessment and Appropriate Assessment of implications for European Sites'). SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“The Board notes that the NIS did not make clear reference to the conservation objectives (COs) of the Natura 2000 sites under consideration and did not provide satisfactorily explicit statements in respect of a determination of whether the project would have an adverse impact on the COs of any site, taking into account cumulative effects. While the Board is aware that site-specific COs have not been established for any of the SPA sites connected to Bantry Bay at this time, it would, however, expect that, as a minimum, the generic COs currently provided by NPWS would provide a basis on which to make an explicit determination.

The Board is therefore seeking a revision of the NIS to address the issue referred to above in compliance with the appropriate guidance.”

335. Pausing to review the sequence to this point, it is notable that the ALAB minute of 31 January 2020 and the s.47 Notice of 23 March 2020 combine to amount to appreciable criticism of whatever NIS MOWI submitted on 18 October 2019. While an implied statement of reasonable scientific certainty of absence of significant adverse impact is technically as valid as an explicit one, the issue is of such importance in AA that it seems careless of a licence applicant, in submitting an expertly-prepared NIS, to rely on inference by the reader on that issue. Indeed, the extent to which judicial reviews in planning and environmental matters revolve around arguments that matters should be inferred which should be, and could easily have been, made express is generally regrettable.

336. Accordingly, I have some, but quite limited, sympathy with IFI’s accusation of “expertise management” – the more polite phrase IFI used for McKechnie J’s concept of “massaging reports”. However, the circumstances seem to me to fall far short of those fatal to the decision impugned in Nurendale. If, as the minute of 31 January 2020 records, MERC considered the meaning of the NIS adequate to justify the inference that it “implied the project will not have a significant adverse impact on the conservation objectives of any connected SPA based on the evidence presented”, then despite MERC’s understandable frustration at being put to infer what should have been made explicit, it might perhaps have been better to leave wellenough alone. Though, as I have not seen the document, I will not second-guess what was done. The salient point is that MERC had taken the view, inter alia on the basis of the NIS in the form in which it was then to hand, that the Salmon Farm would not adversely impact on SCI species or conservation objectives for connected SPA sites. That is the crucial issue for AA purposes.

337. There is a great difference between the council’s purposeful redrafting of reports in Nurendale on the one hand and, on the other, the neutral terms of the s.47 Notice of 23 March 2020 to MOWI. It required that issues be addressed but did not stipulate or suggest – was non-directional as to – what the resulting substance should be. The s.47 Notice tells MOWI of content required of any NIS – of which MOWI ought, in any event, to have been fully aware. All this reflects poorly on MOWI’s competence in preparing the NIS. It also reflects a predisposition as to AA arrived at by MERC and then ALAB – but arrived at properly. I do not think it betokens bias in ALAB in the sense either of prejudgment or premature determination to grant the licence. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

338. MOWI sent ALAB its updated draft NIS on 20 April 2020. This draft has never been published by ALAB either and is not before me. On MERC’s advice591 a further s.47 request to MOWI, dated 8 June 2020, ensued. It sought a further NIS making explicit determinations as to: (a) potential impacts on conservation objectives for each SPA; (b) all potential sources of impact, including in combination effects. In response to inquiry by MOWI, ALAB by letter dated 11 June 2020 clarified its identification of the SPAs and their SCIs as to which it required explicit determinations592 and that “that the NIS is required to only consider those species, but requires the NIS to address the particular conservation objectives for those species”.

339. MOWI sent ALAB its updated NIS on 7 July 2020. This was MOWI’s third attempt to submit a finalised NIS. On 9 July 2020, ALAB decided to send it to MERC for review and on 4 September 2020 ALAB considered the NIS and MERC’s draft report thereon. ALAB’s minute records that it considered MERC’s draft report comprehensive – its only request was that it clarify some references to make it clear that it was ALAB itself which was doing the AA.

340. As it falls in the chronological sequence I mention that on 3 September 2020 the Marine Institute published an updated Screening Matrix for Aquaculture in outer Bantry Bay593 to replace that of March 2019. Perhaps as it only briefly predated the MERC AA Report, the version of March 2019 was appended to the MERC AA Report of 11 September 2020. But ALAB considered the September 2000 version in its AA

Conclusion of May 2021. My attention has not been drawn to any relevant differences.

MERC’s AA Report, 11 September 2020 et seq

341. MERC’s report on the NIS and AA was finalised on 11 September 2020. It runs to 68 pages.594 It:

• grounds its method in, inter alia, Article 6(3) and (4) of the Habitats Directive and the Habitats Regulations 2011,595 in relevant EU and Irish Guidance as to AA596 and in the principle of precautionary, objective, expert assessment.

591 See ALAB minute 15 May 2020.

592 The list was as follows: Beara Peninsula SPA – FULMAR; Iveragh Peninsula SPA – FULMAR, GUILLEMOT; Deenish and Scarriff Islands SPA – FULMAR;

Bull and Cow Rocks SPA -GANNET; Puffin Island SPA – FULMAR; Skelligs SPA — GANNET, FULMAR, GUILLEMOT.

593 FD1/Schedule of Documents/2.36.

Including Bibliography and Appendices. The narrative runs to 46 pages.

595 European Communities (Birds and Natural Habitats) Regulations 2011 (S.I. No. 477 of 2011).

596 Managing Natura 2000 sites: The provisions of Article 6 of the ‘Habitats’ Directive 92/43/EEC. European Commission 2018.

• Assessment of plans and projects significantly affecting Natura 2000 sites; Methodological Guidance on the provisions of Articles 6(3) and (4) of the

Habits Directive 92/43/EEC. European Commission, 2002;

• Department of Environment, Heritage and Local Government Circular Letter PD 2/07 and NPWS 1/07, 2007; Circular Letter 1/08 and NPWS 1/08,

February 2008; Circular Letter L8/08, September, 2008;

• Appropriate Assessment of Plans and Projects in Ireland, Guidance for Planning Authorities. DoEHLG, 2009. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• summarises the NIS content, including its general correlation with the MI Matrix and recognition of data deficits and concludes overall that it provides a sufficiently deep analysis and evaluation of and objective and clear conclusions as to risks, that the Salmon Farm will not impact the species in question.

• then proceeds to MERC’s own, explicit, analysis of these issues to its own conclusion.

342. It suffices here to summarise its conclusions, recommendations and assessment outcome.597 They included:

• An acknowledgement that AA required expert, objective and precautionary interpretation of risks to SCI species.

• Findings that:

o the risks to SCI species had been identified and appraised.

o though greater transparency as to levels of impact could be demonstrated by collecting additional specific data in relation to interactions between aquaculture and seabird species, the available information and data were adequate to support AA and there were no significant lacunae.

• A reasoned conclusion that the Salmon Farm will not impact adversely on SCI species or SPA conservation objectives.

• Recommendations that licence conditions should require, o mitigations to prevent seabirds from predating on salmon smolts and becoming entangled in poorly maintained equipment.

o open-access reporting of interactions between cage aquaculture and wildlife (SCI and protected species) and mortality events.

• A recommendation that Single Bay Management be required for Bantry Bay aquaculture.598

343. Notably in the context of alleged objective bias, ALAB decided on 4 September 2020 “that in the particular circumstances of these appeals it is appropriate in the interests of justice” to request the parties who made submissions or observations, to make submissions or observations as to the NIS and the MERC AA report, prior to ALAB’s concluding its AA and to issue s.46 notices accordingly. So, ALAB published the NIS of July 2020 and the MERC AA Report of 11 September 2020 and, by s.46 Notices dated 23 September 2020 to all parties to the Appeals, invited submissions and observations thereon. ALAB also published newspaper notices announcing the AA and the publication of the NIS and the MERC AA Report and inviting submissions by the public. It identified the specific SCIs and SPAs of concern as the Fulmar (Beara Peninsula SPA, Iveragh Peninsula SPA, Deenish Island and Scariff Island SPA); Gannet (The Bull and The Cow Rocks SPA and Skelligs SPA) and Guillemot (Iveragh Peninsula SPA). ALAB received 5 submissions on AA issues in November 2000.599 An Taisce, Save Bantry Bay, Galway Bay against Salmon Cages and Friends of the Irish Environment expressed concerns as to alleged lacunae in, and the adequacy and conclusions of, the MERC AA Report.

597 §§8 & 9.

598 to manage overall impacts from aquaculture and ensure that development and production is managed and co-ordinated in order to mitigate against

adverse effects. The requirement should be implemented and supported by relevant equipment.

599 From Save Bantry Bay, Friends of the Irish Environment, Galway Bay Against Salmon Cages, Department of Agriculture Food and Marine, and An Taisce.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

344. The Technical Advisor’s Final Report (Saunders) dated 8 December 2020 gives an account of the AA process and agrees that the Salmon Farm will have no adverse effect. It does not substantively add to the AA process.

345. ALAB met on 10 December 2020, 12 January 2021 and 5 February 2021. In the context of an allegation of objective bias, it bears observing that it was now in possession of what it deemed a satisfactory NIS, an expert report supporting a satisfactory AA and submissions on both following a public participation process. A cynic might infer that ALAB was well-equipped to determine the AA in accordance with any prejudice it may have held in favour of the project. Had ALAB’s mind been closed to anything but to grant

the licence or if it was, as is alleged, “bending over backwards” to do so or, as is also alleged “groomed” the

application to succeed, this was the point at which one would have expected ALAB to end the AA process, the presumptively desired objective having been achieved. In my view, the reasonable fair-minded observer would have been considerably reassured by what in fact ensued.

346. In fact, far from even apparently pursuing a predetermination to grant the licence at that point, ALAB minuted600 its duty to be “meticulous” and decided that, before a conclusion could be reached, further review was required of potential effects (i) on gannets (mortality by entanglement with salmon farm nets),

(ii) on SCI birds in combination with kelp harvesting, (iii) on FwPMs in the Trafrask River, (iv) on non-SPA birds. It asked Dr O’Toole to investigate. It also took considerable care as to drafting a further s.47 Notice to MOWI.601 It issued on 15 February 2021 and made searching inquiry as to bird injury and mortality data at other MOWI sites and the reliability of such data. MOWI replied on 2 March 2021 – inter alia stating that total bird entanglement mortality in MOWI sites was 1 herring gull. The data is audited and available on the Aquaculture Stewardship Council website.

347. While I will address it later when considering EIA issues, I should note that the Sweetman Applicants

plead breach of Articles 3 and Articles 5 to 11 in both EIA Directives in that Dr O’Toole’s two reports to ALAB,

dated 28 May 2021, were not circulated for response before ALAB made its Impugned Decision. Those reports were as to AA issues relating to any risk of in-combination effects of the Salmon Farm with Kelp harvesting in Bantry Bay on bird mortality and as to EIA issues regarding any risk of lice-induced salmonid mortality inhibiting FwPM recruitment by diminishing the numbers of available larva hosts. IFI did not plead the issue but made submissions to similar effect. The Sweetman Applicants pleaded a similar point as to the non-circulation of the DMP Statistical Solutions UK Ltd report602 enclosed to ALAB with MERC’s supplemental briefing note of 19 May 2021. I mention these matters here as part of the narrative leading to ALAB’s AA

600 ALAB minutes 10 December 2020.

601 ALAB minutes 12 January 2021 and 5 February 2021.

602 Dated 19 May 2021 and entitled Population Viability Analysis of The Impacts of Additional Mortality Due to Fish Net Entanglement in Gannets from

Bull and Cow Rocks SPA. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Decision. As events internal to the Impugned Decision they do not contribute to any allegation of objective bias.

ALAB to MERC, 15 February 2021 & MERC Reply, 19 May 2021

348. By letter dated 15 February 2021, ALAB had sent the submissions on AA issues to MERC to consider. ALAB’s requirement of MERC bears setting out as indicating that its mind was not closed at this point. MERC was asked to address:

• all points raised in the submissions as to bird entanglement and mortality and any impact on SPA species.

• all points raised by the submissions as to lacunae/gaps in MERC’s AA report of September 2020.

• how MERC formed its conclusions in that report relating to lacunae/gaps (it had found that there were none). MERC was asked to address the foregoing using the best available scientific evidence and expertise (Thus invoking the requirements for performing AA).

349. From this referral to MERC came its recommendation of, and ALAB’s commissioning of, a specialist statistical Gannet Population Viability Analysis to address concerns as to potential gaps in data as to bird mortality due to fish farms. MERC replied to ALAB by Supplemental Briefing Note dated 19 May 2021. It included the DMP Gannet Population Viability Analysis.603 That briefing note was explicitly based on what it asserted to be “best available scientific evidence and expertise” “sufficient to remove any reasonable scientific doubt concerning impacts”. It sequentially addressed each of the concerns raised in submissions.604 Inter alia, it recorded that gannet populations in nearby SPAs are increasing, suggesting that any (if any) mortality at existing fish farms is not having a population level effect. For example, as to risk of gannet mortality by entanglement with fish farm apparatus, MERC advised that,

• such events are highly likely to be rare – at worst occasional – and population level effects are highly unlikely.

• the conclusion that significant mortality is unlikely to arise from the proposed salmon farm is supported by sufficient scientific and empirical evidence, which is considered sufficient to remove any reasonable scientific doubt concerning impacts.

350. MERC attended ALAB’s meeting of April 2021 MERC orally presented and ALAB considered a draft MERC briefing note and the draft Statistical Report.605 MERC confirmed that their AA Report does not suggest that fatal/injurious interactions between gannets and salmon cages never occurs, but does conclude, on the basis of reasonable scientific and empirical evidence, that it was highly likely to be rare (< 10

603 “Population Viability Analysis Of The Impacts Of Additional Mortality Due To Fish Net Entanglement In Gannets From Bull And Cow Rocks SPA”, DMP

Statistical Solutions UK Ltd 19 May 2021.

604 By An Taisce, Save Bantry Bay, Galway Bay Against Salmon Cages, Friends of the Irish Environment.

605 By DMP Statistical Solutions UK Ltd. of Scotland.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Gannets)606 and the level of mortality will not adversely affect the Bull and Cow Rock SPA gannet population. ALAB’s minute records consideration of each objector’s submission on these issues. ALAB agreed that the draft reports were comprehensive and definitive and asked that they be finalised.

351. MERC’s Supplementary Briefing Note to ALAB dated 19 May 2021,

• was explicitly based, in MERC’s view, on “best available scientific evidence and expertise”.

• responded to issues raised by stakeholder submissions in AA. In particular it summarises and responds in some detail to the submissions by An Taisce, Save Bantry Bay, Galway Bay against Salmon Cages and Friends of the Irish Environment.

• overall, it is fair to say, confirms the conclusions of the MERC AA Report of 11 September 2020.

• appends the DMP Statistical Report.

352. The DMP Statistical Report notes, inter alia, that

• Gannets are long-lived and small increases in annual mortality could lead to significant population decline.

Yet the Bull Rock Gannet colony has more than doubled during the operation of existing fish farms in Bantry Bay.

• Gannet mortalities due to fish net entanglement are rare.

• it takes a precautionary approach in assigning all modelled gannet mortalities to the SPA colony – which is unlikely.

• it considers its analysis robust.

• 10 additional adult deaths due to entanglement may cause an impact of low significance to population numbers (0.99% drop from unimpacted levels).

353. MERC’s Supplemental Briefing Note dated 19 May 2021 explicitly invoked the legal standard required of conclusions in AA and, more importantly as to allegations of bias, ALAB’s request for it derived from active interrogation of the issues it described – even though a positive decision on AA issues was already in prospect. To put it another way, it seems clear to me that ALAB was keeping its mind open in a manner inconsistent with objective bias, even leaving aside the factor that all this was internal to the process.

MI AA Screening Matrices for Aquaculture in Outer Bantry Bay

354. SWI plead that the Marine Institute AA Screening Matrices indicated that the proposed fish farm was

not likely to have any significant effects on any protected site – “this in spite of the fact that the decision to

require an NIS involved a finding that significant effects were likely.” However, this narrative is not pleaded as connected to any particular legal ground of challenge. That screening matrix is mentioned in MOWI

606 In fact the minute says that mortality greater than 10 gannets was highly unlikely.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

submissions but to no particular effect and it did not feature in oral submissions. Nonetheless, these matrices are somewhat mysterious and, as they are cited in ALAB’s Impugned Decision and are pleaded, and as on one view they informed ALAB’s decision as to risk to seals, it is necessary to say a little of them.

355. In May607 or June608 2018, the Marine Institute issued an undated “AA Screening Matrix for Aquaculture in Outer Bantry Bay”.609 The Marine Institute updated it in March 2019610 and September 2020.611 Though it wasn’t much addressed inter partes, it is necessary to take some view of the legal status and meaning of these matrices. Their legal status seems to me quite unclear.

356. To 2021 the EU Commission envisaged such matrices – not to be the AA Screening itself but rather to summarise completed AA screening assessments of plans and projects for which development consent was sought.612 A matrix could include, if appropriate depending on the outcome of AA screening, a “Finding of no Significant Effect Report”. The Marine Institute matrices contain content so entitled. However, and confusingly given the Commission’s Methodological Guidance of 2002, it seems that the Marine Institute issues such matrices for a quite different purpose.

357. At least ordinarily, such a matrix, as envisaged by the Commission, will consider one specific plan or

project. While they differ in detail, it suffices to set out the “Brief description of the project or plan” identified

by the Marine Institute in its March 2019 and September 2020 versions:

Marine Institute Screening Matrices for Aquaculture activities in outer Bantry Bay, Co. Cork – Extracts Brief description of March The following species are cultured in outer Bantry Bay (number of the project or plan 2019 licences in parenthesis) – oysters (10) clams (1) abalone (1), sea urchins (1), mussels (8), and finfish (4).

Additionally, applications have been received for the following species

– oysters (12), scallops (5), clams (1), sea urchins (2), mussels (13), kelp/seaweed (1) and finfish (1).

The locations of the sites are shown in Figure 1.

September There are currently 34 licenced aquaculture operations in outer 2020 Bantry Bay with a further 2 applications.

607 MOWI’s NIS July 2020 §4.2.

ALAB Determination 29 June 2021 §2.36.

609 See MOWI NIS dated July 2020 §4.2 & Appendix 1.

610 See MERC’s AA Report of September 2020, Appendix 2.

611 ALAB Determination 29 June 2021 §2.36.

612 Assessment of plans and projects significantly affecting Natura 2000 sites Methodological guidance on the provisions of Article 6(3) and (4) of the

Habitats Directive 92/43/EEC European Communities, 2002 §§2.3, 3.1.5, 3.1.6 & Figures 1 & 2. This guidance was updated in 2021 and the use of screening

matrices was omitted.

613 See MERC’s AA Report of September 2020 Appendix 2.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Marine Institute Screening Matrices for Aquaculture activities in outer Bantry Bay, Co. Cork – Extracts The following species are cultured in outer Bantry Bay (number of licences in parenthesis) – oysters (7), mussels (13), urchins and mussels (1), seaweed (6) and finfish (7).

Additionally, applications are being considered for the following species – mussels (2).

It should be noted that of those sites that are currently considered licenced 4 (2 mussel; 1 salmon and 2 Seaweed) are currently under appeal to the Aquaculture Licence Appeals Board.

The locations of the sites (and licence status) are shown in Figure 1.

I confess that I cannot interpret the “descriptions” above as describing a “project or plan”. Also, in each of

the 2019 and 2020 versions a different description of the project or plan is given in the “Finding of no significance effect report”614 which forms part of the matrix.

358. MOWI’s NIS of July 2020615 cites an undated version of the matrix as dated May 2018 on the footing that “Such matrices have been developed for the majority if not all aquaculture areas in Ireland and are used as a basis for decisions on the granting of licences for shellfish and seaweed licence applications.” MERC’s AA Report of September 2020 cites the March 2019 version and states “AA Screening matrices have been developed by the Marine Institute for all major aquaculture areas in Ireland. The matrices are used as a

decision support tool when DAFM616 are evaluating new aquaculture license applications as well as applications for renewal of existing licenses.” If these rather vague descriptions are correct, and I have no evidence to the contrary, the Marine Institute matrices are not AA Screening Matrices in the sense envisaged by the EU Commission in that they are intended to inform (in some way which is unclear) the conduct of AA screenings rather than summarise the outcome of an AA screening.

359. In any event, if they are intended to summarise the outcome of an AA screening, that AA screening is not before me and there is no suggestion that it was before ALAB. And it was ALAB, not the Marine Institute, which bore the obligation of performing AA screening for purposes of the Appeal.

360. This important difference of purpose and legal status is not apparent on the face of the Marine Institute matrices. They appear to assume that the title “Screening Matrix” is self-explanatory. Neither cites the Commission’s methodological guidelines or any legal instrument. Neither states, or refers to any document stating, how it will function as a decision support tool in AA of specific plans or projects. Neither refers back to or directs the reader to the AA screening of which it may be a summary. They encompass mostly projects already in being. Indeed that of September 2020 encompasses no less than 34 licensed sites

614 Sic.

615 §4.2.

616 Department of Agriculture Food and Marine.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

(including the Salmon Farm617) and a further 2 licence applications. Nonetheless, despite their seeming function as decision support tools, they purport to record AA screening conclusions that AA was not required. The matrices are not susceptible to interrogation as AA screenings.

361. All in all, they are remarkably Delphic documents as to their intended or actual legal and practical purpose and effect. As ever, documents should, as far as practicable, be complete and self-explanatory on their face. If needs be, they should explicitly cross-refer to other documents which explain them – as otherwise they are at risk of interpretation other than in the context of those documents.

362. In the project descriptions in the March 2019618 and September 2020 Matrices, existing and proposed aquaculture activities were described – in the September 2020 Matrix they related to oysters, seaweed, mussels, urchins and finfish (Salmon). The relevant Natura 2000 Sites were described. The assessment of potential effects was recorded in a “Finding of No Significant Effects report”. The following, inter alia appears:

Marine Institute Screening Matrices for Aquaculture activities in outer Bantry Bay, Co. Cork – Extracts Finding of no significance619 effect report:

Description of the March 2019 Aquaculture activities in outer Bantry Bay, Co. Cork.

project or plan September The plan is to licence the620 shellfish and fishfish culture activity in 2020 Bantry Bay, Co. Cork.

The activities in question cover approx. 418 ha. in total, representing approximately 0.9% of the surface area of Bantry Bay.

363. Again, the descriptions are quite unclear as to what it is proposed to licence – not least as between existing and contemplated activities. In any event, there can in law be no plan “to licence” aquaculture. There could at most be a “plan” to consider applications for aquaculture licences and decide them, by way of grant or refusal, in accordance with law. I am not convinced that merely anticipating the performance of a statutory duty to decide aquaculture licence applications in accordance with law is a “plan” within the meaning of the Habitats Directive which could be the subject of AA Screening or AA. Such a view could, for example, suggest a requirement of AA Screening of An Board Pleanála’s “plan” to decide the planning appeals which it expects to come before it. I cannot see any basis in law for such a suggestion or any environmental purpose it would serve given the necessity of AA Screening or AA, where appropriate, of each project for which development consent/approval is sought.

617 In the sense that the licence had been granted by the Minister but was under appeal.

618 See MERC’s AA Report of September 2020 Appendix 2.

619 Sic.

620 Emphasis added.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

364. It suffices to continue the account by reference to the September 2020 version of the matrix.

AA Screening Matrix for Aquaculture in Outer Bantry Bay. (September 2020) (Extract)

Finding of no significance621 effect report:

Describe how the project or The cultivation of shellfish, finfish and macroalgae in outer Bantry Bay is plan (alone or in not likely to affect the conservation features of adjoining Natura 2000 sites.

combination) is likely to affect the Natura 2000 site.

Explain why these effects There is no spatial overlap of the aquaculture activities with Natura sites. In are not considered addition, there would be no interference with key relationships that define significant. the function of the sites. The culture activities will not result in habitat loss, there will not be significant disturbance to key species and there will be no habitat or species fragmentation. There will be no direct discharge of pollutants into the environment during the works and water quality will not be affected. Consequently, it is concluded that the culture of shellfish and finfish, as it is currently constituted and proposed, in Bantry Bay does not pose significant risk to the conservation features of the adjacent sites and as such does not require a full appropriate assessment.

On the basis of the above it is considered that there will be no significant effects on the qualifying interests of the Natura 2000 sites.

365. As will be seen and as is proper to a matrix, the foregoing purports to be a summary of conclusions of an AA Screening. But it is not the AA Screening itself nor does it demonstrate (as opposed to assert) the verifiable scientific certainty required in AA Screening. As was said by Kokott AG in Eco Advocacy,622 in deciding in AA Screening that AA is unnecessary, a competent authority must provide reasons capable of removing all reasonable scientific doubt concerning the harmful effects of the works envisaged on the integrity of the protected site concerned.623 Competent authorities may authorise an activity only if they have made certain that it will not adversely affect the integrity of a European site. That is so where there is no reasonable scientific doubt as to the absence of such effects (Holohan624 and Electrabel625).

621 Sic.

622 Case C-721/21, Eco Advocacy CLG v An Bord Pleanála, Opinion of Kokott AG of 19 January 2023 §§90 & 109(5) & (6). This view was essentially adopted

by the CJEU in its judgment of 15 June 2023.

623 See also Case C-521/12, Briels, §27.

624 Case C‑461/17, EU:C:2018:883, judgment of 7 November 2018, §33 and case-law cited;

625 Case C‑411/17, Inter-Environnement Wallonnie ASBL & Electrabel SA, judgment of 22 June 2019, §120.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

366. Given the lack of clarity as to their status on the face of the MI Matrices, MERC’s recital of its understanding of their function, as an aid to decision-making, is notable. Equally notable however is that MERC, in its AA Report of September 2020 did not simply rely on the AA screening conclusion recorded in the matrix but embarked, in my view properly and prudently, on its own screening evaluation. MERC’s Supplemental Briefing Note dated 19 May 2021 made clear – though it was already apparent – that the matrices were amongst the evidence base for their report. But the Report and Note were based on “a subsequent detailed appraisal and validation of the evidence presented”.626

367. It is important to note also the limited role the MI Matrices played in ALAB’s determination.627 ALAB accepted their findings only as to the relevant SACs and then only as they coincided with the findings of ALAB’s own Seal Screening and Otter Screening Reports. As to the SCI Bird species of relevant SPAs, ALAB rejected the Matrices’ finding that AA was unnecessary in favour of Dr Crowe’s advice that it was necessary as to 3 such species. In exposing MOWI’s application to the rigours of AA, instead of taking the Marine Institute advice that AA was not needed, ALAB’s chosen course is not redolent of bending over backwards to grant the licence.

368. None of the foregoing should be regarded as a definitive finding by me as to the substance of any AA screenings by the Marine Institute of which the matrices are summaries. It merely reflects the evidence before me and, it seems, the material before MERC and ALAB. It may be that in another case a fuller account will be available of the substance underlying, and the resultant legal status of, such MI Matrices.

369. ALAB met three times from February 2021 to review progress and materials as they came in. At its meeting of 28 May 2021 it adopted its AA Conclusion Statement, setting its analysis out at some length, to the effect that the Salmon Farm,

• was not likely to have a significant effect on the integrity of conservation objectives of the relevant Natura 2000 Sites.628

• had no potential for significant effects (on) and was not likely to have an adverse effect on the SCI species of those Natura 2000 Sites.

370. So, by the time it came to make its AA Conclusion Statement of May 2021,629 (by which time it had the updated MI Matrix of September 2000), it was clear that ALAB were proceeding on the footing that the MI Matrices were, as MERC said, intended by the MI to be “used as a decision support tool when DAFM are evaluating new aquaculture license applications” and that, while they informed MERC’s work, they did so

626 Report p46, Note pp9 & 10.

627 §7.4.4.

628 Beara Peninsula SPA (Site code 004155), Iveragh Peninsula SPA (Site code 004154), Deenish Island and Scariff Island SPA (Site code 004175), The Bull

and The Cow Rocks SPA (Site code 004066), Skelligs SPA (Site code 004007), Sheep’s Head to Toe Head SPA (Site code 004156). Glengarriff Harbour and

Woodland SAC (Site Code: 00090), Sheep’s Head SAC (Site code: 000102) and Puffin Island SPA (Site code: 004003).

629 See below.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

only as part of a wider evidence base, all of which MERC had subjected to “detailed appraisal and validation”. ALAB’s acceptance of the conclusions of the matrices was limited to impact on only two adjacent terrestrial SACs630 and to no other relevant Natura 2000 Sites. It must be understood in that light and in light of the extensive other environmental information before it.

371. Even as to Glengarriff Harbour and Woodland SAC, the MI Matrix conclusion that “There is no evidence in the scientific literature to suggest that aquaculture activities impact on seal species” was not, in substance, merely accepted by ALAB. That is clear as ALAB had acted on the recommendation of Prof McIntyre, as chair of the oral hearing, that ALAB to consider AA screening for effect on the seals of that SAC and had commissioned a Seal Impact Assessment of February 2018 which ALAB’s AA Conclusion Statement of May 2021 cited as having found “no evidence of significant negative impacts predicted from the proposed development” – which conclusion it accepted.

372. While it seemed to me contextually necessary to explain my understanding of the role and significance of the matrices and was convenient to do so in considering the AA process at this point, given the paucity of the pleadings as to these matrices, I need take my consideration of them no further.

ALAB’s AA Conclusion, 28 May 2021 & Determination

373. ALAB’s AA Conclusion dated 28 May 2021 is a distinct document of 13 pages631 and is a formal and detailed record of AA too lengthy to record in detail here. It recites much of the content noted above. The following are notable:

• It adopts the findings of the September 2000 version of the Marine Institute Matrix – but only as to screening out the two relevant SACs632 – citing also the Otter and Seal reports. It does not accept the Marine Institute Matrix in its screening out of SPAs.

• In light of the absence of specific, as opposed to general, conservation objectives for SPAs, it explicitly invokes the precautionary principle both generally and in its listing633 of additional conservation objectives reasonably applicable.

• It explicitly found the data and information to hand adequate to support the AA.

• It noted that while Lesser Black-backed gulls, Storm Petrel, and Puffin, could potentially overlap with the proposed Salmon Farm there was no potential for significant effects. All SCI species (they are listed) were screened out other than the three identified by Dr Crowe as meriting AA.634

630 Glengarriff Harbour and Woodland SAC, Sheep’s Head SAC.

631 Plus 1.5 pages listing references.

632 Glengarriff Harbour and Woodland SAC and Sheep’s Head SAC.

633 Table 1: Specific conservation objectives for connected SPA species.

634 Fulmar – as an SCI of the Beara Peninsula SPA, Iveragh Peninsula SPA, Deenish Island and Scariff Island SPA; Gannet – as an SCI of The Bull & Cow Rocks

SPA and Skelligs SPA) Guillemot – as an SCI of the Iveragh Peninsula SPA). SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Of those three species, an individual and reasoned appreciation of and conclusion on each follows, including an appreciation of risk of effects of the Salmon Farm in-combination with other listed activities.635

• Specifically as to the Gannet, ALAB considered, inter alia, that the statistical modelling, demonstrated the high level of Gannet mortality that would have to occur as a result of entanglement before a significant negative effect was seen on Gannet populations at the Bull and Cow Rock SPA. This was far higher the suggested level of mortality for entanglement that may have caused a significantly negative impact on Gannet populations.636

• Therefore, o The potential impact of Gannet mortality due to entanglement was considered to have no potential for significant effects and was not likely to cause a significant negative effect either individually or in combination with other sites, plans or projects. o ALAB concluded that the proposed licensed activity was not likely to have a significant negative effect on the integrity of the conservation objectives of any of the SACs or SPAs.

374. ALAB’s AA is at §§5 and 7.3 of its Impugned Determination and is in terms following those of its AA

Conclusion.

The AA Process – Conclusions as to Bias & Delay

375. As to the allegations in bias, as they relate to the sequence of events as to AA, that an objective observer would reasonably suspect that ALAB was straining to grant the licence despite the deficiencies in MOWI’s application, it appears to me that the foregoing sequence demonstrates rather ALAB’s conscientious anxiety, once it adverted to the need to consider AA, to be meticulous in its conduct of the AA. For example, the objective observer would note that ALAB:

• did not accept the Saunders views that AA was unnecessary.

• did not accept the Marine Institute matrix to the extent that it had purported to screen out AA of the SPAs.

• did not rest on Dr Gitting’s conclusion that only the Gannet needed consideration in AA.

• instead commissioned Dr Crowe’s report and accepted her advice that a wider scope of AA was necessary.

• despite a reassuring NIS and MERC AA Report and after public consultation, identified further AA issues for investigation and requiring a supplemental briefing note from MERC.

• also commissioned, via MERC, the statistical report on Gannet mortality despite MERC’s already having reassured it on this issue.

635 Under headings, Commercial Fishing, Aquaculture, Navigation and Marine Transport, Marine Leisure and Recreation, Mechanical Kelp Harvesting,

Other Activities.

636 Citing Gittings 2018.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

The sequence does not suggest contrivance to grant the licence. It is entirely consistent with a desire in ALAB to reach a correct decision.

376. I consider it unfortunate and undesirable that the “intermediate” versions of the NIS were not published. But ALAB’s pursuit of improved versions speaks primarily to its anxiety to perform a substantively sound AA. And as the final version of the NIS was published for comment I consider that in substance the public participation aims of LZ #2 were met.

377. As to the bias issue and the complaints of failure to circulate the O’Toole report of 25 May 2021 discounting risk to the FwPM and the MERC Supplemental Briefing Note and accompanying Gannet Population Viability Analysis of 19 May 2021, it seems to me that the proper analogy, putting the matter at

its height from the Applicants’ point of view, is with the facts in O’Callaghan637 – in which bias was not found.

Indeed, it seems to me that the failure to circulate the inconsistent witness statements in O’Callaghan was, on any view, of a higher order of error than those alleged failures to circulate here. Even there, the series of errors of law involved, internal to the process (as were the alleged errors here) were held incapable of amounting to objective bias. And, more importantly, these allegations were not pleaded.

378. I therefore reject the allegation of objective bias as it relates to the evolution of the NIS and AA.

Bias – ALAB/Saunders Interactions as to Saunders Final Report 2020

379. IFI asserts that “management of expertise” disclosing bias is seen in ALAB’s e-mail to Dr Saunders on 23rd November, 2020 telling him that his “report would not be deemed complete until the determination is

made”. IFI does not elaborate – it makes the point as if the conclusion of “management of expertise” is self-

evident from that e-mail. I don’t see it. I cannot infer bias from it. In any event, I reject this allegation as not

pleaded.

BIAS & DELAY – FURTHER INFORMATION PROCESSES & NON-CIRCULATION OF REPORTS

Further Information – Introduction

380. As has been seen, s.46 of the 1997 Act empowers ALAB to solicit submissions from a party to the Appeal, or anyone who has made submissions in the Appeal, where ALAB is of the opinion that, in the

637 O’Callaghan v Mahon [2007] IESC 17, [2008] 2 IR 514.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

particular circumstances of the Appeal, it is appropriate in the interests of justice to do so. S.47 of the 1997 Act empowers ALAB to require submission by any person of specified documents, particulars or other information which, in its “opinion” is necessary to enable it to determine an appeal.

381. SWI allege that ALAB made excessive use of these powers such that:

• Determination of the Appeal was unlawfully delayed – in breach of ALAB’s statutory duty of expedition.

• ALAB’s use of those powers created or contributed to a situation of objective bias in the form of a perception that ALAB, instead of taking an independent and impartial approach, was bending over backwards, as it were, to o avoid a refusal of the Aquaculture Licence application. o enable MOWI and its experts to repeatedly remedy deficiencies in its application. o itself procure from others information which MOWI ought to have provided, with a view to remedying deficiencies in MOWI’s application. o enable a grant of the Aquaculture Licence application.

The Sweetman Applicants put the matter similarly – accusing ALAB of ‘patching up’ the MOWI application via repeated s.47 notices to MOWI and others.638

In particular, objection is taken to s.47 Notices issued after the oral hearing.

Further Information – the Law

382. Planning law seems to me to assist. Browne639 states, “It is now well established that once a planning authority has all the information or sufficient information necessary to decide on a planning application, it should not resort” to such procedures. Broadly, I agree. It is clear from planning cases that powers such as those set out in s.46 and s.47 of the 1997 Act may not be exercised to delay making a decision – Forest Fencing640 – or in order to negotiate changes in the proposed development – NCE.641 Browne suggests that the courts allow little discretion or deference to planning authorities as to the legality of requests for further information but rather subject them to what he calls “full blooded”642 review. I confess that I am not clear that the cases cited by Browne lead reliably to that conclusion. Such requests, and the desirability of making them, typically relate to the merits of the proposed development and the decision to make them will

638 Affidavit of Noel Carr, FISSTA National Secretary, sworn 27 September 2021.

639 Simons on Planning Law, 3rd Ed’n (Browne) §3-207 et seq. Many of the cases of alleged invalidity of requests for further information resulted in claims

for default planning permission.

640 Wicklow County Council v Forest Fencing t/a Abwood Homes, [2007] IEHC 242 cited in Simons on Planning Law, 3rd Ed’n (Browne) §3-209.

641 The State (NCE Ltd) v Dublin County Council [1979] I.L.R.M. 249 cited in Simons on Planning Law, 3rd Ed’n (Browne) §3-208.

642 A term used by Browne and which Simons J and I have used in past judgments, but which Humphreys J has recently suggested, in Four Districts

Woodland Habitat Group v An Bord Pleanála [2023] IEHC 335, may not be entirely useful. On reflection, I tend to agree with him.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

typically consist in the exercise of expert judgment by the decision-maker as to the potential of the information likely to be received to contribute to a properly-informed decision. Sometimes of course, as in NCE,643 the very terms of the request for information will disclose an ulterior purpose. But that delay or other ulterior purposes will invalidate such a request does not undermine the proposition that, once its purpose, broadly considered, is valid – as will be presumed absent evidence to the contrary – the courts will accord curial deference to the planning authority as to whether the request may be reasonably regarded as appropriate to that purpose.

383. Browne first cites Murphy v Navan UDC644 – but as an exception to what he sees as the rule and in which, he considers, the court allowed Navan UDC “great leeway”. In Forest Fencing645 the applicant argued that the further information request was for modification of plans,646 as opposed to genuinely seeking further information. Charleton J preferred to apply a “test of genuineness” to any impugned request for further information “rather than to construct what, to my mind, would be a wholly artificial test based on the predominant purpose of the notice or to sift through the notice seeking to discover whether the majority of the queries raised were in respect of modification or seeking further information”. This does not seem redolent of the “full-blooded” review contemplated by Browne. Charleton J held that party seeking to impugn the validity of a request for further information as issued in bad faith for an ulterior motive “bears the burden of proving that this statutory power was not exercised in good faith for the purpose for which it was granted.” Proof of mala fides, while on the balance of probabilities, nonetheless requires proof commensurate with the seriousness of the allegation. As was said in Fyffes647: “ ….. the civil standard is flexible, so that the degree of probability required is proportionate to the nature and gravity of the issue” and “the gravity of an allegation and of the consequences of finding that it has been established are matters which the court must have regard to in applying the civil standard.” In terms of practical proof, it does not seem to me that making the “ugly allegation”648 of objective bias – much lightens this burden.

384. In Conlon Construction649 a posited request for information was held invalid. Butler J said:

“The Council clearly disliked the proposal but equally clearly they understood it and in all its details.

Indeed, it was this understanding which bred their dislike. The letter does not indicate that they lacked any information, explanation or evidence necessary for them to decide upon the merits of the application on the planning grounds ..”

643 State (NCE Limited) v Dublin County Council [1979] ILRM 249.

644 Ó Caoimh J, High Court, Unrep. 31 July 2001.

645 Wicklow County Council v Forest Fencing t/a Abwood Homes [2007] IEHC 242.

646 A purpose as to which Charleton J cited Illium Properties (infra) for the proposition that such requests cannot) be used to vary a planning application.

647 Fyffes v DCC [2009] 2 I.R. 417, §189 et seq.

648 Shatter v Guerin [2021] 2 IR 415, O’Donnell J, §30.

649 State (Conlon Construction Ltd) v Cork County Council Unrep, Butler J. 31 July 1975.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

385. Browne cites NCE650 in which the High Court invalidated a request for further information651 as it was “quite clear … that the Council was not seeking any elucidation from the developers of their plans or that it lacked any information, explanation or evidence necessary for it to decide upon the merits of the application on planning grounds. The letter states the features of the proposal which the Council disliked and asks the developer to alter his proposal to meet the objections.”

386. SWI and Browne cite Illium Properties,652 in which O’Leary J held a request for further information invalid as served to gain more time for the planning authority to do its work and at a time when it already had:

“… enough information to approve the application subject to the same conditions as were ultimately applied but they definitely had enough information to refuse it. The law required them to make a

decision unless they lacked information to make any decision.”

But Illium turned on O’Leary J’s identification of the true and improper purpose of the request as being to delay the decision. And Conlon and NCE seem to have been clear cases of notices served for an improper purpose.

387. The point made by O’Leary J in Illium is that the decision-maker should not seek further information to achieve a particular decision – it must be, ceteris paribus or in principle equally willing to grant or refuse and should do so once it has the information necessary to do either. That flows from its duty of impartiality. However it is not a principle to be taken too far or applied to rigidly: many applications can be refused at any time for the want of the information necessary to grant it. On that strict view, one could argue that decisionmakers should never seek further information. But that argument is untenable given the express statutory powers to do so. And, as I have observed, it is perfectly proper for a deciding authority to come iteratively to a view over time, to take a provisional view subject to testing it by reference to further information. Also, there would seem to be no reason why a deciding authority might not have decided in principle to grant a permission but seek further information to inform its view as to conditions it was considering imposing. Overall, I respectfully adopt Charleton J’s “test of genuineness” as to whether the deciding authority was seeking information with a view to making the correct decision as opposed to a particular decision.

388. Decisions on planning permissions, aquaculture licences and the like are inevitably inquisitorial, multifactorial and iterative. It is in the nature of such decision-making that the decision-maker can make its own inquiries. Hence s.47 permits inquiry not merely of the parties to the process or of prescribed bodies but of “any person”. And recent judgments have emphasised that decision-makers owe duties of

650 State (NCE Limited) v Dublin County Council [1979] ILRM 249.

651 The Supreme Court reversed the High Court on other grounds.

652 Illium Properties Ltd v Dublin City Council [2004] IEHC 327, O’Leary J.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“thoroughly independent and detailed expert scrutiny” – e.g. Balz653 and Jennings.654 It is in the nature of iterative decision-making that a tendency to a particular decision – grant or refusal – may properly evolve over time. Inquiry implies lines of inquiry. Lines of inquiry may tend towards either grant or refusal and may or may not result in the outcome anticipated. As long as an open mind is kept as to the ultimate outcome, there is nothing wrong with a provisional view based on information to hand at a particular point in the process or foreseeing that further information may result in a particular outcome. As McKechnie J said in Nurendale655 “ … an administrative body may have some outcome in mind. That does not amount to prejudgment ..” He contrasted that with a situation in which he found “a rigidity of mind, so that from the start there could have been no other outcome. ……… It was a given from the start.” See also the review of McGrath656 and Lohan657 above when considering bias.

389. As noted earlier, s.47 is activated where ALAB “is of the opinion that any document, particulars or other information is or are necessary for the purpose of enabling it to determine an appeal.” The meaning of the phrase “is of the opinion” and like phrases was elucidated in the State (Lynch) v Cooney and in many cases since.658 Opinion is subject to judicial review – but only in a limited degree. It suffices here to mention three cases. The first is PL659 in which Hogan J held, citing Lynch, that:

“Any opinion formed pursuant to the exercise of a statutory power must be “bona fide held and factually sustainable and not unreasonable” and this phrase connotes “a laxer and more arbitrary level of ... assessment” as compared with a statutory test which requires the decision maker to be “satisfied””

I doubt that any distinction between being “satisfied” and being “of the opinion” will often make a difference

in practice in a particular case, but PL is illustrative in asserting that that the latter “connotes “a laxer and

more arbitrary level of ... assessment. Used Car660 and McCarthy Meats661 are authority that such an opinion may be “subjective”. It is clear that an applicant in judicial review impugning such an opinion faces an uphill task. And, as ALAB points out in written submissions, where ALAB forms such an opinion it “shall” serve a notice accordingly – it has an obligation to seek the “necessary” information.

653 Balz v An Bord Pleanála and Cork County Council and Cleanrath Windfarms [2019] IESC 90, [2020] 1 I.L.R.M. 367, §45.

654 Jennings & O’Connor v An Bord Pleanála & Colbeam [2023] IEHC 14 & cases cited therein.

Nurendale v Dublin City Council [2013] 3 IR 417.

656 McGrath v Trustees of Maynooth College [1979] ILRM 166.

657 Lohan v Solicitors Disciplinary Tribunal [2023] IECA 18.

658 [1982] IR 337. For examples of other cases since see Kiberd v Hamilton [1992] 2 I.R. 257, Kiely v Kerry County Council & Ors [2015] IESC 97, Waltham

Abbey & Ors v An Bord Pleanála [2022] IESC 30, Sweetman v The Environmental Protection Agency [2024] IEHC 55.

659 PL v Clinical Director of St. Patrick's University Hospital & anor [2018] IECA 29, §35.

660 Used Car Importers of Ireland Limited v The Minister for Finance [2020] IECA 298, §171 et seq.

661 McCarthy Meats v The Minister for Housing and Kildare County Council [2020] IEHC 371.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Further Information – Including After the Oral Hearing – Decision.

390. An example from the present case may illustrate the principles just stated: by e-mail of 28 August 2016 to ALAB, Dr Saunders, at a point when what he expected to be his final report was in preparation (in the event it became an interim report) expressed the view that “Overall, almost all of the evidence indicates that the licence should be granted. There are, however, a few issues which haven't been clarified or addressed, or have just been incompletely addressed, but they could turn out to be non-issues with the provision of a small amount of additional information.” Ceteris paribus, I can identify no legal principle or principle of justice or fairness to inhibit the further inquiry thus contemplated merely because the view had been reached that “Overall, almost all of the evidence indicates that the licence should be granted.” If anything, justice required that such inquiry be made.

391. As recorded above, particular objection is taken to s.47 Notices issued after the oral hearing. It is said that the oral hearing was the process whereby ALAB’s information-gathering should have been completed. I accept that in many cases – maybe most – that will prove to be so. It is no doubt often – perhaps even generally – the case that an oral hearing, in planning and similar licensing procedures, will in practice represent the conclusion of information-gathering by the decision-maker and that the only remaining steps will be for the chairman/inspector to report on the oral hearing to the decision-maker and for the decision-maker to make its decision. In the cause of good administration, it may well be generally desirable that this be so. It will encourage participants to put their entire case at an oral hearing rather than anticipate, as it were, further bites at the cherry. It tends to good order in administration of the process and to the efficacy of the oral hearing. But I do not see that such a de facto position is to be elevated to a de jure requirement that the oral hearing must bring inquiry to an end. The statute does not so provide. The very fact that there is power to re-open a concluded oral hearing implies the opposite.662 Even were that not so, I would take the same view. I cannot see why an oral hearing must constitute the end of the process of inquiry if further inquiry can be fairly made and could, as a matter of the discretionary judgment of the

decision-maker, reasonably be expected to improve the quality of the resultant decision. In general terms, such an approach chimes with ALAB’s duty of active inquiry into and scrutiny of the application before it. Indeed, the formalistic approach suggested by the challenge in this respect, absent a statutory requirement, would tend to impoverish decision-making where ALAB is of the view that additional reliable information, likely to be available in a timely manner, would likely appreciably assist its decision-making – whether tending towards grant or refusal of the licence. I reject this argument in its formal sense.

392. As to this argument in its substantive sense, I accept that the fact that ALAB’s information-gathering continued after the after the oral hearing is arguably of some weight in considering questions whether ALAB, instead of taking an independent and impartial approach to the decision of the Appeal, was bending over backwards, as is alleged, to avoid a refusal of the Aquaculture Licence application. That said, I do not think that the point is of such weight as to be even close to decisive on that issue. In my view, that is particularly

662 s.50 1997 Act. That this power was not exercised in the present case does not affect the general point.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

so where the recommendation of the chair of the oral hearing, whom I have found to be unbiased, in his report of November 2017, was that further information was required. It is important to put his recommendations in the context of his advice – both general and conditional – that the Board should issue an aquaculture licence. The further information he advised was required was of a significant order. Inter alia, he advised of what he considered to be important inadequacy of the Minister’s EIA – which had neglected to consider the risk of sea lice infestation of wild salmonid populations in the Dromagowlane/Trafrask Rivers. He also recommended a supplemental EIS, expert reports with a view to supplemental AA screening as to issues relating to otters, seals and birds and he recommended that ALAB make every effort to consider the potential impacts of large-scale farmed salmon escapes. I see no basis for a view that, rather than accept the advice of the chairman of the oral hearing, ALAB was obliged in law to have refused MOWI’s application. Once that advice was accepted and given the range and importance of the matters requiring further consideration, the exercise of further information-gathering powers became entirely predictable, indeed inevitable.

393. It does not seem to me that this judgment will benefit at this point from a blow-by-blow account of events thereafter as they relate to further information processes and similar processes. It is undoubtedly the case that there was significant and undesirable delay from the oral hearing report of November 2017 to ALAB’s determination of July 2021. On the other hand, ALAB was a part-time board faced with a highly complex and controversial decision, involving many “moving parts” as to which it was important that many stakeholders were invited to participate.

The moving parts included, generally, EIA, AA and Water Framework Directive issues, more specifically, issues as to seals, birds (various species requiring individual consideration), sea lice, freshwater pearl mussels, farmed fish escapes, and effects on wild salmonids – this list is incomplete. Stakeholders included MOWI, many objectors, including the Applicants, The Marine Institute, NPWS and BIM663 – and many participated actively. Inter alia, the sEIS and AAA were put to public consultation in which many participated. It must also be remembered that the participation of stakeholders itself generated issues which required further investigation and consultation, often legal advice and consideration.

To pick one example, IFI and its expert Dr Gargan disagreed with the sea lice dispersal analysis and its assumptions as to sea lice buoyancy. While ALAB’s consideration of materials to hand and inquiries of consultees inevitably reflected to some degree the trend of its iterative decision-making, it made searching inquiries. For example by s.47 request to the Marine Institute of 23 April 2019 it sought clarification of the MI’s view as to potential impact on the FwPM in the Dromogowlane/Trafrask river and asked if the MI had evidence to support its statement that the salmonid FwPM hosts in the Dromogowlane/Trafrask river system are predominantly non-migratory brown trout (i.e. that they do not smolt and become sea trout)? As a result of various investigations, ALAB on 15 May 2019 resolved on AA as to birds and required an NIS of MOWI.

Though I am critical of aspects of the manner in which ALAB dealt with the iteration of that NIS, I have already rejected allegations of bias in that regard. I have also rejected allegations of bias as to the steps ALAB took from finalisation of the NIS in July 2020 to its decision in July 2021. In particular, ALAB can hardly be criticised for inviting in September 2020 submissions by participants in the process and the public as to the content of the NIS and MERC report as to AA issues which had concluded that the Shot Head Salmon Farm

663 Bord Iascaigh Mhara.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

will not impact adversely on SCI species or SPA conservation objectives – an invitation taken up by, inter alia, objectors to the salmon farm, including An Taisce and Friends of the Irish Environment.

394. The Sweetman Applicants664 criticise ALAB for not circulating for further comment Dr Saunders’ report of 8 December 2020, Dr O’Toole’s report of 5 March 2021 as to the FwPM, the DMP Statistical Report of 19 May 2021 as to net entanglement risk to Gannets, MERC’s note of 19 May 2021 replying to objectors’ submissions and O’Toole Kelp Report 2021.

395. Browne665 cites Haverty666 as stating, in the context of an argument that an observer on a planning appeal should have been allowed to make a further observation in response to further submissions from an interested party, that:

“The essence of natural justice is that it requires the application of broad principles of commonsense and fair play to a given set of circumstances in which a person is acting judicially. What will be required must vary with the circumstances of the case.”

396. What is required, as is recorded in Genport,667 is that a quasi-judicial decisionmaker such as ALAB take reasonable steps to ensure that every party interested should have a reasonable general opportunity of making representations to it and of replying to submissions or representations by any other party. And so, as is said in, inter alia, Haverty, Wexele and Southwood,668 someone must have the last word. The practical fact is that publication for further comment must stop somewhere and, notably, the documents listed just above, of the non-circulation of which complaint is made, are not submissions from partisans but are reports by consultants to ALAB. It seems to me that the following observation of Murphy J in Haverty broadly applies here:

“I can appreciate their concern that they might have wished to expand upon their argument or to raise counter-arguments to those made in reply by the developers but I have no doubt that the real substance of their case was before An Bord Pleanála and duly considered by it.”

397. The iterative trend of ALAB’s thinking over time, in reliance on the advice of its experts, towards granting the aquaculture licence, is apparent in the documents. There is nothing wrong with that. There is no doubt that, as that trend advanced, ALAB went to considerable lengths in further investigation, including but not only, by s.47 and s.46 notices. The combination of a properly iterative movement towards a grant and a

664 Affidavit of Noel Carr – sworn 27 September 2021.

665 Simons, Planning Law, 3rd ed’n (Browne) §6-136.

666 State (Haverty) v An Bord Pleanála [1987] I.R. 485, §493.

667 State (Genport Limited) v. An Bord Pleanála, High Court, Unreported 1st February 1982, Finlay P, §8.

668 Southwood Park Residents Association v An Bord Pleanála & ors [2019] IEHC 504.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

scrupulous concern nonetheless to resolve potentially significant environmental issues can be ambiguous. It is perhaps a matter of disposition or the light one chooses to shine on such a combination whether one sees that as ALAB, as the Applicants characterise it, bending over backwards to grant the licence or, as ALAB might put it and to deploy the idiom without adopting it, bending over backwards to ensure they would make a correct decision. I would be very reluctant to criticise ALAB for trying to be thorough. Necessarily, this issue is adjudged holistically in all the cumulative circumstances. It may be that ALAB would have been entitled at points in the process to throw their hat at the application and refuse it on the basis that MOWI had not provided what was needed. Indeed, ex hypothesi, that is possible in any case in which a request for further information is contemplated. But that is a very different thing from saying they were obliged to do so. There is no obligation on a decision-maker to mechanistically refuse an application as soon as, and merely because, it becomes possible to do so. Generally, it is perfectly reasonable for a decisionmaker to consider generally that a licence should probably be granted but may have to be refused for particular reasons and to defer a decision, inquire accordingly and decide in light of the results of those inquiries. That is good decision-making.

Bending Over Backwards To Grant The Licence – Conclusion

398. The Applicants bear the onus of proof, on the balance of probabilities to a standard reflecting what is, in the end, a very serious allegation of improperly bending over backwards to grant the licence and grooming it to that end. In that end, and for the reasons set out above, I do not see that they have discharged that onus.

ALAB’S ACCEPTANCE OF THE RPS REPORT

399. PS CG3 pleads that the Aquaculture Licence is invalid as ALAB, in November 2015, accepted the RPS Water Quality Report 2015 from MOWI contrary to Ss.40(4),669 41(3)&(4),670 and/or s.44(2)&(4)671 of the 1997 Act. I reject this plea as misconceived. I accept that the RPS Report was validly part of a reply by MOWI to the many other appeals to ALAB, which reply ALAB had requested under s.44(2) of the 1997 Act and which reply MOWI was entitled to make. I also accept MOWI’s submission by that the RPS Report was before ALAB by two other legitimate means – via the oral hearing (which was adjourned specifically to allow of its consideration) and via the sEIS of April 2018. If I am wrong in these regards, I take the view that it would at this remove of time and effort be entirely unreal and unwarranted by any consideration of justice to quash the Aquaculture Licence for ALAB’s having considered what was, on any view, highly relevant information

669 ALAB shall determine the appeal as if the licence application had been made to ALAB in the first instance.

670 Appellants may not elaborate on or make further submissions on or submit further documents, particulars or other information on their appeals and

the Board shall not consider any such.

671 The other parties to the appeal have 30 days from notice of the appeal to respond. ALAB shall not consider late responses, elaborations on responses

or further submissions. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

(whether or not one agreed with it). It was also information which, it is fair to say, very much set the scene for the process from at least its publication in mid-2017 on and was information to which all participants had full opportunity to respond. If needs be, I would refuse relief in my discretion on that account

DELAY – TIME EXTENSION/BREACH OF TIMELIMITS – S.56 of the 1997 Act

S.56 of the 1997 Act – Expedition – Duty and Objective of ALAB

I have set out s.56 of the 1997 Act, as relevant, above. S.56(1) not merely imposes on ALAB a general duty of

expedition in determining appeals, it does so in quite imperative terms: “… shall …. ensure …. all steps …”.

However, the duty is softened appreciably by “as far as practicable” and the reference to “unnecessary” delays. There is a wealth of caselaw672 on the meaning of the word “practicable” in a statute and, as ever, context is important. Amongst the useful elucidations are “capable of being carried into action in a practical way having regard to such practical difficulties as exist” and, more prosaically pithy, “doable”.

400. Of itself, s.56(1) imposes no quantified time-limits. S.56(2), introducing the target of four months, softens the duty considerably further: the word “ensure” appears again but is preceded by “should endeavour to”. While that is clearly not strict, nonetheless it does put an order of magnitude on the duty of, and expectations of, expedition. S.56(3) permits service of notices extending the time for decision but only where the impossibility or inappropriateness of a decision within the 4 months derives from the “the particular circumstances” of the appeal and the notice must give reasons which, it follows, must relate to those particular circumstances. That said, the concept of inappropriateness seems a wide one. The notice must specify a date by which the ALAB “intends” to decide the appeal, and it is not apparent that such date must be within the following four months. The obligation is softened still further in that once a notice is served, ALAB’s obligation, whatever its stated intentions and despite the extension of time, remains only to “endeavour” to ensure a decision by that date.

401. On an overview of its text and as to expedition, it is difficult to avoid the strong impression that, at least as to its legal effect, s.56 is more aspirational than obligatory. It may be going only a little too far to describe it as virtue signalling – especially given the appointment of part-time, as opposed to full time, ALAB members, with a very modest secretariat indeed, to deal with inevitably complex and controversial licensing applications (at least as to salmon farming). This contrasts sharply with the impetus to expeditious determination of appeals evident in the strict time limits imposed by the Act on all but ALAB – see for example ss.40(1)&(3), 44, 45, 46, 47, 48 and 50. Indeed, failure to comply with a s.47 notice in time is an offence – as is failure to obey a notice to attend an oral hearing.673 One could observe that the State cannot

672 Caselaw reviews are found in McC v EHB [1996] 2 IR 296 and Ruigrok v. Commissioner of An Garda Síochána [2005] IEHC 439.

673 s.57(6)(d).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

be accorded the compliment of requiring others to do only that which it is willing to do itself. But doubtless there are good reasons for that. Indeed, the present case demonstrates the necessity of flexibility. It is quite impossible to see on the facts of this case that ALAB could have decided the Appeal in four months – even if it was a full-time body with much greater resources. While s.56 must be interpreted as “text in context”, it does not seem to me possible to consider that scheme of the 1997 Act as to expedition requires interpretation of s.56 as imposing stricter duties of expedition than might otherwise be found in its ipissima verba. Indeed, the very contrast between the requirement that ALAB move expeditiously with the terms in which deadlines are imposed on others suggests not.

402. S.56(3) is, however, clear as to one time limit: the first s.56(3) time extension notice must be served before the expiry of four months from ALAB’s receipt of the appeal. It is common case that the first s.56(3) notice was not served until on 8 April 2016 – approximately 6 months after the Appeals were filed in October 2015 and so in breach of s.56(3).

403. In considering the adequacy of the reasons given in s.56(3) notices as explicitly required by s.56(3), it must be remembered that such notices are justiciable as capable of being quashed in judicial review. Accordingly, the reasons must both:

• be accurate and complete (though not necessarily detailed) in identifying the actual reasons which moved ALAB in issuing the notices and

• give information sufficient to enable the recipient to decide whether to seek to quash the notice in judicial review – Connelly.674

The s.56 Notices and Notes Thereon

404. The Appeals were filed in October 2015 and determined by ALAB in June 2021. As stated, the first s.56(3) notice extending the time period issued on 8 April 2016 – about 6 months after the receipt of the appeals. Later similar notices were dated 16 November 2016; 15 March 2017; 11 October 2017; 20 December 2017; 31 August 2018; 26 June 2019; 24 March 2020; 14 August 2020; and 16 December 2020. There were ten such notices in all and the elapsed period between receipt of the appeals and ALAB’s

decision of was 68 months – or approximately 17 times the 4-month norm set in s.56(2)(a).

405. The sequence of s.56 notices and some notes thereon can be tabulated as follows:

674 Connelly v An Bord Pleanála & Clare County Council & McMahon Finn Wind Acquisitions Limited, [2018] IESC 31, [2021] 2 IR 752 §36 et seq, citing,

inter alia Mallak v Minister for Justice [2012] IESC 59, [2012] 3 IR 297 & EMI Records v Data Protection Comm [2013] IESC 34, [2014] 1 ILRM 225.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

# Date of S.56 Extension Duration of Reason Stated Notice to extension (approx.)

1 8 April 2016 30 7 months Time to date expended in taking legal advice on and November consideration of Appellants’ concerns as to conflicts of 2016 interest. Substantive consideration of appeal delayed as a result.

Will consider requests for oral hearing.

Note – This notice was served about 6 months after the receipt of the appeals and about 2 months after the expiry of the 4-month time limit set by s.56 for the issue of the first extension notice.

2 16 November 31 May 6 months Not in a position to conclude within the expected 2016 2017 timeframe.

Notes

• It is difficult to see this notice as, in its express terms, compliant with the requirement of s.56(3) of the 1997 Act that a s.56 notice state ALAB’s reasons why it is not possible to determine the Appeal in the time available. Reasons may be brief but they must be meaningful. They may assume such knowledge of the context as is available to the recipients. But the reason given here is merely conclusionary – indeed it merely expresses the only possible conclusion underlying a s.56 notice. The reason must state why ALAB is not in a position to conclude the appeal within the expected timeframe.

• ALAB’s minutes of 15 November 2016 record, in substance, ample reason why a decision could not be made by 30 November 2016.675 They record a decision to issue s.47 notices accordingly but not a

decision to issue s.56 notices accordingly.

• SWI, by letter dated 17 November 2016 both complained of delay and the lack of explanation for it and also reiterated its call for an oral hearing. While broadly understandable, this letter is appreciably selfcontradictory in that an oral hearing would inevitably delay matters.

3 15 March 31 October 7676 or 5677 Adjournment of oral hearing and complexity of matters to 2017 2017 months be determined.

Note – The Oral Hearing had been adjourned on 15 February 2017, inter alia to permit publication of the RPS Report 2015, publication of which ALAB had overlooked.

675 ALAB agreed that further information was required from (1) MOWI, (2) IFI, (3) NPWS and the (4) Marine Institute before any further consideration can

be given by the Board to the Appeal or whether to hold an oral hearing. The issues requiring clarification were: 1. Salmonids and freshwater pearl mussel

in Dromagowlane River; 2. Well boat discharges; 3. The suitability of the cage and mooring system; 4. The use of Emamectin Benzoate; 5. Seabed impacts

in respect of the requested change in the licence conditions to accommodate four additional cages; 6. The presence of potentially harvestable Nephrops

in the licence area.

676 From date of s.56 notice.

677 From expiry of previous notice on 31 May 2017.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

# Date of S.56 Extension Duration of Reason Stated Notice to extension (approx.)

4 11 October 31 January 3 months Adjournment and reconvening of the oral hearing and 2017 2018 complexity of the matters to be determined Notes

• The Oral hearing had ended on 20 September 2017. The chairperson’s report was awaited.

• This deferral can have come as a surprise to no-one.

5 20 December 31 October 10678 or 9679 Chairperson’s Oral Hearing Report680 published.

2017 2018 months Accordingly, ALAB is seeking additional information:

• sEIS681 re: Sea lice risk to wild salmonids and in turn to FwPM682

• Reports on risks to Otters, Seals and Birds.

6 31 August 30 June 10683 or 8684 Arising from its consideration of submissions (which had 2018 2019 months been requested on 10 April 2018) and ALAB’s review of the timescale on 28 August 2018.

Notes

• S.46(1) letters from ALAB to all on 10 April 2018 had enclosed the Saunders Otter Report of 24 November 2017, the Coram Seal Report of 1 February 2018, the Gittings Wild Birds impact report of 5 February 2018.

• Also relevant here was the s.47 request dated 27 February 2018 to the Marine Institute as to bird impacts & its reply dated 28 April 2018.

• ALAB had considered the replies at its meetings on 1 May 2018 and 12 June 2018 and resolved to o seek legal advice on the reports, o require MOWI to publish its sEIS dated 16 April 2018 (which included MOWI’s Pest Management Plan as to sea lice).

o ask Dr Saunders to evaluate all information to hand and update his draft report.

o on receipt of Dr Saunders’ updated draft report, issue a further s.47 notice to NPWS as to the need for AA, the adequacy of the EIS and the Minister’s EIA, the FwPM, Gannets and other birds, seals and otters.

o defer a decision as to a time extension.

• ALAB at its meeting on 28 August 2018 had also

678 from date of s.56 notice.

679 from expiry of previous notice on 31 May 2017.

680 dated 8 November 2017.

681 Supplemental EIS

682 Freshwater Pearl Mussel.

683 from date of s.56 notice.

684 from expiry of previous notice on 31 May 2017.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

# Date of S.56 Extension Duration of Reason Stated Notice to extension (approx.)

o resolved (again) to require MOWI to publish its sEIS and Pest Management Plan and invite submissions thereon.

o resolved that “Arising from the receipt of the submissions and observations received to the Section 46 Notice and bearing in mind the steps yet to be taken, including publication of the notice concerning the Supplemental EIS as well as the requirement to allow time for submissions concerning same and their subsequent evaluation, the Board reviewed again the timescale for determination of this appeal.”

• Viewed together, and in isolation from the more general progress of the Appeals, these resolutions seem to me to justify a further extension of time. However, they might well have been, but were not, described by way of reasons in the s.56 Notice of 31 August 2018.

• SWI by letter to ALAB dated 1 October 2018 submitted that, o ALAB clearly was not in a position to conclude in AA as a certainty that there would be no impact on any European Site or protected fauna or flora.

o MOWI, the Minister and ALAB cannot keep on seeking some scientific basis for such certainty.

o “ALAB cannot keep on adjourning and deferring a decision indefinitely. While there may be a statutory basis for the deferral of a decision, the licence Applicants should not be indulged with continuous adjournments.”

o The licence application should be immediately refused.

• In November 2018, on ALAB’s instruction, MOWI published its sEIS. 8 weeks were limited for public response and, by letters dated 29 November 2018, ALAB invited all parties to respond.

7 26 June 2019 31 March 9 months AA Screening recommended AA as to Bird SCIs.

2020 ALAB has sought an NIS from MOWI by 23 September 2019.

8 24 March 31 August 5 months Not in a position to conclude within the expected 2020 2020 timeframe.

Notes

• IFI impugns the reasons for this s.56 Notice in particular as inadequate at approximately 4.5 years into the licence application process.

• As stated above, it is difficult to see a s.56 notice in these terms as complaint with ALAB’s obligation to state the reasons why it is not possible to determine the Appeals in the time available.

• This s.56 notice was anticipated in ALAB’s meeting of 31 January 2020 and decided upon at ALAB’s meeting of 19 March 2020, the minutes of which meetings record:

o The need to allow new ALAB members to read into the file.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

# Date of S.56 Extension Duration of Reason Stated Notice to extension (approx.)

o ALAB’s acceptance of MERC’s view that AA was not a barrier to grant of the licence and that MOWI’s NIS of October 2019 so implied but should explicitly so state.

o ALAB’s decision to require an amended NIS by notice to MOWI under Art 42(3)(b) of the Habitats Regulations.685 o That thereupon, MERC’s AA report would need updating.

o That the amended NIS and MERC’s AA report would be circulated to allow submissions by all parties.

o That on receipt of those submissions, Dr Saunders would update his report.

o The documents on ALAB’s website would be reordered/renamed “so it is clear and easy for all parties to follow”.

o Extension of time was necessary “due to the complexities of the appeals”.

• It is not apparent why ALAB minuted these reasons but did not state them, if only briefly, in the s.56 notice.

• Nonetheless, it is not apparent that IFI protested the inadequacy of reasons at the time or suffered any prejudice due to their inadequacy.

9 14 August 31 4.5 months Not in a position to conclude within the expected 2020 December timeframe “due to the complexities of the appeal and the & 18 2020 availability of reports”.

September 2020686 Notes

• IFI impugns the reasons for this s.56 notice in particular as inadequate at almost 5 years into the licence application process.

• ALAB’s minutes of 15 August 2020 records that it o awaits MERC’s finalized AA Report & Dr Saunders’ finalized Report.

o decided to extend time due to the possibility of further public participation on foot of s.46 notices.

• It is not apparent why ALAB minuted these reasons but did not state them in the s.56 Notice.

• Nonetheless it is not apparent that IFI protested the inadequacy of reasons at the time or suffered any prejudice due to their inadequacy.

685 European Communities (Birds and Natural Habitats) Regulations, Art 42(3) At any time following an application for consent for a plan or project, a

public authority may give a notice in writing to the applicant, directing him or her to — (b) furnish any additional information that the public authority

considers necessary for the purposes of this Regulation.

686 E-Mail “To whom it concerns” from ALAB to, inter alia, IFI. Exhibit GF4 Tab 1.14 to Affidavit of Gregory Forde sworn 18 September 2020. It is unclear

why this was sent given the S.56 Notice 14 August 2020 but no point is taken on the issue. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

# Date of S.56 Extension Duration of Reason Stated Notice to extension (approx.)

10 16 December 30 June 6 months Assessment of responses received as part of the public 2020 2021 participation process and conclude the AA.

24 & 29 June ALAB’s Determination made & signed.

2021

Note – The total time from ALAB’s receipt of the Appeals to ALAB’s decision was about 68 months – or about 17 times the 4-month norm set in s.56(2)(a) of the 1997 Act.

Delay – Pleaded Bases of Challenge

406. Under the general rubric of alleged breach of its duty of expedition, IFI plead687 that ALAB’s decision should be quashed for the following reasons:

i. ALAB lost jurisdiction to determine the Appeal by failing to ensure unbroken continuity of 4- monthly renewals of s.56 extensions. At trial, this morphed into a particular and arguably unpleaded argument that by not serving its first s.56(3) time extension notice within the 4 months stipulated ALAB lost jurisdiction to determine the Appeal. On balance, I think it would be unfair to shut IFI out of this argument for want of pleading.

ii. More generally, ALAB breached its duty of statutory duty of expedition. (S.56, which imposes the general duty of expedition, is pleaded generally.) The 6 years ALAB took to determine the Appeal was excessive. ALAB’s unreasonable and “severe delay” in a “long drawn out” process breached s.56(2),688 (3)689 and (4)690 of the 1997 Act. It included ALAB’s continuous facilitation of MOWI by way of “inordinate amount of time and opportunity” so it could address environmental concerns as they arose in the appeal. ALAB in effect decided the appeal process in accordance with its technical advisor’s first report691 and managed the appeal thereafter to allow MOWI to accumulate the necessary documents to permit the granting of the licence.

iii. IFI argue that many s.56(3) notices breached ALAB’s explicit duty to give reasons for extending time. S.56(3) is pleaded and the notice it mandates is a notice containing only two elements: the

687 Grounds 1/12/22 CG2 & §32.

688 The Board should endeavour to ensure that every appeal is determined within four months.

689 Notices of the reasons for delay and the date by which a decision is intended.

690 Board shall endeavour to ensure that the appeal is determined by the date specified in the notice.

691 Saunders, 31 December 2016.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

reasons for delay and the date by which a decision is intended. I reject ALAB’s argument that this issue was not pleaded as to the obvious examples, being the notices of 16 November 2016,

24 March 2020 and 14 August 2020.

407. At trial, IFI argued that there is no minute of an ALAB decision authorising the s.47 notice of 16 November 2016.692 I reject this point as not pleaded.

408. SWI693 and the Sweetman Applicants694 make similar, if more general, pleas as to delay and breach of the duty of expedition imposed by s.56. They allege failure to avoid unnecessary delays and compare the 68 months taken to decide the Appeals with the 4 months contemplated in s.56.

409. All Applicants say that however elastic the general duty of expedition set by s.56(1), ALAB must be in breach of it by taking 68 months to decide the Appeal – to find otherwise would eviscerate the duty of meaning.

410. That egregious delay is also characterised as explicable only in terms of objective bias by way of a determination – “bending over backwards” – to grant the Aquaculture Licence. For reasons already given I reject this argument

Delay – Opposition

411. ALAB say generally that the time lapse is explained by the complexity of the case and the necessity of its thorough investigation and consideration before decision.

412. ALAB pleads695 to the IFI case,696 beyond traverses, that

• It did, as far as practicable, determine the Appeal expeditiously and avoid unnecessary delays in accordance. The time it took was to ensure inter alia that it o fulfilled its legal obligations, properly determined all issues raised in the Appeal, including by IFI, o did so on sufficient information and on the appropriate or best available scientific advice and

692 Transcript Day 2 p40.

693 SWI CG4 & §§4.5 & 4.6.

694 PS CG5 & §22.

695 Opposition 279/7/22 §9 et seq.

696 With one exception, it does not seem to me necessary to set out here its pleas on these issues in the other two cases.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o ensured full participation by all parties.

• In these regards it pleads, o ss.35697, 44698, 46699, 47700, 49701 of the 1997 Act o the 2012 Aquaculture Appeals EIA Regulations.702 In particular, Arts. 3(5) and 3(6) empower ALAB to “require the production by the appellant of any additional or supplemental information that it considers necessary to enable it to make an assessment” and to require information to remedy inadequacies in an EIAR. Art. 3(7) requires ALAB, in effect, to consider all materials properly before it and Art. 3(9) empowers ALAB to have regard to reports by its advisors.

• No-one protested the delay in determining the appeal or any of the s.56 notices or the reasons given therein and all continued to participate in the Appeals.

• The requirement to serve a first s.56 notice (dated 8 April 2016) within the applicable 4 month period is procedural. Absent complaint, ALAB was entitled to continue consideration of the Appeal and it was appropriate that it do so.

• No-one has suffered, pleaded or particularised prejudice by reason of any defects in the s.56 notices.

• IFI lacks standing to complain of these issues, has waived its right to complain and/or has acquiesced in the matters of which it complains.

• IFI is out of time to complain – especially as the last s.56 notice was served on 16 December 2020.

413. ALAB does concede703 that SWI complained of delay by letter of 1 October 2018704 but not by reference to s.56 or any other legal obligation. ALAB pleads that SWI

• made no further complaint and took no further action in that regard.

• has no standing to make such a complaint now in and/or has waived any such claim and/or has acquiesced and/or is outside the time prescribed by O.84 RSC to make such a claim and/or relief should be refused for SWI’s delay in making this claim now in these proceedings.

ALAB “may, from time to time, engage such consultants or advisers as it considers necessary for the performance of its functions”.

698 Parties’ entitlements to make submissions.

699 ALAB may request submissions or observations “in the interests of justice”.

700 ALAB shall require the submission of documents, particulars or other information where of the opinion that they are necessary to determine an appeal.

701 Power to hold an oral hearing.

702 S.I. No. 468 of 2012 – Aquaculture Appeals (Environmental Impact Assessment) Regulations 2012.

703 Opposition to SWI 23/1/23 §25.

704 Summarised in the Chronology above.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

General Principles & Cases as to Timeliness

414. It bears observing that timeliness of decision-making is a general principle of both domestic and EU administrative law. In EU law it derives from the general right to good administration under Article 6(1) TEU705 and Article 41 CFREU706 – see Holland.707

415. No matter how complex, controversial and hard-fought the individual case, it is difficult to see that a saga in which the present licensing applications were made in 2011 and the Appeal was decided in 2021 conforms to any sensible principle of timeliness of decision-making. Whatever the competing views of participants in these licensing processes, if the facts of this case are in any real degree typical as to the complexity of the decisions required and the time and resources required to make them, it is in general terms difficult to see that the 2017 Report708 of the Independent Aquaculture Licensing Review Group can have been anything but broadly correct in describing the system as unworkable, outdated and in need of urgent root-and-branch reform. One can only see this observation, made 7 years ago, as expressing an authoritative view that the system, at least at that time, languished in disrepute. However, I am concerned only with the lapse of time before ALAB, which ran from the Appeals filed in November 2015 to its determination in June 2021. I note that in Behan709 Barrett J observed that “there is no rule of law that staffing issues within an administrative body must inexorably rebound to the detriment of a person making application to same, even if that detriment is ‘but’ that that person must wait an inordinate/unreasonable time for a decision on such application …” Ousley J observed, albeit obiter, in UK Coal Mining710 as to the Article 6 ECHR right to a determination of civil rights within a reasonable time that,

“… the question of the normal workload of the relevant authorities, or a continued shortage of resources do not excuse a delay711 ….. This is because if the normal level of resources and workload or the normal decision-making process itself gives rise to unreasonable delay, the right would be nullified by the structures of government when instead it should be respected, and structures adjusted accordingly. This was not a case where the Secretary of State referred to some particular temporary problem that he had in dealing with letters expeditiously or some problem particular to this case or affecting the particular decision-maker e.g. ill-health or a rash of such decisions. I am of course sympathetic to the problems of normal workload but that is not a matter which appears under the Convention to be relevant.”

705 Treaty on the European Union. Article 6(1) The Union recognises the rights, freedoms and principles set out in the Charter of Fundamental Rights of

the European Union …… which shall have the same legal value as the Treaties.

706 Charter of Fundamental Rights for the European Union. Article 41(1). Every person has the right to have his or her affairs handled impartially, fairly

and within a reasonable time by the institutions, bodies, offices and agencies of the Union.

707 Holland v The Minister for Justice [2023] IECA 73.

708 Report of the Independent Aquaculture Licensing Review Group; Review Of The Aquaculture Licensing Process May 2017.

709 Behan v An Bord Pleanála [2020] IEHC 133.

710 UK Coal Mining Ltd v Secretary Of State For Local Government, Transport and Regions [2001] EWHC 912 (Admin).

711 Citing Zimmerman and Steiner v Switzerland (1983) 6 EHRR 17.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

As compared to Article 6 ECHR guarantees it is not apparent that a different view would be taken at Irish law of the right to a decision within a reasonable time. However, ALAB did not plead staffing shortages or lack of resources – it relied rather on the inherent, and undoubted, complexity of the issues to be decided.

Holland 2023, Lafarge Redland 2000, Diesel 2020, Wexele 2010, Behan 2020.

416. On a review of the authorities as to delay in decision-making712 the Court of Appeal recently drew some general conclusions in Holland.713 Simplifying the facts appreciably, in Holland the applicant for a residence card for his stepdaughter sought mandamus compelling a decision on the application given a 23- month delay in doing so. The decision had been made by the time the proceedings came on for trial but the Court of Appeal decided to decide the moot. It declared the respondent in breach of the Irish law duty to make a decision within a reasonable time – though declining to rely on Article 41 CFREU or on the appellant's asserted EU law right to an effective remedy and/or good administration. Below I adjust the conclusions of the Court of Appeal somewhat to present circumstances but am happy that what follows accords with those conclusions.

• Where an authority has power to make a decision, but no time is fixed by law for it to be made, there is nevertheless a duty to make the decision within a reasonable time.

• What time is reasonable will vary with the circumstances of each case, including the investigations required of the authority and the time taken by an applicant and others to respond to enquiries made by the authority.

• A time limit applicable to a first instance decision is informative as to what may be considered a reasonable time to take a decision on an appeal from, or on a review of, a first instance decision.

• In a case of egregious delay, amounting either to a violation of rights or so as to be tantamount to a refusal, mandamus may issue requiring that the decision be made.

• Mandamus will generally become moot if the decision, the making of which it is sought to compel, is made before trial of the judicial review.

712 Druzinins v Minister for Justice [2010] IEHC 84, El Menkari v Minister for Justice [2011] IEHC 29, Saleem v Minister for Justice [2011] IEHC 49 , Mahmood

v Minister for Justice [2016] IEHC 600, Lofinmakin v Minister for Justice [2013] IESC 49, [2013] 4 IR 274, Okunade v Minster for Justice Equality and Law

Reform [2012] IESC 49, [2012] 3 IR 152, Nearing v Minister for Justice [2009] IEHC 489, [2010] 4 IR 211, Case C-604/12 H.N. V Minister for Justice, Equality

and Law Reform, Ireland and the Attorney General, Judgment of 8 May 2014. H.N. v Minister for Justice, Equality and Law Reform and Others; Straczek v Minister for Justice [2019] IEHC 155.

713 Holland v Minister for Justice [2023] IECA 73.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

417. As compared to a statute which does not fix a time limit, an explicit statutory duty of general expedition, at very least, emphasises the importance of timeliness. In a general sense, it also tends to reduce the time which may be thought to be reasonable.

418. In Lafarge Redland,714 the principles stated were similar to those set out in Holland. A delay of over 9 years or 7,715 despite which the planning application remained undecided, was described, even allowing for special circumstances including the complexity of the case, as “scandalous” – indeed in breach of the planning applicant’s right to a decision within a reasonable time as provided in Article 6 ECHR716 The petitioner, who was the planning applicant, got declarations directed at ensuring a decision within a further reasonable time. The planning applicant/petitioner in North Lowther717 failed on the facts as to a 3.5 year delay but Lafarge Redland was upheld.

419. In Diesel SPA718 and though delay was not an issue in the case and no relief flowed from her finding of 18 years of delay, Irvine J felt it would be “remiss not to express the view that the delay of the parties in bringing to fruition, in May 2012, an application for registration719 lodged in 1994, is nothing short of egregious. Delay of that magnitude, absent extraordinary circumstances, should not occur and cannot be in the interest of the parties apart altogether from the fact that it has the potential to bring the statutory

scheme into disrepute.” Though the delay involved here is a lot less, it is difficult not to apply this rationale to

the present case at least in general terms.

420. In Wexele720 a disappointed developer failed to quash the refusal of its planning application. Inter alia, it complained of a 2½ year delay in deciding the application. O’Neill J thought the complaint arguable but insubstantial. All the steps taken by the respondent which added to the overall time it took to dispose of the appeal, were necessary for its fair disposal of a very difficult and complex problem, having regard to the size of the proposed development, the infrastructural constraints, the interests of various statutory agencies, the interests of the local residential and business communities, the interests of the wider community in the area of the planning authority. He was satisfied that the time taken was unavoidable if the appeal was to be heard and determined fairly to all concerned.

421. In Behan721 the applicant challenged the Board’s refusal of substitute consent for his quarry. The case is unusual in that Barrett J quashed the refusal on the basis that Mr Behan’s application for substitute

714 Lafarge Redland Aggregates Ltd v Scottish Ministers [2000] 4 PLR 151.

Depending on how one calculated it.

716 European Convention on Human Rights. No such argument was made here – though SWI cited the equivalent Art 47 of the Charter on Fundamental

Rights of the EU, for the entitlement to “an independent and impartial tribunal previously established by law.”

717 North Lowther Energy Initiative v Scottish Ministers [2021] CSOH 104.

718 Diesel SPA v Controller of Patents [2020] IESC 7 (Supreme Court, Irvine J, 19 March 2020).

719 Of a trademark.

720 Wexele v An Bord Pleanála [2010] IEHC 68.

721 Behan v An Bord Pleanála [2020] IEHC 133 (High Court (Judicial Review), Barrett J, 12 March 2020).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

consent had been invalid and should have been rejected as such and so formed no proper basis for the impugned refusal. Barrett J also held the Board’s culpable delay of 4 years and 5 months in deciding to refuse was in breach not merely of the general duty of expedition imposed by S.126 PDA 2000 on the Board but also of Mr Behan’s constitutional right to fair procedures, and/or his right to good administration under Art. 41 CFREU.722 S.126 is very similar to s.56 of the 1997 Act. The judgment is not entirely clear what reliefs flowed from that finding of delay – though it is clear that at least a declaration did. However, as Barrett J found that the substitute consent application had been invalid in any event, the circumstances and legal basis of the decision were far simpler than those of many cases of delay and of the present case. The case would not seem to be authority for relief beyond a declaration.

422. Since the decision in Holland, reliance on Article 41 now seems unnecessary. That said, the rationale underlying the Irish law right to a decision in a reasonable time is the same as that underlying the Art. 6 ECHR right to a determination of civil rights within a reasonable time. Ousley J observed, albeit obiter, in UK Coal Mining723 that the rationale is “to keep to a minimum the period of uncertainty during which a person's civil rights and obligations are subject to determination by a judicial decision.”

Browne 2011, Dolan 1975

423. In Browne724 Hedigan J held that the 2-year time-limit from registration of a quarry, imposed by s.261(6)(a) PDA 2000 on the Planning Authority’s imposition of conditions on its operation, was mandatory. At issue were the constitutional property rights of the quarry-owner. S.261(6)(a) commenced with the words “not later than”. No power to extend time was given and no provision was made for what was to happen if the limit was exceeded. Hedigan J followed Dolan725 – citing Lowry LCJ as to statutory time limits to the effect that:

“Certain principles, although they are not rules, may at least be adopted as guides:

1. A time limit is likely to be imperative where no power to extend time is given and where no provision is made for what is to happen if the time limit is exceeded;

2. Requirements in statutes which give jurisdiction are usually imperative;

3. Where the act is to be done by a third party for the benefit of a person who will be damnified by non-compliance, the requirement is more likely to be directory;

722 Recognised as a general principle of European Union law in Case C-604/12 H.N. v Minister for Justice, Equality and Law Reform [ECLI:EU:C:2014:302].

723 UK Coal Mining Ltd v Secretary Of State For Local Government, Transport and Regions [2001] EWHC 912 (Admin).

724 Browne v Kerry County Council [2011] 3 IR 514.

725 Dolan v O’Hara [1975] NI 125.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

4. Impossibility may excuse non-compliance even where the requirement is imperative.”

The immediately preceding words of Lowry LCJ bear citation as context for those principles:

“The question is whether the requirement that the applicant transmit the case stated to the Court of Appeal within fourteen days is imperative, with the result that non-compliance deprives the appellate court of the jurisdiction to hear the case stated, or merely directory.

It has proved impossible to lay down a general rule and only rarely does the form of words used in the statute provide a clue. Even then, the decisions are hard to reconcile.726 ……. prior authority on other statutes is not of great assistance to the court in determining the intention of the Legislature.

The learned judge acknowledged the help he had received from the judgment …….of Middleton J.727, who posed the test in the form of a question:

“Was the matter in which there was disobedience so essential and fundamental that the noncompliance with the statute rendered it void, or was it so subsidiary and collateral that it may be safely ignored?”

I likewise find this last question/test helpful – though there may be a considerable a middle ground between “essential and fundamental” and “subsidiary and collateral”.

424. The third point made by Lowry LCJ is of some relevance. Paraphrasing, where the decision is to be made by ALAB for the benefit of a person who will be damnified by delay, the requirement of expedition is more likely to be directory. Clearly many, including the Applicant/Objectors, have an interest in expedition by ALAB. They do not want to be interminably active in their opposition to the salmon farm. But it seems to me that the party primarily interested specifically in expedition (as opposed to in the substantive outcome) is MOWI.

Cork County Council 2010, Celtic Roads 2013, Irish Refining 1990, Saleem 2011.

425. Cork County Council v The Valuation Tribunal728 is of present interest less for its obiter that the relevant statutory time limit729 was directory rather than mandatory, than for its analysis underlying that finding. The statute required the tribunal to decide valuation appeals within 6 months. The Council’s attempt to have its decision quashed as made out of time failed for various reasons. In deciding that the time limit

726 Maxwell, 12th ed. pp. 314-322, Craies, 7th ed. pp. 260-274, Montreal Street Rail Co. v Normand in [1917] A.C. 170 cited by Jones L.J. and, more

recently, B. v B. [1961] 2 All E.R. 396.

727 in a Canadian case, R. v McDevitt (1917) 390 O.L.R. 138.

728 Cork County Council v The Valuation Tribunal, and ESAT Telecom and The Commissioner of Valuation, [2010] 1 IR 57.

729 s.37(2) Valuation Act, 2001.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

was directory rather than mandatory, Dunne J observed, in terms which seem to be widely applicable to time limits within which decisions are to be made in which participants have opposing interests:

“One only has to consider what would happen to appeals if a decision was not furnished within six months. Does that mean that all such appeals are automatically invalid or liable to be struck down?

That cannot be the purpose of the provision. Rather it seems to me that the legislature was setting a time limit within which such decisions should be furnished and providing some certainty to parties as to when they are entitled to require or compel the Valuation Tribunal to furnish a decision.”

426. In other words, such time limits may entitle interested parties to mandamus to enforce expedition but a decision made late is not thereby invalidated. Of course, much may depend on the nature of the

decision to be made and the terms of the statutory scheme. But it is not difficult to infer that Dunne J’s remark “That cannot be the purpose of the provision” was grounded in a view that invalidating such decisions would serve no useful purpose and cause great inconvenience. Often, in reality, invalidation for delay would generate even further delay assuming the application is made again – it is difficult to see that an applicant could be shut out from having his application decided on its merits. And if the delayed decision was made in an appeal process (as here before ALAB) difficult questions could arise as to the position remaining after quashing the appeal. Would the original decision be re-instated and would the appellant – in the present case, appellants on both sides of the issue – have thereby been deprived of his appeal through the fault only of the appellate tribunal? And if the original decision was not re-instated, the successful party would have been deprived of that decision through the fault only of the appellate tribunal. It is easy enough to see that, often at least, the lesser evil is that delay or breach of time limits does not invalidate a decision on the merits of the application or appeal.

427. Celtic Roads730 challenged the Valuation Tribunal’s adjournment of certain valuation appeals to a date beyond the 6-month time limit to await the outcome of a case stated to the High Court in another valuation appeal, which outcome, the Tribunal considered, was likely to bear on the appeals before it. The Valuation Tribunal cited the Cork County Council case to the effect that the time limit was directory not mandatory. Peart J granted a declaration that the Tribunal had no power to so adjourn the appeals, at least absent exceptional circumstances, which the pending case stated did not constitute.

428. By the time Celtic Roads was tried by Peart J, the case stated pending which the valuation appeals had been adjourned had been decided and the Tribunal had set a date to hear the Celtic Roads appeal. Peart J held that by the time he gave judgment “in all likelihood … by now the appeals have been determined”. Peart J therefore held that the issue of the Tribunal's power to adjourn an appeal beyond the 6-month

730 Celtic Roads Group [Portlaoise] Ltd v The Valuation Tribunal [2013] IEHC 180 (Peart J, 29 April 2013).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

period was “moot as between these particular parties”.731 That conclusion can only have been based on the Tribunal’s continuing jurisdiction to decide the appeal despite its breach of the time limit (though it must be said that an application to quash any determination of those appeals was not before Peart J). Peart J considered that the Valuation Tribunal had no jurisdiction to voluntarily ignore the 6-month time limit, at least absent exceptional circumstances. But he accepted, obiter, the view taken by Dunne J in the Cork County Council case that the 6-month time limit was directory rather than mandatory “in the sense that a

decision made outside that period specified should not be regarded as invalid”.

429. Irish Refining732 shows the Supreme Court’s attitude to delay by a decision-maker in its effect on the validity of an appeal – albeit in this case delay by the decision-maker at first instance. The relevant Act733 entitled a dissatisfied party to a valuation appeal to require the circuit judge to state a case to the High Court within 21 days after being so required. The Commissioner of Valuation so required and the circuit judge missed the deadline.734 Irish Refining unsuccessfully sought prohibition of his doing so belatedly – alleging his want of jurisdiction and observing that the Act provided no power to extend time. Egan J in the High Court held the time limit directory as “it could never have been the intention to deprive a party of his right of appeal by way of case stated.”

430. The Supreme Court agreed, considering it “quite clear as a matter of general principle, that time limits contained in statutory provisions may be either mandatory or directory and that the court in reaching a

conclusion into which category any individual provision falls must seek to interpret the meaning, intention and objective of the legislation concerned.” Finlay CJ held that interpreting the time limit as mandatory would be unfair in that the person seeking the case stated would be entirely at the mercy of the judge and such manifest unfairness should not be assumed to have been the real intention of the legislature. Finlay CJ observed that counsel had cited decisions in which persons seeking to appeal in statutory procedures, who failed to carry out a particular step within a statutory time limit, were held to have lost their right to invoke the further jurisdiction of the courts and continued:

“He was not, however, able to point to any case in which it had been decided that a person's right to ……. appeal, ……. could be lost ……. by the failure of the court to carry out some specific duty within some specified limited time.”

It seems to me that the relevant analogy here approximates ALAB’s failure of expedition to that of a court in this passage.

731 Peart J decided to determine the case nonetheless on the basis that “It is probable that the issue will arise again, and since the resolution of the issue

is a matter of statutory interpretation and not particularly dependent on the particular facts of the present case, it ought to be decided now, as there is

a wider benefit to other property owners, the Commissioner of Valuation and indeed the Tribunal itself.” On that basis he held that, though the

proceedings were moot as between these particular parties, they were not moot in the sense of the judgment of Murray C.J. in Irwin v Deasy [2010] IESC

35. I respectfully suggest that this amounts not to a finding that the proceedings were not moot but to a finding that they should be decided despite

being moot and on foot of one of the well-recognised exceptions to the general rule against deciding moots.

732 Irish Refining v Commissioner of Valuation [1990] 1 IR 568.

733 Section 10 of the Annual Revision of Rateable Properly (Ireland) Amendment Act, 1860, as amended by s.31, sub-s.3 of the Courts of Justice Act, 1936.

734 For what seem to have been good reasons: the parties unsuccessful attempts to agree the terms of the case stated.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

431. Saleem735 merits mention in light of an argument that a timely decision by ALAB of the appeal would have been a refusal of the licence on the basis of insufficiency of the information tendered by MOWI to support a grant – such insufficiency being apparent from ALAB’s repeatedly bespeaking further information of MOWI and others. Mr Saleem first sought mandamus to compel decision of his application for long term residency, complaining of a 15-month delay. That claim was compromised. The Minister then refused him long term residency on the basis that, at the date of that decision, Mr Saleem’s permission to remain had expired shortly after his proceedings had started. A current permission to remain at the date of decision of an application for long term residency was a condition of the grant of long term residency. Mr Saleem sought certiorari of the refusal of long-term residency on the basis that, had the decision been made in a timely manner, his permission to remain would have been current at the date of such decision, the basis on which he was later refused long term residency would not have existed and he would have been granted long term residency. Cooke J considered that too speculative a proposition to justify certiorari:

“A finding to that effect would require the court to determine that the time actually taken was so unreasonable as to have been unlawful and to fix a period that would have been reasonable; and then to find that the decision would necessarily have been different had it been taken at that point.”

Cooke J concluded that it was “impossible to make such a determination”. That conclusion was drawn “Having regard to the evidence as to the volume of such applications received and processed by the Department and to the absence of any basis for concluding that the resources applied for that purpose were so manifestly inadequate as to amount to a breach of duty on the part of the respondent”. Also, no case could be made that the Department ought to have decided the first application before he ceased to qualify for long term residency 4 months after he had applied for it.736 It does seem that Cooke J’s finding against certiorari was fact-specific. Also, it was concerned with a relatively simple form of an administrative law

decision – and one involving no objectors. I do not see it as authority that, in the inevitably iterative, highly complex, highly regulated, multifactorial, multi-party process at issue here, in which an early and proper preliminary view had been that a grant of the licence was likely, it should nonetheless have been refused in order to make an expeditious decision.

Dietacaron 2004

432. In Dietacaron737 the Board repeatedly adjourned an oral hearing part-heard in May 2003 pending receipt of modified development proposals it had sought of the developer. Before the Board had decided the appeal Dietacaron, an objector to the proposed development, alleged breach of the Board’s duty under

735 Saleem v The Minister for Justice, Equality and Law Reform, Respondent [2011] 2 IR 386.

736 This represents a simplified account of the facts in Saleem but suffices here.

737 Dietacaron Ltd v An Bord Pleanála and Everglade Properties Ltd, South Dublin County Council, Quarryvale Two Ltd and Quarryvale Three Ltd [2005] 2

I.L.R.M. 32. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

s.126 PDA 2000 and sought certiorari quashing the Board’s notice deferring consideration of the appeal and inviting revisions to the proposed development.

433. As earlier observed, s.126 is very similar to s.56 of the 1997 Act.

• It first imposes a general duty “to ensure that appeals and referrals are disposed of as expeditiously as may be … to ensure that … there are no avoidable delays at any stage in the determination of appeals and referrals.”

• It next imposes an objective to determine appeals within 18 weeks.

• It permits the Board, where it is not “… possible or appropriate, because of the particular circumstances of

an appeal” to determine it within the 18 weeks, to extend the period by notices stating, “the reasons why it would not be possible to determine the appeal or referral within that period” and specifying “the date before which the board intends that the appeal or referral shall be determined”.

Nine deferrals had extended the 18-week period from January 2003 to the point of delivery of the revised plans in February 2004.

434. Quirke J accepted that s.126 was intended primarily to benefit applicants for planning permission Quirke J did not accept that it was to benefit only such applicants and he rejected the developer’s argument that Dietacaron lacked locus standi on the issue of delay. But he held that the weight which the court will attach to a complaint of delay by the Board in its procedures will normally take into account the nature and extent of the applicant's interest in those procedures. In this he seems to me to take the same view as that which I have expressed above as to the application of the third principle stated by Lowrey LCJ in Dolan – that the party primarily interested specifically in expedition by ALAB (as opposed to in the substantive outcome) is MOWI.

435. Quirke J described the objective stated in s.126 to determine appeals within 18 weeks as “heavily qualified”. In my view the same can be said of s.56 of the 1997 Act.

436. On the facts Quirke J considered it common sense that the scale and proportions of the proposed development would have rendered it difficult for the board to determine the appeal within the normal timeframe applicable in routine planning appeals and he was not satisfied that the Board’s delay had been “avoidable”. He continued:

“I am satisfied on the evidence that the applicant has been in no way prejudiced by the deferrals of which it complains and will have ample opportunity to participate fully in all stages of this appeal and in the planning process relevant to it. I am satisfied also that public participation in this process has not been in any way undermined by any delay on the part of the board avoidable or otherwise.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

The relief which has been sought by the applicant in these proceedings is discretionary in nature.

Even had there been technical breaches by the board of individual provisions of s.126 of the Act of 2000 (and I do not find that there have been such breaches), I do not think that, on the facts of this case, it would have been appropriate for the court to exercise its discretion to quash the board's

decision on grounds of such technical breaches, or on grounds of the delay complained of by the applicant.”

In my view, viewing public participation as a predominantly a right of, rather than a burden on, a participant, and taking an overview of this case, the first paragraph above applies in the present case.

437. It must be said that the delay in Dietacaron was far shorter than the delay in the present case. But the case is notable for

• its invocation of the discretionary nature of relief in judicial review,

• a willingness to exercise that discretion against relief for technical breaches of s.126,

• its attribution of significance to the absence of evidence of prejudice to Dietacaron and

• a primary concern that Dietacaron’s rights and public rights to participate in the planning process would have been vindicated.

• a willingness to exercise that discretion against relief where Dietacaron’ rights and public rights to participate in the planning process would have been vindicated.

The S.56 Notices – Delay in Serving the First and Lack of Reasons

438. Clearly, the first s.56 notice, served on 8 April 2016, was served out of time – about 6 months after the appeals were lodged738 and about 2 months after the expiry of the 4-month time limit set by s.56(3) of the 1997 Act. ALAB has not satisfactorily explained that breach. ALAB’s submissions explain its plea on this issue: it submits that as IFI plead that by and on its breach ALAB “lost jurisdiction” to determine the Appeal, IFI was obliged to, but did not, object at that time. Instead IFI allowed the Appeals to continue at cost and expense to the many involved – ALAB, MOWI, and the objectors. It asserts that IFI by failing to protest or act at that time and by its continued participation in the Appeal, acquiesced in its continuance despite any defect of jurisdiction.

439. Nor has IFI shown any prejudice driving from specifically the delay in issuing the first s.56 notice. Assuming for the sake of argument that IFI is correct – that the first s.56 notice was invalid as too late and that from the expiry of the 4 month period without service of a s.56 notice – it follows that IFI could, at that time, have started judicial review to quash the s.56 notice and for a declaration that ALAB had lost

738 On various dated in early/mid-October 2016.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

jurisdiction to determine the Appeal. Assuming such relief, all concerned would have been, as matters turned out, saved about 5 years of considerable trouble, effort and, not least, expense – and, indeed, more of the same thereafter in these proceedings. That has amounted to a very considerable prejudice to all involved save IFI – who ex hypothesi cannot allege prejudice which they could and should have avoided. Of course IFI could not have predicted the duration and quantum of that trouble and expense as matters turned out, but I do not think that affects the question. There is no doubt but that it must have been apparent to IFI at that point in the process, which had hardly begun, that a long, difficult and expensive road lay ahead –

even if not as long, difficult and expensive as it turned out to be. The point is that IFI’s case is that from 2016

ALAB lacked any jurisdiction whatsoever to consider and decide the Appeal. Yet IFI did not protest at the time or take any action until by these proceedings in 2023.

440. MOWI cites Podariu739 in observing that:

• In general terms, ALAB had jurisdiction to determine the appeal and to extend the time within which it would do so – this is not a “no subject matter jurisdiction” case.

• IFI acquiesced the exercise of those jurisdictions – in any defective exercise thereof – by not challenging their exercise at or about the time of that exercise.

441. Podariu was not a delay case. The Fitness to Practise Committee of the Veterinary Council acted ultra vires in adding, during a hearing before it, an additional complaint to a notice of inquiry against a veterinary surgeon, of which complaint it found him guilty. It was held that by failing to object and by accepting that the committee had the power to amend the notice of inquiry, the vet had acquiesced in its dealing with the new complaint and so was estopped from arguing that the Committee had acted ultra vires.740 Hogan J observed that an entirely new jurisdiction cannot be created by estoppel.741 But the committee had not lacked jurisdiction “in this particular sense” of “no subject matter jurisdiction”. The statutory preconditions to jurisdiction742 in question existed fundamentally as protections for the vet – which protections he could waive and had waived.

442. IFI seeks to distinguish Podariu on the basis that, on the expiry of the 4-month period limited by s.56 of the 1997 Act, ALAB simply fell out of jurisdiction to serve a first notice to extend time for its determination of the appeal and thereby fell out of jurisdiction to determine the appeal.

739 Podariu v The Veterinary Council of Ireland [2018] 3 IR 124.

The finding was quashed on other grounds.

741 Citing the law as to estoppel by conduct as illustrated by a trilogy of leading Supreme Court decisions from the 1970s: In re Green Dale Building Co

[1977] IR 256, Corrigan v Irish Land Commission [1977] IR 317 and The State (Byrne) v Frawley [1978] IR 326.

Giving as examples that a decision of the Medical Council purporting to sanction a veterinary surgeon would be wholly void and ineffective, even if the

veterinarian in question had somehow submitted to the jurisdiction of that council and that District Court cannot exceed its own geographical limitations

by purporting to deal with offences which had not been the subject of a complaint made within the appropriate District Court district.

742 That the complaint have been referred to the Committee by the Preliminary Investigation Committee of the Veterinary Council, whose function was

to decide if a complaint merited such referral. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

443. MOWI submit that it is relevant that IFI’s submission that the late service of the first S.56 notice deprived ALAB of jurisdiction in the Appeal, would either leave the Minister’s determination standing and unappealable – as no decision on the appeal would have activated the annulment of the Minister’s decision by virtue of s.40(6) of the 1997 Act – or would leave a vacuum in which the status of the licensing application would be in limbo. The latter is an unacceptable conclusion and the former would mean the deprivation of all appellants of their statutory right of appeal by reason of, as MOWI say, a relatively minor error by ALAB, by which no-one claims to have been prejudiced – and an error not of the Appellants’ making. In the latter respect, I respectfully adopt the reason of the Supreme Court in Irish Refining.

444. While I say more on the issue below, I agree with ALAB and MOWI, having regard to the caselaw

generally but in particular to Dietacaron, Podariu and Irish Refining, that ALAB’s failure to issue its first s.56

notice within the 4-month time limit, though rendering it in breach of s.56 did not render that notice ineffective and did not deprive ALAB of jurisdiction to decide the appeals.

445. From here, I will address also the allegation, which I accept, that some s.56 notices, in breach of s.56(3), gave no meaningful reasons for the extensions of time in question.

446. This is not a “no subject matter jurisdiction” case. IFI acquiesced in all of the s.56 notices as to the fact of their service, as to any want of reasons therein and as to the delay resulting – not just in the sense that it did not seek to quash them when they issued but in that it did not object to them when they issued and until after the substantive decision was made in the Appeal. If anything, Mr Podariu’s was stronger than the present case. Yet he was estopped from complaining of the addition to the notice of inquiry against him of an accusation of professional misconduct in the form of dishonesty. And as to the first s.56 Notice, IFI’s case is that, once the 4 months expired, ALAB’s jurisdiction to extend time for its decision and hence its jurisdiction to decide the appeal was lost irremediably. On that view,

• IFI should have sought judicial review then and would not have been shut out for prematurity in not awaiting the substantive outcome of the appeal.

• had IFI done so, over 5 years of pointless public administration and very considerable public and private resources could have been saved. The same is true in degrees decreasing with time as to later s.56 Notices allegedly irremediably invalid for want of reasons.

447. I hold that IFI by its failure to object, acquiescence and continued participation in the licensing process, waived any objections to the validity of the s.56 notices as to the fact of their service, as to any want of reasons therein and as to the delay resulting. In addition, I hold that as to the first s.56 Notice of 2016 which, it is said, ALAB would have irremediably lost jurisdiction to determine the Appeals, IFI should

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

have commenced judicial review at that time and was out of time to do so in 2023. I reject the challenge on this account. I also accept ALAB’s submission that SWI’s objection to delay in October 2018 was general and not related to the validity of the s.56 Notices.

448. In any event and if I am wrong in the foregoing view, having regard to Dietacaron and in the absence of evidence of specific prejudice to IFI, I exercise my discretion against granting relief on this account. Leaving aside the issues of bias, the sequence of events clearly establishes in general that IFI had ample opportunity to participate in the decision-making process. It can have no complaint in that regard to the

conclusion in September 2017 of the oral hearing at which it was heard. Thereafter, it was sent updated materials for comment in April 2018743 and was sent the sEIS in November 2018. In December 2018 it submitted to ALAB that the latter was deficient in assuming sea lice were neutrally buoyant. In September 2020, ALAB sent IFI for comment both MOWI’s NIS744 of July 2020 and MERC’s AA report of September 2020. IFI’s only specific complaint is that it was not sent the O’Toole report of May 2021 on the FwPM for comment. However that can be considered as a discrete issue. Similar observations can be made as to the other Applicants.

449. I cannot see that any of these alleged errors by ALAB should, in justice require that the valid appeals of 14 Appellants, including ALAB’s, should go undecided and thereby create a limbo in which the Minister’s decisions would wander. Nor does the Act require such an outcome.

Delay Generally & the Proper Remedy

450. Almost exactly 10 years elapsed from MOWI’s application to the Minister to ALAB’s determination of the Appeal. Of that 10 years, 5 years and 8 months was spent in ALAB. While the total delay here was not as egregious as that in Diesel, overall it was of the same order as that in Lafarge Redland. It seems to me that, prima facie, the same general observations apply as were made in both those cases. Whatever the complexities of the case, such a delay tends to bring the licensing scheme into disrepute and must be regretted.

451. In principle one can readily see that public and other participants, who may oppose a proposed development but doubtless have much else to do with their time and resources, should not be enmeshed in demanding, stressful, time-consuming, resource-consuming and tiring licensing processes for any more than a reasonable time – and certainly not for the 10 years of their involvement in the licensing process in the

743 ‘Supplementary Briefing Notes by Dr. Saunders’ dated 24th November, 2017 assessing potential impact on otters, by Alex Coram of St Andrews Marine

Research dated February, 2018 assessing potential impact on the common seal – both as a species of interest in the nearby Glengarriff Harbour and

Woodland SAC and by Dr Tom Gittings dated 5 February, 2018 assessing potential impact on wild birds in nearby SPAs.

744 Natura Impact Statement within the meaning of the Habitats Directive.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

present case. No evidence was tendered of particular prejudice along those lines and by reason of the delay in this case. I was invited to infer general prejudice from the passage of the 10 years – for 5 years and 8 months of which ALAB were responsible. I think such an inference is reasonable and one may infer, at least, some expense accordingly.

452. However, it is very difficult indeed to weigh such prejudice as a discrete increment on what would have in any case been the inevitable burden of an undeniably complex case which was never, on any view, going to be decided within the statutory target of 4 months – or even close to it. Public participation can be very burdensome. That, at an individual level, each objector can choose his/her level of involvement is some, but not a complete, answer to that observation. Some things just are complex and defy simplification. That was undoubtedly so in this case. As Clarke J said in Connelly:745

“ ….. the issues involved may themselves be complex. The reasons put forward either in favour or against a proposed project may involve detailed scientific argument or complex calculation. If such issues arise then it will inevitably be the case that the reasons themselves may be complex and scientific. Where a party wishes to engage with the planning process in a case which raises complex issues of that type ….. then it is inevitable that the party concerned will also have to engage with such matters if any part of their opposition or challenge derives from such complex or scientific questions.”

453. It must be said that much of the delay in this case derived from ALAB's perception that the information provided by MOWI was inadequate. It is not for me to question that perception. I need not review that issue comprehensively but notably,

• Dr Saunders’ interim report of December 2016 identified “notable omissions” from the EIS (though the EIS had passed muster with the Minister).

• the 2017 recommendations by the chair of the oral hearing derived largely from a view that the information to hand was inadequate and notably suggested that an sEIS be required of MOWI.

• Dr Gittings, advisor to ALAB, in February 2018 considered the EIS flawed as to wild birds.

• the ALAB minute of 31 January 2020 and the s.47 Notice of 23 March 2020 combine to amount to significant criticism of whatever NIS MOWI submitted on 18 October 2019 and MOWI required 3 attempts to July 2000 to satisfy ALAB in that regard.

• Even in December 2020, Dr Saunders’ Final Report746 considered that the cumulative effect of the proposed Salmon Farm with other salmon farms in the bay as to sea lice had not been specifically considered. No doubt the Applicants will take limited, if any, solace from the insight that MOWI’s part in this delay has doubtless cost it dearly.747 But whatever about its causal contribution to the delay, clearly, ceteris paribus,

745 Connelly v An Bord Pleanála, [2021] 2 IR 752 §74.

746 p64.

747 Some sense of the losses involved and profits in prospect can be gleaned from the Saunders report of December 2020 which, at p49, cites MOWI’s

projections of harvest values of €14,234,500 in year two and €15,088,040 in year four, with as expected profit margin of 10% to 15% of gross turnover.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

delay did not enure to its financial benefit. Nonetheless, ALAB was in control of and is, ultimately, responsible for the fairness and duration of the process.

454. As the review above discloses, the decided cases on delay in administrative and quasi-judicial processes are almost invariably brought by applicants for licences, benefits, declarations of status, permissions and the like or by persons seeking to halt investigations or prosecutions on foot of complaints made against them. And, in the cases as to licences and the like, in the ordinary way, they seek mandamus requiring that the decision be made. Dietacaron is an exception but was prosecuted in advance of a substantive decision on the planning appeal at issue in that case and, of course, the applicant for judicial review, an objector, failed in that case.

455. No authority has been cited to me in which opponents of development consents, in judicial review proceedings initiated after a decision was made, impugned for delay the final and substantive result of the statutory process. It can often be said of certiorari of a permit or licence on any ground, even absent remittal, that the disappointed applicant can apply again and that is not often, if ever, a ground for refusing certiorari. It would however be paradoxical that, if a decision otherwise valid is quashed at the instance of an objector to the permit/licence application and is quashed only for egregious delay by the decision-maker in making it, the prospect of a renewed or remitted application is one in substance of exacerbating the very delay and the very burden of public participation of which the applicant for judicial review had complained.

456. A useful framework for analysis is found in the following propositions from the Sweetman Applicants’ submissions which I set out below – reordered slightly and numbered for ease of reference – and which will address sequentially:

i. “It is not ALAB’s function to help the developer over the finishing line.

ii. “If the information furnished by the Developer is so deficient as to take years to remedy, then the application should be refused.”

iii. “The consequence of this was that the Appellant members of the public were embroiled in the process for many multiples of the 4-month timeframe that the Oireachtas contemplated. That is an unfair demand on the resources of the public.”

“Article 9(4) Aarhus and Article 11(4) EIA Directive require that the appeal procedure be “fair, equitable, timely and not prohibitively expensive”.”

iv. “That does not prevent the developer submitting a further and properly constituted application.”

457. Proposition i., though dramatically phrased and not without merit, seems to me too absolute. I do not see error, at least generally, in a decisionmaker who has formed a provisional view in principle that a proposed development is broadly acceptable seeking further information which might render development consent possible via a decision substantively correct in all the circumstances. That seems to me at least one of the purposes underlying further information procedures. The absolute view would require that all SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

development consent applications be fit for grant immediately they are made, would turn development consent applications into obstacle races for applicants requiring refusal of permission at every failed hurdle even for minor and reparable deficiency and unjustly lead to a needless multiplicity of development consent applications as applicants seek to recover from defeats in earlier applications. It would also disable development consent applicants from engaging with and adapting to issues raised by objectors. None of this would enure to good public administration, which is the project of judicial review – (Saleem, Murtagh, ETI748).

458. Proposition ii. has more general merit but, in my view, raises issues of degree and requires to be applied in context of the circumstances of the particular application. In my view, decision-makers can be allowed leeway for reluctance to force applicants for development consents to start again when remedial information may solve a problem in the application and statutory powers to seek and/or receive it are to hand.

459. Nor do I see that, as the Applicants also suggest, decision-makers are confined to information provided by or to seeking such remedial information from the applicant for development consent. The very purpose of prescribing consultees is to make their relevant data and other information, expertises, perspectives, and opinions available to the decisionmaker. As long as the decisionmaker’s requests are properly respectful of consultees’ independence, I see no reason why, within the statutory regime, a decisionmaker cannot seek their contribution with a view to discerning whether a licence should be granted. As a general proposition, this must be so not least as:

o Recital (7) of the EIA Directive requires EIA “on the basis of the appropriate information supplied by the developer, which may be supplemented by the authorities and by the public likely to be concerned by the project in question.”

o The obligation in EIA of a private-sector applicant for a licence to provide “appropriate” information is to do so “within the limits of what may reasonably be required of a private operator” – EIA Directive Article 5(1)(b), Abraham,749 IL v Land Nordrhein-Westfalen,750 and Coyne.751 Of course, what may reasonably be required will be particular to the substance of the development consent application and the applicant in question: ceteris paribus, the more complex and/or environmentally sensitive the project, the more commercial the venture and/or the more well-resourced the development consent applicant, the more may reasonably be required. But given the obligation of

748 Saleem v Minister for Justice, Equality and Law Reform [2011] IEHC; Murtagh v Kilrane [2017] IEHC 384 §14 citing Graham v Police Service

Commission [2011] UKPC 46 in turn citing R v Lancashire CC ex p. Huddleston [1986] 2 AER 941: Environmental Trust Ireland v An Bord Pleanála [2022]

IEHC 540 §232.

749 Case C-2/07 Abraham v Wallonia, Opinion of Kokott AG of 29 November 2007, [2008] ECR I-1197.

750 Case C-535/18 Land Nordrhein-Westfalen, Judgement of 28 May 2020, §82.

751 Coyne v An Bord Pleanála & Enginenode [2023] IEHC 412, §129.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

comprehensive EIA – Namur-Est752 – the limit on what may be required of a developer necessarily implies that necessary information may be sourced other than from the developer.

o consultees, many of whom are public authorities and/or specialist agencies, are often able to supply relevant data and other information, expertises, perspectives, and opinions not reasonably to be required of the aspirant developer.

460. I have considerable sympathy with proposition iii. But assuming the objectors’ specific complaint of “unfair demand on the resources of the public” is justified the regrettable practical fact is that certiorari cannot recover their spent time, effort and resources. Melding practicality with principle, Christian is authority that “the overriding principle behind any remedy in civil proceedings should be to attempt, in as clinical a way as is possible, to undo the consequences of any wrongful or invalid act but to go no further.”753 Ceteris paribus, assuming the ALAB’s decision is otherwise legally correct and isolating a remedy to the objectors’ specific complaints of delay in the process and a resulting unfair burden of public participation, certiorari to deprive MOWI of the benefit of a substantively correct licence would be punitive rather than remedial – and punitive of a party other than the body primarily responsible.

461. On the other hand, declaratory relief, in formally recording illegality in the expectation that decisionmakers will respond in their practice accordingly, has an important role in achieving what is a fundamental value of and the strategic project of judicial review – the promotion and maintenance of the highest standards of public administration and public confidence therein (Saleem, Murtagh, ETI, O’Lone 754). Indeed, so firm is that expectation that, often, mandamus does not issue when otherwise it might – a declaration is assumed sufficient to produce the required effect. As Humphreys J said in Save Cork City,755 a declaration of non-compliance with legal requirements “puts down a formal marker so that any future non-compliance can be assessed by reference to whether there has been a pattern of action or inaction that amounts in effect to disregard of the legal obligations concerned.”

462. Proposition iv. – that the developer may make a further licence application – illustrates the very considerable risk of throwing good resources of all involved, MOWI, objectors and public bodies alike, after bad. The logic of certiorari of a decision ex hypothesi otherwise correct, as a remedy for, specifically and discretely, the imposition of an unfair burden of public participation, is that there would be both no remittal and no decision by the decision-maker on the merits of the application and so would be the prospect of yet more burden of public participation and expenditure of private and public resources in a new and inevitably highly repetitive, and the extent of such repetition, wasteful, application.

752 Case C-463/20 Namur-Est Environnement ASBL v Région Wallonne, Opinion of Kokott AG of 21 October 2021 §§42, 44, 46, 49, 50, 57, 59, 60, 72.

753 Christian v Dublin City Council, [2012] IEHC 309, Clarke J.

754 Saleem v Minister for Justice, Equality and Law Reform [2011] IEHC; Murtagh v Kilrane [2017] IEHC 384 §14 citing Graham v Police Service

Commission [2011] UKPC 46 in turn citing R v Lancashire CC ex p. Huddleston [1986] 2 AER 941: Environmental Trust Ireland v An Bord Pleanála [2022]

IEHC 540 §232. O’Lone v. An Bord Pleanála [2023] IEHC 136, Humphreys J, §§41 & 42.

755 Save Cork City Community Association CLG v an Bord Pleanála [2021] IEHC 509 §65.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

463. Of course, it would not be a precise repetition. A new application may differ from its predecessor. All protagonists would be better informed (inter alia by l’esprit de l’escalier) and forearmed for the fray. Some existing parties may not engage – new parties may. New arguments may be made and old ones abandoned. There is at least a prospect that relevant considerations of environmental law, policy and science would have changed. It is even possible that a new application would not be made, though MOWI’s dogged pursuit of

this project since at least 2009 – when it retained Dr Bass and initiated a pre-EIS ‘scoping study’756 – strongly

suggests otherwise, to the point that I attribute the possibility no great weight. But these factors, even if viewed as potential benefits, are, in the end, incidental to and not the proper remedial aim of any certiorari which might be granted. The very fact that the Impugned Aquaculture Licence is granted for a maximum of

10 years (and is subject to management conditions compliance with which may impose ongoing requirements changing with time and circumstance in that period) illustrates both that revisiting its grant will be required at a time considered appropriate and that ALAB’s judgment is considered good until that time – for a decade – despite the potential for changed circumstances in that decade.

464. Accordingly it is my view that if this is a case of unreasonable delay by ALAB in deciding the Appeal, the appropriate remedy is a declaration to that effect.

Unreasonable delay – Decision

465. As to whether ALAB was actually guilty of unreasonable delay, it is tempting to say of the period of close to 6 years to decide the Appeal that it is a case of res ipsa loquitur. Certainly, and as I have said, whatever the complexities of the case, such delay tends to bring the licensing scheme into disrepute and must be regretted. That it was superimposed on a lengthy time awaiting Ministerial decision in a total of 10 years between the making of the application and the decision of the Appeal exacerbates that regret. While

the Ministerial delay was not ALAB’s fault, it should have heightened ALAB’s sensitivity to delay – though not

at the expense of proper decision-making.

466. A prohibition on unreasonable delay must not be confused with a demand for relentless and perfect efficiency and expedition. I will attempt briefly to describe what seem to me time-critical and timedetermining events in the process before ALAB. But I emphasise that it will perforce omit considerable and

756 Sweetman Grounds §42.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

more or less continuous activity on the Appeal file. Inter alia, ALAB considered the file at its meetings more or less monthly.757 This account also ignores some controversies I address elsewhere in this judgment.

467. ALAB received the 14 Appeals in October 2015. It immediately sought to retain, by tender, technical advice and also invited each appellant to comment on all the other appeals. This prompted, inter alia, submission of the RPS Report 2015 and complaints of conflicts of interest affecting ALAB members. ALAB very properly took those complaints seriously and took legal advice. It decided in April 2016 that there was no such conflict and it retained Dr Saunders as its technical advisor. By this time ALAB had considered the file on 5 occasions. I see no unreasonable delay to this point.

468. By late August 2016 Dr Saunders had indicated,758 with caveats but significantly and unobjectionably, that “Overall, almost all of the evidence indicates that the licence should be granted” and had recommended an oral hearing.759 His caveats resulted in considerable s.47 requests to MOWI,760 IFI, the Marine Institute & NPWS which yielded commensurately considerable responses, which informed Dr Saunders’ interim report of December 2016. It recommended761 that final licence refusal or approval should be determined on further consideration, after an oral hearing, of issues of the potential lice risk to the salmonid and freshwater pearl mussel populations in the Dromagowlane/Trafrask River system. Though he doesn’t precisely say it, it does seem clear that he considered that the licence could be granted if ALAB was reassured on these issues.

469. ALAB properly accepted his recommendation of an oral hearing – which started in July 2017. It might

– even should – have started more quickly but I can’t say the delay was culpable. It was adjourned partheard when it became apparent that the RPS Report 2015 had not been circulated to participants. While unfortunate, that adjournment was to ensure, and was the cost of, fair procedures. ALAB continued to address the file at its monthly meetings through 2017 – inter alia dealing with its requirement that the Minister produce the reports on the 2014 salmon escape. It proved impossible to reconvene the oral hearing until September 2017. The Chair’s report thereon came to hand in November 2017. As has been seen, it recommended that ALAB should issue an aquaculture licence – conditional on the outcomes of the following:

757 ALAB’s Impugned Determination records that it “considered the appeal at its meetings on the 20 October 2015, 24 November 2015, 19 January

2016, 7 March 2016, 5 April 2016, 13 September 2016, 22 November 2016, 10 January 2017, 9 February 2017, 7 March 2017, 5 May 2017, 13 June

2017, 19 July 2017, 25 August 2017, 31 October 2017, 9 November 2017, 13 December 2017, 19 February 2018, 1 May 2018, 12 June 2018, 28 August

2018, 9 October 2018, 14 November 2018, 11 December 2018, 22 January 2019, 26 March 2019, 30 April 2019, 15 May 2019, 25 June 2019, 9 October

2019, 14 November 2019, 10 December 2019, 31 January 2020, 26 February 2020, 19 March 2020, 22 April 2020, 15 May 2020, 11 June 2020, 9 July

2020, 6 August 2020, 4 September 2020, 8 October 2020, 5 November 2020, 30 November 2020, 10 December 2020, 12 January 2021, 5 February

2021, 2 March 2021, 1 April 2021, 29 April 2021, 28 May 2021 and 24 June 2021.”

758 Exhibit MCC2 T53 to the Affidavit of Mona-Claire Costelloe.

759 Exhibit MCC2 at pages 682 to 689 to the Affidavit of Mona-Claire Costelloe.

760 Seeking information on Salmonids & Freshwater Pearl Mussel in the Dromagowlane/Trafrask River, well boat discharges, cage design detail, an

updated assessment of the effects of the proposed revised cages, sea lice treatments and the presence of Dublin Bay Prawn in the licensed area.

761 10.0 Recommendation of Technical Advisor with Reasons and Considerations; 11.0 Draft Determination Refusal /or Grant.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

ALAB should:

• require a Supplemental EIS (“sEIS”) on o sea lice risks to wild salmonids migrating from/to the Dromagowlane and Trafrask Rivers, o any resulting implications for local Freshwater Pearl Mussel (“FwPM”) populations o potential impact of salmon farm waste on water quality, having particular regard to the required maintenance of WFD ‘good water status’.

• do desk-top studies, which may indicate the need for supplemental AA screening of risks to: o The otters of the Dromagowlane and Trafrask catchments, including the potential impact of declining wild salmon stocks, o The common seal population of the Glengarriff Harbour & Woodland SAC, and o Birds in nearby SPAs.

• “make every effort to consider the potential impacts of large-scale farmed salmon escapes.”

470. Importantly, the Chair’s underlying, if significantly conditional, recommendation was that a licence be granted. In my view, ALAB cannot be faulted for not calling a halt to the licensing process at this point. It properly set about recommended processes which would inevitably take appreciable time. The sEIS would take time to prepare and would predictably result in further interaction with prescribed consultees and public participation. The AA screening had obvious potential to result in AA – which would require preparation of an NIS and yet further interaction with prescribed consultees and public participation. In each case – EIA and AA – it was already apparent that public participation would be on issues of vehement controversy and interaction with prescribed consultees and the public was likely to be iterative. The resultant documents would require consideration in appreciable detail by ALAB’s technical advisor and by ALAB itself. That these matters would require considerable time to process was quite predictable. In the event, all these features came to pass – though their full extent was not doubt unpredicted at that point.

471. Expert desk-top reports on seals, otters and birds, the sEIS and MOWI’s Pest Management Plan were

all to hand by mid-April 2018 – just after s.46 requests had issued to all as to risks to seals, otters and birds.

Dr Gittings’ desk-top study concluded that the Minister’s EIA had been flawed as to birds and also recommended formal AA screening. In March 2018, the MI opined that AA as to birds was not required. At its June 2018 meeting, ALAB decided to seek legal advice and to ask Dr Saunders to review the considerable additional material now to hand and to consult the NPWS by s.47 Notice as to AA and EIA issues related to birds and the FwPM once Dr Saunders’ review was to hand. In its August 2018 meeting ALAB was awaiting Dr Saunders’ review and in any event decided to publish the sEIS and invite submissions. In particular, ALAB decided to extend its decision date to 30 June 2019. This was, of course, a very lengthy extension – arguably surprisingly so. But it is minuted as deriving from ALAB’s review of the timetable and as ALAB was at the time advertising the sEIS for public consultation and the position as to AA remained unclear, it cannot be deemed unreasonable. A s.56 letter issued accordingly on 31 August 2018.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

472. I should mention that in this interim and in response to that s.56 letter of 31 August 2018, SWI wrote to ALAB on 1 October 2018 to the effect that “ALAB cannot keep on adjourning and deferring a decision indefinitely” on AA and was not now able to conclude a certainty of no impact on any European Site or protected fauna or flora. It said, “While there may be a statutory basis for the deferral of a decision, the licence Applicants should not be indulged with continuous adjournments.” SWI submitted that the licence application should be immediately refused “without further delay”. By its acknowledgement of a statutory basis for deferral, it is clear that SWI considered that ALAB had a discretion in the matter but urged its exercise by way of refusal of the licence. On 9 October 2018, ALAB directed a letter to SWI under s.41(3)762 of the 1997 Act refusing to consider the unsolicited letter of 31 August 2018. While I am not convinced that ALAB’s was a proper response to what was in essence a legal submission, nothing seems to me to turn on that given ALAB’s autonomous duty to decide the licence application within a reasonable time.

473. ALAB appears to have awaited the NPWS response before publishing and advertising the sEIS. It was dissatisfied with NPWS’s initial response that it had no comment to make and NPWS gave a more substantive response in mid-November 2018 advising consideration of the FwPM in EIA and that AA should include risk to gannets of the Bull & Cow Rocks SPA. The sEIS was published and advertised later in November. In mid-December 2018 ALAB, having considered the NPWS response and legal advice thereon, decided on the formal AA screening as to birds recommended by Dr Gittings. It also decided on s.47 notices to IFI and NPWS as to the FwPM. IFI replied per Dr Gargan disputing, citing academic authority, the sEIS

conclusion of no sea lice risk to wild salmonids – notably on the basis that sea lice are positively buoyant – not neutrally so as the sEIS modelled. Accordingly, IFI asserted, their dispersal pattern is less determined by hydrography than by wind. It is obvious that this required careful consideration. Public and consultee’s submissions, including by applicants in these proceedings, were in by mid-January 2019. In February 2019, the MI issued a reassuring s.47 reply (to the NPWS and IFI submissions) as to risk of lice-induced mortality in wild salmonids and risk to the FwPM by loss of salmonid larva hosts. In April 2019, Dr Crowe recommended AA as to specified birds763 and in May 2019 ALAB resolved to require an NIS of MOWI and to seek tenders for consultants to assist ALAB in doing an AA. Of course, AA now had to be done to the high standard of reasonable scientific certainty. On ALAB’s direction, s.56 Notices issued on 26 June 2019 extending the

decision deadline to 31 March 2020. Again, this may seem surprisingly long, but, not least given the AA process just starting, I cannot see that it was so long as to be unreasonable.

474. To this point, while it may be suggested that the various workflow streams might have been better managed to advance in parallel, that can be very difficult to manage in a process in which the views of one participant are often in sequential response to the views of another – or, indeed, others – and each

762 S.41(3) reads: Without prejudice to section 46, an appellant shall not be entitled to elaborate in writing on, or make further submissions in writing in

relation to, the grounds of appeal stated in the notice of appeal or to submit further grounds of appeal, and any such elaboration, submissions or

further grounds of appeal received by the Board shall not be considered by it.

S.41(4) reads: Without prejudice to section 47, the Board shall not consider any documents, particulars or other information submitted by an appellant

other than the documents, particulars or other information which accompanied the notice of appeal.

763 Fulmar, Gannet & Guillemot.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

participant has views on multiple issues. There were no significant periods of inactivity on the file after the Oral Hearing report came in and I cannot see that ALAB can be held in dereliction of its duty of reasonable expedition to this point.

475. MOWI’s first attempt at an NIS was submitted in October 2019. ALAB had retained MERC to advise it on AA and in January 2020 presented a draft report of its view the effect that the fish farm would not adversely impact on SCI764 species or SPA765 conservation objectives. However, MERC considered that while the NIS implied as much, it should say so expressly. ALAB accepted MERC’s advice – both the implication and the requirement of expression – and in March 2020 issued a notice for additional information to MOWI under Art. 42 of the Habitats Regulations766 seeking an amended NIS. It also issued s.56 notices deferring the expected decision date to 31 August 2020. In April, it minuted its intention to put the NIS and AA report out to public consultation and in June issued another Art. 42 notice for additional information to MOWI as its second attempt at an NIS remained inadequate. After further clarificatory correspondence, MOWI’s third attempt came to hand in July 2020. To any extent this can be said to have caused delay it was from the expiry of the March 2020 deadline to July 2020. MERC gave its initial response thereto in August and its AA report in September 2020. It advised that there were no significant lacunae in its analysis, advised imposition of certain conditions and stated its reasoned conclusion that the Shot Head farm would not impact adversely on SCI species or SPA conservation objectives. In August 2020, MOWI had issued s.56 notices deferring the expected decision date to 31 December 2020 and in September 2020 ALAB invited submissions by participants, consultees and the public on the NIS and the MERC AA Report, with a mid-November deadline. ALAB met virtually767 on 30 November 2020 and decided that it could not decide the AA issues pending legal advice.768 Dr Saunders’ Final Report ensued in December 2020 – despite his criticisms of the information to hand (in particular, I consider the issue of the risk of fish escape separately) he could “offer no substantive technical reasons to refuse the current licence application.”

476. Strikingly, at this point and despite Dr Saunders’ advice and MERC’s reasoned conclusion as to AA that the Shot Head farm would not impact adversely on SCI species or SPA conservation objectives, ALAB in meetings of December 2000 and January 2021 recorded that it had to be “meticulous” and directed further review of 4 issues – as to (i) gannets, (ii) kelp, (iii) the FwPM and (iv) non-SPA birds. It accordingly

• extended the decision date to June 2021.

• made a further s.47 request of MOWI as to risks to birds.

• bespoke a further MERC report on risk to birds. In turn, MERC bespoke the DMP statistical analysis of net entanglement mortality in Gannets.

• sent Dr O’Toole on a site visit to assess and report on the issue of salmonid hosts of FwPM larvae in the Dromagoulane/Trafrask River.769

764 Species of Conservation for the protection of which SPAs are designated.

765 Special Protection Areas within the meaning of the Wild Birds Directive.

766 European Communities (Birds and Natural Habitats) Regulations 2011 as amended.

767 no doubt for Covid reasons.

768 The exhibited minutes of this meeting are redacted but see the chronology in the Affidavit of Francis Dowling sworn 28 June 2022.

769 PS description.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Bespoke a report from Dr O’Toole on the kelp issue. These reports came to hand in May 2021 and, on 28 May 2021, ALAB formally determined in AA that the project would not have an adverse effect on the integrity of European sites. ALAB’s decision to grant the Aquaculture Licence ensued at its next meeting, the following month and was published accordingly.

477. In my view and leaving aside whether certiorari would be an appropriate remedy for unreasonable delay (in my view it would not – a declaration would be the proper relief) on close analysis of the sequence of events in part set out above, ALAB’s decision cannot be quashed for unreasonable delay in making it. Such analysis bears out its defence that this was a highly complex matter and there was no period of time of any consequence during which ALAB was not properly active. I confess that I am, in a sense, surprised by my own

conclusion. I freely acknowledge that it is highly counterintuitive to refuse to find unreasonable delay where a decision statutorily but non-bindingly expected in 4 months (which was never even remotely realistic) and subject to a statutory duty of expedition, took almost 6 years to make. I remain of the general view that such an outcome is regrettable and tends to disrepute of the system. I respectfully suggest such revisitation of ALAB’s systems as may be possible despite the inherent, inevitable and time-consuming demands of, inter alia, the complexity of the marine environment, EIA, AA and the high controversy which attends such applications. However, I must decide the issue objectively on a fair and close analysis of the evidence. Without deciding that O’Keeffe irrationality rules apply, I must nonetheless bear in mind that a conclusion of irrationality represents a high hurdle. I cannot find that the Applicants have surmounted that hurdle.

478. With one exception only and by reference to the general duty identified in Holland770 to make the

decision within a reasonable time, and as informed by the applicable statutory provisions, I take the view, not without hesitation, that the time taken by ALAB to decide this matter, while both very considerable and clearly regrettable, cannot be said, on close analysis of the evidence, sequence of events and particular circumstances of the case, to have been unreasonable.

479. The exception seems to me to lie in the fact that ALAB delayed unreasonably from the making of the Appeals in October 2015 to embarking on AA Screening. While one could analyse that delay in various ways, I think the sensible analysis is that ALAB did not start to get to grips with the issue until after the Oral Hearing Report of November 2017 recommended so-called “supplemental” AA Screening. Though progress was slow thereafter, and it is striking that it was only in mid-2019 that a screening determination to perform AA was made. However, as I have said, close analysis of activity in this interim restrains me from finding unreasonable delay in that period.

480. If I am wrong in this conclusion I would, in all the circumstances of this highly complex and difficult process, apply Dietacaron in exercising my discretion to refuse to quash the Impugned Aquaculture licence

770 Holland v Minister for Justice [2023] IECA 73.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

for ALAB’s delay where the rights of all participants and public rights to participate in the process have been vindicated.

481. For reasons I have stated above, in my view the appropriate remedy in this regard should be limited to a declaration that ALAB delayed unreasonably as to AA Screening from the making of the Appeals in October 2015 to embarking on AA Screening after the Oral Hearing Report of November 2017. In my view, no other relief is appropriate.

BIAS – GENERAL, PREJUDGMENT & DELAY

482. I have already dealt with the specific allegations of bias. I return to the issue of bias at this point as it is pleaded that ALAB’s conduct of the appeal evinced objective bias as well as causing delay by

• the large number of time extensions granted,

• directing the production of further documents,

• bespeaking additional reports, thereby collecting the material to justify its pre-determined decision to grant the Aquaculture Licence.771

483. This plea, in effect, invites me to stand back and take an overview of the circumstances and delay to which IFI point and infer objective bias from them, in the round, as it were. It will have been seen that I have rejected the various specific complaints which allegedly contributed to this overall picture of bias and save I one respect I have rejected the complaint of delay. In doing so, I have, en passant, considered the more general plea. However, at the cost of some repetition but given the emphasis laid in the case on issues of bias and delay, I think I should return to it.

484. To put it colloquially, as they did, the Applicants argue that ALAB, being predisposed to, or having decided to, grant the licence, “bent over backwards” to ensure correction of deficiencies in the MOWI application. They say ALAB continually extended the appeal process, apparently with a view to gathering the information necessary to permit it to issue the licence. They point to the overall time taken to dispose of the appeal and say that ALAB, by multiple s.56,772 s.46773 and s.47774 notices, repeatedly afforded MOWI opportunities to mend its hand as to deficiencies in its application and also repeatedly compensated for those deficiencies by obtaining from others, such as prescribed bodies, information which it was MOWI obligation to supply. IFI say that ALAB gave the impression that it was effectively operating in a consultative role with MOWI, in seeking out and collecting the material it needed in order to grant the aquaculture

771 The Statement of Grounds says “determined to grant the appeal” but the correct meaning is clear in context.

772 Extending the time for decision of the appeal.

773 Power to request submissions or observations from a party or other person who has made submissions or observations.

774 Power to require documents, particulars or other information from a party or on any person who has made submissions or observations.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

licence. IFI say that an objective observer, would reasonably believe the result of the appeal was a foregone

conclusion given the repeated delays and “the fact that at every turn MOWI was afforded ample opportunity to address any issues that arose”.

485. As part of this general argument, SWI also argue more specifically that ALAB’s information-gathering phase should have come to an end with the oral hearing report, whereas thereafter significant further information-gathering in fact ensued. I have rejected this argument already and need not repeat myself here.

486. In my view and as to bias viewed on an overview, the reasonable observer would be appreciably reassured that each minute of an ALAB meeting commences with an agenda item “Conflicts of Interest/Section 31 Declarations” or similar – in which the members’ responses, and in some cases, recusals from consideration of the Shot Head file amongst others, are recorded. While not exhaustive of possibilities of bias, this agenda item can only have served to repeatedly remind ALAB members of their general duties of independence and impartiality. Many such minutes refer to the avoidance of the possibility of bias. In November 2015, as to the Shot Head file, a discussion of bias included discussion of the then-recent decision in Reid.775 It would be clear to the reasonable observer from these minutes that ALAB takes corporate governance as to such matters seriously and its doing so can only keep the necessity of the avoidance of bias and their obligations to leave any partisan baggage “at the door” live in its members’ minds.

487. The argument for bias in Orange #2776 – a case I have considered above – was, in significant respects, not dissimilar to the present one. The allegation was of objective bias infecting the decision on competitive tenders for the grant of a licence to operate a mobile phone network. Geoghegan J described the allegation as follows:

“The case on bias is a most unusual one. What was suggested was that the evaluators made so many erroneous decisions all one way, that is to say, against the plaintiff, that there must have been bias on their part and that therefore in a vicarious sense there was bias on the part of the first defendant.”

The elements of the allegation of bias in Orange #2 which are similar to the present case are that the bias alleged was:

• internal to the impugned decision-making process.

• that of evaluators of the tenders – that is to say persons in positions similar to those of the advisors to ALAB whose involvement is criticised.

• said to have comprised of “erroneous decisions all one way” against the plaintiff’s bid.

775 Reid v IDA [2015] IESC 82, [2015] 4 I.R. 494.

776 Orange Ltd v Director of Telecoms (No. 2) [2000] 4 IR 159 at 250.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Said to have included bias evident from alterations in drafting iterations of the evaluator’s report as to compared to its final report.

488. Orange succeeded at trial. But the Supreme Court unanimously dismissed its claim. As here relevant, the headnote to the report reads as follows:

“3. That a court was not entitled to infer from the establishment of errors in the impugned

decision, or the process leading to the decision, that the decision itself was vitiated by the existence of bias which could be equated to objective bias.

4. That bias could not be established from the nature of a decision made, as the allegation of bias had to be made on foot of circumstances outside the actual decisions made in the case itself. The manner in which proceedings were conducted could not in itself create a reasonable suspicion of bias.”

489. The report of the five judgments of the Supreme Court in Orange #2 (some of which I have set out earlier) is lengthy but in substance they differ little and the headnote excerpt above well-describes their effect. As described above, Barron J held that “Bias always exists before the hearing or other process” – though it “may become apparent in the course of a hearing”. While, as also described above, that view has been diluted somewhat, it remains the general rule. In particular it remains the rule as to errors in the impugned decision-making process – even a sequence of errors going all the one way. The majority in Kelly,777 held that “decisions made or taken in the course of an investigation or proceedings cannot, of themselves, give rise to an inference of bias” and O’Callaghan et al,778 is authority that bias cannot be inferred from errors in the decision-making process or from a pattern of such errors.

490. §7.1 of ALAB’s determination illuminates its rationale and method. It notes that different technical advisors provided information as ALAB requested, thereby ensuring that ALAB had access to the best available scientific advice which allowed ALAB to make informed decisions, confident that,

• in AA, the standard of no reasonable scientific doubt has been reached regarding all ecological and environmental queries as to SACs and SPAs

• a sufficient standard of information was provided for all other queries.

§7.1 displays ALAB’s consciousness of the burden of its duties – including duties of inquiry, investigation and

scrutiny, in EIA, AA and more generally and it is consistent with its aiming to be “meticulous” in such matters.779

777 Kelly v The Minister for Agriculture, Fisheries and Food, et al [2021] 2 IR 624.

778 O’Callaghan v Mahon [2007] IESC 17, [2008] 2 IR 514.

779 ALAB minutes 10 December 2020.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

491. Decisionmakers are sometimes properly criticised for failure to engage with submissions, failure to perform their investigative functions and failure to bring their critical faculties to bear on the application and other information before them. In my view, where a decisionmaker has taken a provisional and proper view that a licence or permission might be granted, it is proper for it to pursue any doubts, uncertainties and outstanding issues, as much to guard against the possibility that its provisional view may be incorrect as to confirm it as correct and in any event to seek to make a substantively correct decision. These countervailing purposes are very difficult to distinguish in practice and which one ascribes to the decision-maker can often be a matter of perspective. While, inevitably, one “side” of the argument will be dissatisfied with the result and will often be dissatisfied with how it was reached, the best means of advancing public faith in the integrity of the process is, indeed, that decision-makers be meticulous to the extent practicable. Scrutiny by

decision-makers is necessary not least as it is reassuring to the disappointed. That, of course, has to be balanced with questions of resource allocation and efficiency and expedition of the process (statutorily mandated in this case and relevant to the burden on all sides of the argument).

492. Also relevant is balancing

• the general requirement that the aspirant licensee must provide the information necessary to the success of its application insofar as it may be reasonably expected of a (typically) private applicant and do so “up front” in its application (which is part of the price of requirements of expedition),

• against the reality that in complex licensing processes, the contributions of others to the assemblage of relevant information – the public, relevant NGOs and other state agencies – is not merely proper but is necessary to informed decision-making. The tensions between these considerations are obvious – perhaps most notably as between the duty to make a sufficiently informed decision and that to make an expeditious decision. The resolution of such tensions by the decision-maker will be a matter of judgment and balance, often difficult and multi-factorial in a dynamic and iterative process and in the particular circumstances of each decision. Their resolution in favour of scrutiny and inquiry should not lightly be second-guessed by a court with the full advantage of hindsight.

493. This tension is well-illustrated in the present case in which ALAB’s decision is impugned both for want of expedition and for failure to circulate information for further public comment.780 In my view, a court should not at all lightly conclude that a decision-maker is bending over backwards to give effect to a prejudgment in favour of any particular decision as opposed to properly scrutinising the matter before it. The court should not require of decision-makers that their lines of inquiry be limited to those which may tend towards refusal of development consent or exclude inquiry which may remove impediments to development consent which they consider at least generally deserving of such consent. The presumption of validity of impugned decisions, the “East Donegal” presumption that public servants act in accordance with Constitutional duties of fairness, the duty to interpret a process so as to validate rather than invalidate a

decision and the fact that the onus of proof is on an applicant in judicial review all imply that, where events

780 Which is not to suggest that applicants in judicial review may not make contrasting pleas.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

in the decision-making process are credibly explicable in terms of the decision-maker’s anxiety to make a proper decision, then at least generally, that is the preferable interpretation. Nurendale is far from an authority to the contrary as, in that case, the facts were egregious and compelling.

494. For example, I consider that, in accordance with law as to taking, in an iterative process, a provisional or preliminary view set out above, ALAB was entitled to take a view from MERC’s presentation on

31 January 2020 that its AA was likely to favour the project. I see no reason as to the substance of what ALAB did thereafter in its route to its AA Conclusion to criticise ALAB for objective bias. On the contrary, given its entitlement as of January 2020 to take the view that its AA Conclusion was likely to favour the project, its “meticulous” pursuit of issues with the potential to confound its preliminary view is to its credit and is inconsistent with ALAB’s having a closed mind by way of a rush to premature judgment. Its course of action from 31 January 2020 betokens an anxiety to carry out a meticulous investigation based on the best available scientific evidence and expertise and with a view to legally sound conclusions.

495. For these reasons, I take the view that, even standing back and looking at the facts of this case by way of overview, the Applicants’ case in bias, prejudgment and delay is not made out save, as to delay, in terms of the declaration I have indicated above.

496. However some further alleged bias issues are addressed below.

EIA – 2011 OR 2014 DIRECTIVE?

497. The first EIA Directive was adopted in 1985 and amended in 1997, 2003 and 2009781. The second, EIA Directive, replacing the first, was adopted in 2011.782 It was amended in 2014783 resulting in what I will for convenience call the “2014 EIA Directive”.784 The statutory processes in the present cases started in 2011785 and ended in 2023.786 Hence the question arises, which of the 2011 and 2014 EIA Directives applied to the process at the date of the Impugned Decisions. ALAB applied the 2011 Directive.

781 Directives 85/337, 97/11, 2003/35 and 2009/31.

782 Directive 2011/92/EU on the assessment of the effects of certain public and private projects on the environment.

783 Directive 2014/52/EU.

784 Directive 2011/92/EU on the assessment of the effects of certain public and private projects on the environment as amended by Directive 2014/52/EU.

785 Application for Foreshore and Aquaculture Licenses.

786 Foreshore and Aquaculture Licenses issued.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

498. SWI CG5 and PS CG15 plead that ALAB applied the 2011 EIA Directive,787 contrary to Article 3 of the Amending 2014 EIA Directive and Article 3 TEU,788 even though no scoping opinion was given and the information required to be submitted pursuant to Article 5(1) of the EIA Directive was not submitted until after 16 May 2017. ALAB thereby failed to comply with Articles 1(2)(g)(iv),789 3 and 5 to 11 of the 2014 EIA Directive.790 Properly, they say, the 2014 Directive applied. The Sweetman Applicants also plead that the State inadequately transposed Article 3791 of the 2014 Directive which provides the relevant transitional provisions.

499. In reality this is turns entirely on the mistransposition ground PS CG18, as ALAB complied with domestic law – the relevant (and belated) transposing Aquaculture Appeals EIA Regulations 2019,792 Article

8 of which provides that:

“The Principal Regulations793 will continue to apply to licence applications received on or before 16 May 2017 without the amendments contained in Regulations 3 to 7 of these Regulations.”

The effect, if valid, is that the EIA Directive 2011 applied in this case.

500. Article 3 of the 2014 Directive provides for the transition from the 2011 to the 2014 EIA Directive as follows:

“1. Projects in respect of which the determination referred to in Article 4(2) of Directive 2011/92/EU794 was initiated before 16 May 2017 shall be subject to the obligations referred to in Article 4 of Directive 2011/92/EU prior to its amendment by this Directive.

2. Projects shall be subject to the obligations referred to in Article 3 and Articles 5 to 11 of Directive 2011/92/EU795 prior to its amendment by this Directive where, before 16 May 2017:

(a) the procedure regarding the opinion referred to in Article 5(2)796 of Directive 2011/92/EU was initiated; or

(b) the information referred to in Article 5(1) of Directive 2011/92/EU was provided.”

787 i.e. the 2011 EIA Directive unamended by the amending 2014 EIA Directive.

788 Treaty on European Union. Article 3 obliges the EU to work for the sustainable development of Europe based on, inter alia, a high level of protection

and improvement of the quality of the environment.

789 The Grounds actually cite “Article 2(iv)(g)”. There is no such Article and its citation is clearly a typo. The identity of the relevant Article 1(2)(g)(iv) is

readily discernible and requires that EIA include a reasoned conclusion on the significant effects of the project on the environment.

790 i.e. the 2011 EIA Directive as amended by the amending 2014 EIA Directive.

791 “” signifies derivation from the Amending 2014 EIA Directive.

792 Aquaculture Appeals (Environmental Impact Assessment) (Amendment) Regulations 2019 (S.I. No. 276 of 2019).

793 Aquaculture Appeals (Environmental Impact Assessment) Regulations 2012 (S.I. No. 468 of 2012) which transposed the 2011 EIA Directive.

794 i.e. EIA Screening.

795 Article 3 and Articles 5 to 11 set out the substantive and procedural requirements of EIA.

796 i.e. EIA Scoping.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

This means that if an EIA screening determination or and EIA Scoping determination predated 16 May 2017 the 2011 EIA Directive applied. Neither occurred here so the issue is the effect of Article 3(2)(b).

501. Article 3(2)(b) applies where the “information referred to” in Article 5(1) of the 2011 Directive was supplied before 16 May 2017 that the “developer supplies in an appropriate form the information specified in Annex IV” as elaborated in Article 5(3) – which requires that the information provided by the developer “shall include at least …..”. Substantive requirements as to content follow. In substance, and as transposed to Ireland, Article 5(1) required submission of an EIS.797 It is clear that in the present case the EIS preceded

16 May 2017. It was submitted in 2011, such that ordinarily and by Article 3(2)(b) the 2011 EIA Directive should apply. They say that where the competent authority considers that further information is required from the Developer, the information referred to in Article 5(1) has not been supplied until all that information is received. Only then can EIA be done.

502. ALAB cite both the binding effect and the rationale of the decision of Twomey J in Foley & Hayes798 refusing relief in a challenge to the EPA’s grant of a revised industrial emissions licence to a cement plant, permitting to change the fuel it incinerates. Inter alia, Twomey J rejected an allegation that the EPA had wrongly applied the 2011 Directive when it should have applied the 2014 Directive. Irish Cement’s licence application, accompanied by an EIS, had made before 16 May, 2017. Ms Hayes relied on Article 3(2)(b) of the 2014 Directive, on the fact that additional information was requested after that date and on the fact that it was not until 3 September, 2019 that the EPA deemed the EIS compliant with the relevant Regulations. That argument seems to me indistinguishable from that made in the present proceedings.

503. Twomey J cited EIA Directive Recital 39, M28799 and case C-431/92800 to the effect that:

“It is important to note that a key driver for the transitional provisions is ensuring legal certainty and that, if ‘any procedural steps have already been initiated’ under the 2011 Directive, there should be no doubt arising about which Directive applies, since it is the 2011 Directive, rather than the 2014 Directive, that will apply.”801

Twomey J held that the crucial criterion for determining whether the 2011 Directive or the 2014 Directive applied was the date on which the EIS was formally lodged and “It would defeat legal certainty if the fact

797 Environmental Impact Statement – submitted pursuant to the Aquaculture (Licence Application) Regulations 1998 and Aquaculture (Licence

Application) (Amendment) Regulations 2010.

798 Foley & Hayes v Environmental Protection Agency [2022] IEHC 470, §185 et seq. Recently upheld on this issue: sub nom. Hayes & Foley v

Environmental Protection Agency [2024] IECA 162.

799 M28 Steering Group v An Bord Pleanála [2019] IEHC 929.

800 Case C-431/92 Commission v Germany [1995] E.C.R I-2189.

801 §193.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

that the EPA sought further information after this date regarding the EIS meant that the 2011 Directive did not apply to the application.”802

504. Twomey J went on to elucidate what he considered to be the illogicality of a contrary position but I need not. It suffices that on Worldport803 principles, I consider myself bound by the decision of Twomey J unless I thought that decision clearly in error or not based upon a review of significant relevant authority, such that there are substantial reasons for believing that decision wrong. I am not of that view, despite the cases cited by the Sweetman Applicants804 and hold myself bound by Foley such that I must reject the challenge based on the view that ALAB ought to have applied the 2014 EIA Directive and not the 2011 Directive. Foley & Hayes805 requires me to hold that ALAB was correct to apply the 2011 Directive.

505. For the avoidance of doubt, I do not consider that the increase in the number of proposed cages as between the proposal to the Minister and that considered by ALAB so alters the identity of the project as to require disapplication of Foley & Hayes.806 While ALAB necessarily considered the increased number of proposed cages and they will cover a larger area, for this purpose the proposal remains in essence the same project as that proposed. Notably, the location remains and the quantum of salmon production will remain the same. I do not rule out the possibility that changes to particular projects will require a different outcome but do not consider that the changes in this case do so.

506. O therefore respectfully reject the grounds of challenge as they relate to application of the wrong EIA Directive and mistransposition of the 2014 EIA Directive.

802 §197.

803 In the Matter of Worldport Ireland Limited (In Liquidation) [2005] IEHC 189. Clarke J – “It is well established that, as a matter of judicial comity, a judge

of first instance ought usually follow the decision of another judge of the same court unless there are substantial reasons for believing that the initial

judgment was wrong. ……… Amongst the circumstances where it may be appropriate for a court to come to a different view would be where it was clear

that the initial decision was not based upon a review of significant relevant authority, where there is a clear error in the judgment, or where the judgment

sought to be revisited was delivered a sufficiently lengthy period in the past so that the jurisprudence of the court in the relevant area might be said to

have advanced in the intervening period. In the absence of such additional circumstances it seems to me that the virtue of consistency requires that a

judge of this court should not seek to second guess a recent determination of the court which was clearly arrived at after a thorough review of all of the

relevant authorities and which was, as was noted by Kearns J., based on forming a judgment between evenly balanced argument. If each time such a

point were to arise again a judge were free to form his or her own view without proper regard to the fact that the point had already been determined,

the level of uncertainty that would be introduced would be disproportionate to any perceived advantage in the matter being reconsidered. In the absence

of a definitive ruling from the Supreme Court on this matter I do not, therefore, consider that it is appropriate for me to consider again the issue so

recently decided by Kearns J. and I intend, therefore, that I should follow the ratio in Industrial Services and decline to take the view, as urged by counsel

for the Bank, that that case was wrongly decided.”

804 Both Case C-461/17 Holohan and Case C-431/92 Commission v Germany were cited by Twomey J in Foley. Case C-396/92 Bund Naturschutz in Bayern

and Others ν Freistaat Bayern [1994] ECR 1-3717 was not cited by Twomey J in Foley but is relied on here as having been relied on in Case C-431/92

Commission v Germany. Clifford / O’Connor v An Bord Pleanála (No. 3) [2022] IEHC 474 does not seem to me to demonstrate error by Twomey J.

805 Foley & Hayes v Environmental Protection Agency [2022] IEHC 470, §185 et seq.

806 Foley & Hayes v Environmental Protection Agency [2022] IEHC 470, §185 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

EIA – PROCEDURAL PURPOSE, PRECAUTIONARY PRINCIPLE AND SIGNIFICANCE OF UNCERTAINTY

507. A recurrent theme of the criticisms of ALAB’s decision is that it fails to resolve uncertainties – for example as to risk posed by farm-generated lice to wild salmonids and in turn to the FwPM. As is invariable in cases such as this, the parties have deployed the linked Precautionary and Preventive Principles to their respective ends. It is frankly impossible in a judgment already far too long, to address every last of these criticisms. I will address some but it can be said more generally that such criticisms are often misplaced attempts to argue the merits of ALAB’s decision (as to which I am not competent save in the event or irrationality) by setting up a straw man in the form of a requirement of certainty which EIA does not require.

508. As it is a basic principle of EU environmental law, laid down in the TFEU, it is undoubtedly surprising that the precautionary principle is mentioned only cursorily in the EIS and the sEIS807 does not mention it at all. Dr Saunders’ final report808 mentions the precautionary principle only once – and then only to note its invocation by Save Bantry Bay. ALAB’s Determination does not mention it at all – though its AA Conclusion Statement does invoke it.

509. IFI pleads that ALAB failed to apply the precautionary principle and s.61 of the 1997 Act809 and submits that this failure occurred on questions where there was conflicting factual and scientific evidence. IFI invokes the precautionary principle in impugning ALAB’s decision as irrational – in its written submissions impugning,

• the sea lice dispersal modelling, as allegedly incorrectly assuming that sea lice larvae are neutrally buoyant.

• selective deployment of evidence as to risk posed by farmed salmon to wild migratory salmon and transfer of sea lice from farmed salmon to wild migratory salmon.

• fundamental error in preferring “pro-development” scientific theses as to the dependency of the Trafrask River FwPM on migratory salmonid larva hosts as opposed to non-migratory brown trout.

• ALAB’s technical advisor’s acceptance that these issues remain “a matter of some conjecture, since there is

presently insufficient scientific data to make incontrovertible conclusions on current fish or mussel population status, or indeed the relationship between current salmon populations and fresh water pearl mussel recruitment.”810

IFI submits that ALAB’s conferring the benefit of the doubt on MOWI’s proposal contravened the precautionary principle. I will consider these specific issues in due course. But some general consideration of the precautionary principle is useful.

807 See below.

808 §9.1.

809 ALAB, in deciding a licence application or appeal, shall take account of the suitability of the place or waters for the aquaculture proposed, existing or

potential other beneficial uses of and any statutory status of the place or waters, likely effects of the proposed aquaculture on the economy of the area

in which it will take place, likely ecological effects and likely effects on the environment generally in that vicinity.

810 Page 107, Technical Advisor’s Final Report.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

510. In cases like this, the Precautionary Principle is often invoked in argument in almost talismanic terms and without real elaboration or analysis. It tends to be deployed in judicial review to disguise in legal garb an attempt to appeal the merits of a decision or to lower the hurdle presented by the law on irrationality. The Precautionary Principle also tends to be deployed in an attempt to evade the limits of the essentially procedural purpose of EIA. That purpose is to ensure that environmental risk is objectively assessed and that development consent decisions are informed by that assessment. But EIA does not require a particular

decision as to the acceptability or otherwise of a particular risk identified in EIA or require refusal of development consent by reason of such risk. The EIA Directive prescribes EIA of projects to which it applies but “crucially”811 “does not lay down the substantive rules in relation to the balancing of the environmental effects with other factors or prohibit the completion of projects which are liable to have negative effects on the environment” – Leth.812 As Kokott AG said in Namur-Est,813 the EIA Directive does not specify what the consequences of certain findings of the EIA are for the development consent. Even significant negative effects may be permissible. Kokott AG in Leth cited the “procedural nature of the EIA Directive” and said:

“….. the EIA Directive does not contain any rules as to which projects can at all be carried out. In particular, contrary to the view taken by the Commission, the requirement of an assessment of environmental effects cannot be understood as an obligation to conduct an appraisal of environmental effects with other factors. For that reason, the EIA Directive does not preclude the implementation of a project even in the case where the environmental impact assessment establishes that there are significant negative effects on the environment.”814

511. This seems to me consistent with the objective of the EIA Directive815 – “to ensure a high level of protection of the environment and of human health, through the establishment of minimum requirements for the environmental impact assessment of projects” “with regard to the type of projects subject to assessment, the main obligations of developers, the content of the assessment and the participation of the competent

authorities and the public”. That is to say, the EIA Directive’s contribution to a high level of protection of the

environment is essentially procedural. And, as has been said, the EIA Directive is not necessarily the

environmentally optimal legal solution and the “high level of protection” is not a trump card to render it so –

Case C-396/92 Bund Naturschutz in Bayern and Kingston.816

Kingston, Heyvaert and Cavoski, European Environmental Law (Cambridge, 2017), p388.

812 Case C-420/11 Leth v Austria [2013] 3 CMLR 13 CJEU § Leth was recently cited to this effect in An Taisce v Minister for Housing, Local Government and

Heritage and others [2024] IEHC 248. See also Browne, Simons on Planning Law (3rd ed., 2021) §14-18.

813 Case C-463/20 Namur-Est Environnement ASBL v Région Wallonne, Opinion of AG Kokott of 21 October 2021, §69.

814 Kokott AG §42. She repeated the point in case C-196/16, Comune Di Corridonia.

815 Recitals 1 and 41 (Consolidated EIA Directive 2011 & 2014).

816 Case C-396/92 Bund Naturschutz In Bayern v Freistaat Bayern, Opinion of Gulmann AG of 3 May 1994, §§66, 67. European Environmental Law,

Cambridge, 2017 p10. Both recently cited in Power v An Bord Pleanála [2024] IEHC 108, §64. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

512. I do not think that the citation in FitzPatrick817 of the view of Gulmann AG in Bund Naturschutz in Bayern818 that the purpose of the EIA Directive is, inter alia, “to have projects designed in such a way that they have the least possible effect on the environment” upsets the primarily procedural and informative method of the EIA Directive, wherein environmental considerations must be taken into account, so as to minimise effect on the environment. Gulmann AG was, it seems to me, expressing the general aim of the EIA Directive as opposed to a justiciable standard by which to assess compliance with EIA Law.

513. True, the view that EIA is procedural only can be overstated. Clarke CJ in Connelly819 said:

• it is important to note that EIA does not require any particular result to the relevant process but rather is concerned with the process itself.

• EIA law does require that permission be granted only where EIA favours the grant.

• the reasons for the grant must o demonstrate that the decision-maker properly had regard to the results of the EIA in coming to its

conclusion. o state the basis on which the EIA leads to the conclusion that a permission can be granted.

514. At one point820 ALAB suggested that after EIS screening has applied the precautionary principle in deciding whether EIA is needed, the precautionary principle has no further role in EIA because EIA is procedural and does not dictate the outcome of development consent applications. It is difficult to accept such a restrictive view of the role of the precautionary principle in EIA. In Djurgården 821 Sharpston AG grounded the well-established view that the EIA Directive ‘has a wide scope and a broad purpose’ in “the very purpose of the directive, which seeks to achieve a high degree of environmental protection on a precautionary basis, on the grounds that ‘the best environmental policy consists in preventing the creation of pollution or nuisances at source, rather than subsequently trying to counteract their effects’”. There is no prohibition on EIA, de facto, determining the outcome of a development consent application. All but invariably it does so by prompting the imposition of conditions as to mitigation of environmental risk. EIA may well prompt refusal of development consent if residual environmental risk822 is considered unacceptable. ALAB softened that position later823 – conceding, for example, that the precautionary principle may apply in the information-gathering stage of EIA. I agree – as the Commission and Kingston say, it requires comprehensive assessment of risks to the environment based on the best available evidence, the most reliable scientific data available and the most recent results of international research.824 But ALAB argued, citing Scott,825 that it is very difficult to see how the precautionary principle would have a practical legal effect of requiring a particular outcome to a development consent application. Counsel for ALAB also

817 FitzPatrick v An Bord Pleanála & Apple [2019] 3 IR 617.

Case C-396/92 Bund Naturschutz in Bayern v Freistaat Bayeran [1994] E.C.R. I-3717.

819 Connelly v An Bord Pleanála, [2021] 2 IR 752 §102 – I have edited and paraphrased his view somewhat.

820 Day 10 p12.

821 Case C-263/08 Djurgården-Lilla Värtans Miljöskyddsförening v Stockholms kommun genom dess marknämnd, Judgment of 15 October 2009.

822 i.e. residual despite mitigation.

823 Day 12 p133 et seq.

824 See below.

825 Infra.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

argued that the applicants had not specified precisely how, they say, the precautionary principle required ALAB to make a particular decision in this case.

515. The precautionary principle is a general principle of EU Law826 – Artegoden,827 Bayer,828 France829 and Hygeia.830 The Precautionary and Preventive Principles are “enshrined”831 as essential to the high level of protection of the environment required by Article 191 TFEU to the effect that EU environmental law is

“based on the precautionary principle” and on the principle that “preventive action should be taken”. But the

principle is not defined, or even described, in the EU Treaties or other EU legislation. Scott832 asserts that “The precautionary principle is controversial, in part because it is often mis-represented and misunderstood” and that there are many different versions and no agreed definition of it.

516. Kingston et al833 cite the CJEU in Gowan834 to the effect that

“It follows from the precautionary principle that, where there is uncertainty as to the existence or extent of risks to the health of consumers, the institutions may835 take protective measures without having to wait until the reality and the seriousness of those risks become fully apparent.”

For “risks to the health of consumers” we may read “risks to the environment”.

It will have been noticed that this formulation, at least, is enabling rather than coercive – “may take”.

517. The caselaw review by Macken J836 in Hygeia leads her to conclude (as does Scott) that “the

precautionary principle is highly technical in nature and subject to considerable constraints in its application.”

This seems consistent with the observation of the EU Commission in 2000837 that “The issue of when and how to use the precautionary principle, … is giving rise to much debate, and to mixed, and sometimes contradictory views” and its warning that the precautionary principle “should not be confused with the

826 Though that has been doubted – Scott, Legal Aspects of the Precautionary Principle, A British Academy Brexit Briefing November 2018. Professor

Joanne Scott FBA. Professor of European Law at the European University Institute in Florence and at University College London and Co-Director of the

Academy of European Law at the European University Institute. She considers the precautionary principle to be “generally applicable as an autonomous

principle throughout these different areas of EU law, rather than constituting a general principle in the formal sense as understood by the European

Court” However her distinction and doubt need not detain us as, by Article 191 TFEU, the principle is clearly enshrined as a basis of EU Environmental

Law.

827 Case T-74/00 & others, Artegodan GMBH & Others v Commission of the European Communities.

828 Cases T-429/13 and T-451/13, CJEU 17 May 2018.

829 Case T-257/07, France v Commission.

830 Hygeia Chemicals v Irish Medicines Board [2010] IESC 4 – Macken J, citing Case T-74/00 Artegodan and Case T-13/99 of Pfizer Animal Health v Council

of the European Union [2002] ECR II/ 3318.

831 Hygeia Chemicals v Irish Medicines Board [2010] IESC 4 – Macken J.

832 Op cit.

833 European Environmental Law, 2017, p94.

834 Case C-77/09 Gowan Comércio Internacional e Serviços Lda v Ministero della Salute, Judgment of 22 December 2010.

835 Emphasis added.

836 A Judge of the European Court of Justice from 1999 to 2004.

837 Communication from the Commission on the precautionary principle, Brussels, 2.2.2000, COM(2000) 1 final.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

element of caution that scientists apply in their assessment of scientific data”. The Commission said: “It is also necessary to clarify a misunderstanding as regards the distinction between reliance on the precautionary principle and the search for zero risk, which in reality is rarely to be found.”

518. The issue is often further complicated by the fact that analysis of the nature and effect of the precautionary principle is often considered by the EU courts in the context of challenges to EU secondary legislation prompted by policy considerations838 – a somewhat different context to that of environmental

decision-making in individual cases. The question before the EU courts is often one of justification of precautionary positive action rather than one of refusal for precautionary reasons to permit an economic activity – for example by a development consent decision. That said, the underlying principles apply. Kingston et al839 say:

“ … the precautionary principle should not be used as a carte blanche for discretionary decisionmaking: at least enough indications must be available to corroborate the existence of a genuine risk, and member states must still deploy the best available evidence in decision making. …. … CJEU rulings … tend to insist on scientific evidence but at the same time afford a broad margin of discretion to EU authorities to determine which evidence should be relied on and how it should be interpreted. At times, this imparts a certain seesaw quality to CJEU judgments with the court seemingly vacillating between the importance of expertise on the one hand and discretion on the other … CJEU rulings aim to affirm expectations of evidence-based decision making without leaving EU public authorities hamstrung under conditions of scientific uncertainty. It is a difficult balance to achieve …… instances when the CJEU annuls an EU legal act for not being scientifically justified remain extremely rare.”

519. The precautionary principle prompts intervention where the scientific evidence demonstrating the existence of a risk to the environment remains equivocal. While the Commission in (as late as?) 2000840 explicitly gave general guidance only and did not claim the final word on these issues, it described the precautionary principle as a risk management strategy used by decision-makers to reach “proportionate, non-discriminatory, transparent and coherent decisions”. It is applicable “where scientific evidence is insufficient, inconclusive or uncertain and there are indications through preliminary objective scientific evaluation that there are reasonable grounds for concern that the potentially dangerous effects on the environment …. may be inconsistent with the chosen high level of protection.”

838 For example, in Artegoden what was at issue was a challenge to the Commission’s withdrawal of marketing authorisations for medicines with a view

to the precautionary protection of public health.

839 European Environmental Law, 2017, p35.

840 Communication from the Commission on the precautionary principle, Brussels, 2.2.2000, COM(2000) 1 final. See also Kingston et al, EU environmental

Law, Cambridge 2019. P94. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

520. Unlike the CJEU in Gowan,841 Hygeia842 formulated the Precautionary Principle in coercive rather enabling terms: It

“…… requires competent authorities to take all appropriate measures to prevent specific potential risks to public health, safety and the environment by giving precedence to the requirements related to the protection of those interests, and without having to await definitive evidence of the risk involved.” 843

521. Bayer,844 offering it seems a “Definition”, more recently confirmed that the “precautionary principle is a general principle of EU law” and described the circumstances in which it applies and of what its application consists:

“Where there is scientific uncertainty as to the existence or extent of risks to …… the environment, the precautionary principle allows845 the institutions to take protective measures without having to wait until the reality and seriousness of those risks become fully apparent…”

However, it also describes the principle as “requiring” 846 competent authorities “to take appropriate measures to prevent specific potential risks to …. the environment, by giving precedence to the requirements related to the protection of those interests over economic interests ….”

The precautionary principle was recently considered in a Heather Hill case847 – albeit in the somewhat particular context of the Habitats Directive.

522. Kingston et al848 say:

"According to the CJEU, the precautionary principle requires a two-stage analysis.

First, the potentially negative consequences for the environment must be identified.

Secondly, it requires a comprehensive assessment of the risks to the environment based on the most reliable scientific data available and the most recent results of international research.

841 Case C-77/09 Gowan Comércio Internacional e Serviços Lda v Ministero della Salute, Judgment of 22 December 2010.

Hygeia Chemicals v Irish Medicines Board [2010] IESC 4 – Macken J, citing Case No. T-74/00 & others, Artegodan GMBH & Others v Commission of the

European Communities, supra, and Case No. T-13/99 of Pfizer Animal Health v Council of the European Union [2002] ECR II/ 3318.

843 Emphasis added.

844 Cases T-429/13 and T-451/13, CJEU 17 May 2018. §§109 & 110.

845 Emphasis added.

846 Emphasis added.

847 Heather Hill Management Company CLG v An Bord Pleanála [2022] IEHC 146.

848 European Environmental Law, 2017, p95

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

And if, following this analysis, it proves impossible to determine with certainty the existing or extent of the alleged risk because of the insufficiency, inexclusiveness or imprecision of the results of studies conducted that there remains a likelihood of real harm to the environment if the risk materialises, the precautionary principle will apply to justify objective and non-discriminatory measures that would otherwise be prima facie precluded due to their restrictive nature."

523. It seems to me that the following principles can be, it has to be said somewhat tentatively, derived from the cases cited above and the cases cited therein:

i. The guarantees in administrative proceedings conferred by the EU legal order are of fundamental importance and include in particular, the duty of good administration849 – the duty of the competent institution to examine carefully and impartially all the relevant aspects of the individual case.850 These rights broadly correspond to Irish law guarantees of fair procedures and independent, objective, impartial decision-making.

ii. In a particular case, it is first necessary to identify the specific risk to which, it is suggested, the precautionary principle applies.

iii. It is next necessary to achieve, by way of assessment of that risk, such thorough level of scientific certainty as is possible. The best scientific knowledge851 reasonably available852 must be availed of.

The Commission asserts853 that the precautionary principle requires a scientific evaluation of risk which is as objective, thorough and complete as possible and includes an assessment of scientific complexity, disagreement and uncertainties. As applied to EIA, this seems to me consistent with the general position that EIA must be comprehensive – as complete as possible.854 iv. However, it may prove impossible to carry out an absolutely full or definitive risk assessment, for example because the available scientific data is inadequate or because such a risk assessment would require long and detailed scientific research. Also, the obligation on the development consent applicant in EIA is to provide information “within the limits of what may reasonably be required of a private operator” – EIA Directive Article 5(1)(b), Abraham,855 IL v Land Nordrhein-Westfalen,856 and

849 Art 41 CFREU – 1. Every person has the right to have his or her affairs handled impartially, fairly and within a reasonable time by the institutions,

bodies, offices and agencies of the Union.

850 Pfizer Animal Health v Council of the European Union [2002] ECR II/ 3318 as to the duties of EU institutions.

851 It seems to me that the word “scientific” is used here in a broad sense as including, for example, technological knowledge.

People Over Wind Environmental Action Alliance Ireland v An Bord Pleanála [2015] IECA 272. O’Sullivan v An Bord Pleanála & Rowing Ireland [2022]

IEHC 117, ETI v An Bord Pleanála & Cloncaragh Investments [2022] IEHC 540. These were AA cases but if the word “reasonably” applies in that strictly

demanding context, a fortiori it applies generally.

853 Communication from the Commission on the precautionary principle, Brussels, 2.2.2000, COM(2000) 1 final. See also Scott op it.

854 Case C-50/09, Commission v Ireland, Judgment of 3 March 2011; Case C‑463/20 Namur-Est Environnement ASBL v Région wallonne, Opinion of Kokott

AG of 21 October 2021.

855 Case C-2/07 Abraham v Wallonia [2008] ECR I-1197, Opinion of Kokott AG of 29 November 2007.

856 Case C-535/18 IL et al v Land Nordrhein-Westfalen, Judgment of 28 May 2020, §82.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Coyne.857 It is also recognised that scientists may have widely diverging views, not least as to what may be extremely complex scientific and technical assessments.

v. Competent authorities may (must if the competent authority lacks the relevant expertise) entrust scientific risk assessments and the provision of scientific advice based thereon, to experts. Such assessment and advice must be thorough and based on the principles of excellence, independence and transparency.858 Given that the scientific advice on which competent authorities rely often emanates from the protagonists in the development consent process, these principles place considerable demands on the experts retained by those protagonists – perhaps especially demands of independence, including of their clients. It also requires careful and properly sceptical consideration of that scientific advice by competent authorities. The deployment of internal or contracted expertise by the decision-maker itself, as occurred in this case, can represent an appreciable reassurance.

vi. For the application of the precautionary principle to continue to be relevant after such risk assessment, there must remain, despite such assessment, scientific uncertainty as to the existence or extent of the risk.

vii. The uncertainty requirement is not a licence or requirement to act on merely hypothetical risk.

There must be “reasonable grounds for concern”.

viii. What follows is risk management – not necessarily prohibition of development. It is also based on scientific assessment – of like quality and comprehensiveness as that which informed the risk assessment. The competent authority is to decide which measures appear to it to be appropriate and necessary to prevent the risk from materialising or to minimise it. So, as applicable to EIA, the precautionary principle requires consideration of mitigation859 and assessment of any residual risk expected to remain despite mitigation. Such residual risk must, to the extent possible, be identified, characterised, weighed and assessed before being deemed acceptable or unacceptable.

ix. Once the assessment is adequate to identify the residual risk of a proposed development (i.e. the risk remaining despite mitigation – remembering that there may remain scientific uncertainty as to the degree of such risk) it is not (leaving aside situations such as AA under the Habitats Directive, in which the level of acceptable risk is prescribed) generally required, including in EIA, to achieve “zero risk” 860 or even, necessarily, the highest technically possible level of protection (i.e. the lowest level of risk). The competent public authority has a broad discretion to determine the levels of risk deemed acceptable and unacceptable for society by balancing risks and benefits in the particular

857 Coyne v An Bord Pleanála & Enginenode [2023] IEHC 412, §129.

858 This particular observation was made in Pfizer Animal Health v Council of the European Union [2002] ECR II/ 3318 as to matters relating to consumer

health and the Court cited the preamble to Commission Decision 97/579 and the Commission's Communications on the Precautionary Principle and on

Consumer Health and Food Safety. However it is difficult to see it as anything less than a generally applicable principle.

859 In both the preventive and curative senses of that word.

860 Case T-13/99 Pfizer Animal Health v Council of the European Union [2002] ECR II/3318.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

circumstances of each individual case. Scott861 says that “The threshold above which risk will no longer be regarded as acceptable depends upon the wording of particular legislation.”

x. This seems to me consistent with the position that EIA is primarily procedural and informative and EIA law requires that environmental considerations be taken into account in deciding development consent applications – as opposed to requiring specific substantive development consent decisions – FitzPatrick.862

xi. The EU cases cited in Hygeia express similar views to those applied in judicial review at Irish law as to EIA – the court is concerned with the legality of the decision and may not substitute its assessment of the facts for that of the competent authority.

524. Scott863 cites the CJEU as emphasising the role of proportionality in the application of the precautionary principle. She says that

“Proportionality encompasses three elements: a measure must be necessary and appropriate to attain its aim, must not be more restrictive or onerous than is necessary to do so, and the disadvantages caused by the measure must not be out of proportion with the aim pursued. … This last element of the proportionality principle is regarded as raising the question of whether a decision strikes an appropriate balance between the protection of .. environment ..864 on the one hand, and economic interests, including the interests of traders, on the other” and this “overlaps with the risk management principle identified by the Commission which requires an examination of the costs as well as the benefits of action and inaction.”

525. It seems to me that particular tension attends the question what are competent authorities to do once the precautionary principle has identified the “reasonable grounds for concern” of a risk to the environment of which definitive evidence remains elusive or where, as here and which amounts to the same thing, the various experts are deeply disagreed. Bayer and Hygeia describe this situation as requiring competent authorities to take “appropriate measures” or, indeed, “all appropriate measures” to prevent specific potential risks. The Commission seems to take a less demanding view of the duties of decisionmakers.

526. Scott identifies a “key question”: “Does the precautionary principle ever require, rather than merely permit, the adoption of protective measures?” She suggests that despite the CJEU’s sometimes using the language of requirement, overall its case law tends to the permissive, with some exceptions. She identifies as

861 Op cit.

862 FitzPatrick v An Bord Pleanála & Apple [2019] 3 IR 617.

863 Op Cit

864 More fully, the referred to “public health, environment and consumer safety”.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

such an exception the requirement of scientific certainty as to the absence of adverse effect in AA under Article 6(3) of the Habitats Directive.865 To those engaged in environmental law, the contrast with the primarily procedural effect of EIA immediately occurs. Scott suggests that “It is very rare, though not unknown, for the precautionary principle to serve as a basis for challenging decisions because they are insufficiently protective.” This echoes the views of Kingston et al, set out above, as to the tension, and the difficult balance to achieve, between requiring scientifically evidence-based decisions on the one hand and curial deference to the exercise of discretion on the other such that, empirically, instances when the CJEU annuls an EU legal act for not being scientifically justified remain extremely rare.

527. It does seem to me, speaking generally, that opponents of projects tend to mistake the precautionary principle for a requirement that, to be permitted, projects requiring EIA must be environmentally risk-free. If, as is clear, zero risk is not required, it necessarily follows that someone – the statutory decision-maker – must identify and quantify the residual risk posed by a project, decide whether it is acceptable and grant or refuse permission accordingly.

528. It appears to me, taking the cases together, that once EIA is done and adequate reasons are given, decisionmakers are entitled to considerable curial deference to their multifactorial assessment balancing disparate considerations (a legalistic description of the necessary task of comparing and choosing between apples and oranges and the exercise of judgment such tasks inevitably require866) in deciding whether, in context – which context may properly include non-environmental considerations – an identified residual risk (i.e. after prescribed mitigation) is acceptable.

529. In short, the precautionary principle imposes considerable and detailed burdens on applicants for permissions/licenses and on competent authorities in deciding such applications. But it also imposes very considerable burdens on applicants for judicial review seeking to rely on the principle and successful challenges to the multifactorial assessment whether residual environmental risk is acceptable when weighed against the benefits of a project are likely to be rare.

530. Closely associated with the foregoing principles is the treatment of uncertainty in EIA. Uncertainty is inevitable and far from necessarily flaw in EIA. EIA law does not unreasonably prescribe the assessment of unassessable risk. It does not require certainty in the assessment of risk. But it does require recognition, recording and assessment of uncertainty. It is entirely unsurprising – indeed eminently desirable – that EIA should articulate uncertainties – see Jones,867 Long868 and Thakeham.869 And EIA law requires applicants for

865 Scott cites C-127/02 Waddenzee.

866 To pick an irrelevant example, the task of balancing the need for quantum of housing against the amenity value of a view.

867 R (Jones) v Mansfield District Council and another [2003] EWCA Civ 1408, [2003] All ER (D) 277 (Oct) – cited in Monkstown Road Residents Assoc’n v

An Bord Pleanála [2022] IEHC 318.

868 R (Long) v Monmouthshire County Council [2012] EWHC 3130 (Admin).

869 R (Thakeham Village Action Ltd) v Horsham District Council [2014] EWHC 67 (Admin).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

licences to provide information only within the limits of what may reasonably be required of a private operator.870 Lindblom J observes in Thakeham that in Jones it was considered possible in principle, depending on the circumstances of the case, to have sufficient information to enable a decision reasonably to be made as to the likelihood of significant environmental effects even if certain details are not known and further surveys are to be undertaken. The EPA Guidance on EIARs871 observes that

“Some uncertainty is unavoidable in EIA, especially about matters that involve an element of judgement, such as assigning a level of significance to an effect. Such judgements should be explicit and substantiated rather than presented as objective fact.”

“Where relevant the EIAR should describe the forecasting methods or evidence used to identify and assess the significant effects on the environment, including details of difficulties (for example technical deficiencies or lack of knowledge) encountered in compiling the required information and the main uncertainties involved.”

531. Of course, this is last passage is taken directly from Article 5(1) and Annex IV of the EIA Directive which requires that an EIS/EIAR include:

“6. A description of the forecasting methods or evidence, used to identify and assess the significant effects on the environment, including details of difficulties (for example technical deficiencies or lack of knowledge) encountered compiling the required information and the main uncertainties involved.”

While this passage derives from the 2014 amendment to the EIA Directive, it merely made explicit the preexisting requirement. Jones, Long and Thakeham preceded the 2014 amendment of the EIA Directive and the necessity in EIA to address uncertainty was already established.

532. MRRA872 cites Jones873 to the effect, inter alia, that,

“Although a planning authority must have sufficient information about the impact of the project to be able to make an informed judgment as to whether it was likely to have a significant effect on the environment, that does not mean that all uncertainties must be resolved or that a decision that an EIA is not required could only be made after a detailed and comprehensive assessment had been made of every aspect of the matter. Uncertainties might or might not make it impossible reasonably to conclude that there was no likelihood of significant environmental effect. Everything depends on the circumstances of the individual case.”

870 Coyne v ABP, Ireland & EngineNode [2023] IEHC 412 §129.

871 Guidelines on the information to be contained in Environmental Impact Assessment Reports, May 2022 §§3.7.2 & 3.7.5.

872 Monkstown Road Residents Association v An Bord Pleanála & Lulani [2022] IEHC 318.

873 R (Jones) v Mansfield District Council and another [2003] EWCA Civ 1408, [2003] All ER (D) 277 (Oct).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

533. It will be seen from the foregoing that SWI’s reduction in its submissions of the effect of the precautionary principle in EIA to an assertion that “where uncertainty remains, the requirement is to refuse consent” is a crude and far from a useful statement of the law and is destructive of the essentially procedural requirement of comprehensive EIA.

EIA – AQUACULTURE LICENCE – WAS IT DONE?

EIA – Introduction

534. MOWI’s 2011 EIS, submitted with its Aquaculture Licence Application, assumes, but does not state the legal source of,874 the necessity of EIA of the proposed aquaculture project. However, as its “Conclusions”875 reveal, it proceeds on the entirely and fundamentally misconceived basis that “This Environmental Impact Statement arises from an Environmental Impact Assessment… ”. Just as mistakenly it lists “the main findings of the EIA for the proposal, as reported in this EIS”. Of course, these assertions reverse the proper order of things. An EIS is not the product of an EIA – the developer does not do EIA. The EIS informs the EIA by the authority competent to perform EIA – in this case and in the first instance, the Minister.

535. However, the Minister did not suffer from the same misconception. His EIA of June 2015876 identified

the legal obligation to do EIA as “section”877 4A of the Aquaculture (Licence Application) Regulations, 1998 –

which is set out in Appendix 1 to the EIA. I find the interplay between Articles 4A and 5878 of those 1998 Regulations as amended, complex – at least as to obligation to submit an EIS. But Article 4A appears to impose the obligation to do EIA. Though the 2006 Regulations879 amending Article 5 of the 1998 Regulations continued to cross-reference Part II of the First Schedule to the EIA Regulations 1989,880 six years earlier s.176(3) PDA 2000881 had superseded that Schedule such that the applicable Schedule is now, it seems, Schedule 5 PDR 2001.882 The net result of negotiating this maze is that EIA is required of an aquaculture licence application for a “seawater salmonid breeding installation” (amended in 2016883 by the addition of

874 Apart from a brief and erroneous reference to “the Fisheries (Amendments) Act 1977” – presumably intending the Fisheries (Amendments) Act 1997.

875 2011 EIA Vol 1 p295 Section 10.

876 p4.

877 Sic.

878 Aquaculture (Licence Application) Regulations, 1998 Regulation 5 as amended by the Aquaculture (Licence Application)(Amendment) Regulations

2006; Aquaculture (Licence Application) (Amendment) Regulations 2010.

879 Aquaculture (Licence Application)(Amendment) Regulations 2006.

880 European Communities (Environmental Impact Assessment) Regulations 1989 (S.I. No. 349 of 1989)

881 (3) Any reference in an enactment to development of a class specified under Article 24 of the European Communities (Environmental Impact

Assessment) Regulations, 1989 (S.I. No. 349 of 1989), shall be deemed to be a reference to a class of development prescribed under this section. It was

Article 24 of the 1989 Regulations which had adopted the First Schedule to those regulations.

882 Planning and Development Regulations 2001.

883 Aquaculture (Licence Application)(Amendment) Regulations 2016.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

the words “(other than for trial or research purposes where the output would not exceed 50 tonnes)”, or a “Seawater fish breeding installations with an output which would exceed 100 tonnes per annum;”.884 These provisions clearly seek to transpose the requirement imposed by the EIA Directive885 as to EIA of “intensive fish farming”.

536. ALAB, by its Impugned Determination, did not identify the source of its obligation to do EIA. As to appeals to ALAB, EIA is required886 of the same categories of installation as stipulated above – but also, and

more generally, of “intensive fish farming”887 if ALAB determines it “likely to have significant effects on the

environment”.

537. The Aquaculture Licence in the present case is for the “cultivation” of salmon. One might equally say it is for the “farming” of salmon. The intended method is by introducing smolts888 bred elsewhere and farming them for up to 22 months to maturity, for human consumption. Indeed, it seems that literally, the descriptor “seawater salmonid breeding installation” is a misnomer as salmonids breed only in freshwater. It seems to follow that the phrase is intended to describe not breeding but rearing and why the regulations depart from the EIA Directive word “farming”889 is unclear. However, as all accept EIA of the Salmon Farm was required nothing seems to turn on the misnomer.

EIA – Screening

538. IFI complains of an absence of an EIA Screening decision.890 ALAB protests that IFI did not plead that case.891 IFI admits that the point was not specifically pleaded but says it was generally pleaded and cites its plea of a failure to carry out a “properly executed EIA”. I have addressed the rules of pleading in judicial review earlier. It suffices the mention here the requirement of precise pleading in judicial review. The plea of failure to carry out a “properly executed EIA”, if considered sufficient on its own, would encompass a vast range of challenges to an EIA. On the other hand, failure to screen for EIA is a simple point entirely susceptible to being simply and directly pleaded. It was not and I reject the point for want of pleading.

884 specified in point 1(f) of Part II of the Fifth Schedule PDR 2001 Point 1(f) reads in full:

(f) Seawater fish breeding installations with an output which would exceed 100 tonnes per annum; all fish breeding installations consisting of cage rearing

in lakes; all fish breeding installations upstream of drinking water intakes; other freshwater fish breeding installations which would exceed 1 million smolts

and with less than 1 cubic metre per second per 1 million smolts low flow diluting water.884

885 Annex II, §1(f).

Aquaculture Appeals (Environmental Impact Assessment) Regulations 2012 – Article 3(2)(a) and by cross-reference to Aquaculture (Licence

Application) Regulations, 1998 Regulation 5 as amended.

887 Aquaculture Appeals (Environmental Impact Assessment) Regulations 2012 – Article 3(2)(b) reads “aquaculture of a class specified in Annex II of the

Council Directive” – Annex II, §1(f) refers to “intensive fish farming”.

888 A Smolt is a young salmon or trout which has grown to the stage of moving from freshwater to saltwater.

889 Annex II, §1(f) refers to “intensive fish farming”.

890 §4.27 of IFI’s written submissions.

891 §42 of ALAB’s written submissions in the IFI Proceedings.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

539. However, if I am wrong in that rejection I can briefly state that the point, if pleaded, would have foundered. It is highly formalistic and without merit given ALAB accepted that it was obliged to do EIA and given the purpose of the EIA Directive, as set out in its Art 2.1, “to ensure that, before development consent

is given, projects likely to have significant effects on the environment by virtue, inter alia, of their nature, size

or location are made subject to a requirement for development consent and an assessment with regard to their effects on the environment.” Even assuming (which I do not find) any breach of an EIA Screening obligation in this case, I have no hesitation in rejecting the challenge in this regard in the exercise of my discretion to refuse relief. It is very clear is that, on whatever precise legal basis of obligation to do so, both the Minister and ALAB recognised the need to do and purported to do EIA. Any absence of EIA Screening made no difference to the attainment of the purpose of the EIA Directive.

Did ALAB do EIA?

540. Dr Forde892 and Dr Gargan893 for IFI complain that ALAB “did not commission or conduct its own EIA”. In large part that proposition is based on the observation that in §4 of its Determination, headed “Environmental Impact Assessment” ALAB referred to “the EIA” in terms clearly referable to the Minister’s EIA of 2015 which the Minister had sent to ALAB pursuant to s.43 of the 1997 Act. That observation is unfortunate as tending, taken out of context, to confuse. But it does not provide a proper basis for an inference that ALAB did not itself do an EIA.

541. IFI aptly cite

• Rushe,894 to the effect that “[t]he starting point for a consideration of the applicants’ case on EIA grounds

is the decision of the Board itself as recorded in the Board Order”

• Aherne,895 in which that the Board had failed to do an EIA was pleaded. Noonan J observed that the Board’s decision clearly recorded that it had done an EIA and that “the onus of proving otherwise rests upon the applicants. No evidence has been adduced to contradict the assertion of the Board contained in its decision.” CHASE896 applied Aherne as to onus of proof.

542. As stated at §4 of its Determination as to EIA, and far from relying only on the Minister’s EIA, ALAB significantly canvassed issues relevant to EIA including, significantly, information provided to ALAB after the date of the Minister’s EIA. They included information supplied:

892 Head of Operations IFI (now retired). He swore Affidavits of 23 September 2021, 12 October 2021, 29 June 2021 and 30 September 2022.

893 Affidavit 30 September 2022, inter alia §28.

894 Rushe v An Bord Pleanála [2020] IEHC 122 §227.

895 Aherne v An Bord Pleanála [2015] IEHC 606 §22.

896 CHASE v An Bord Pleanála [2021] IEHC 203 §326. See also Ratheniska v An Bord Pleanála [2015] IEHC 18, §67. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• By Prof McIntyre to the effect that the Minister’s EIA had been inadequate and suggesting an sEIS. ALAB accepted his views on 9 November 2017 and 19 February 2018. At this latter meeting ALAB noted inadequacy of the EIS as to birds and that the Minister’s EIA had been “clearly inadequate”.

• by MOWI (notably the sEIS of 2018).

• by the Marine Institute and IFI itself.

• By NPWS – to the effect that EIA was required as to effect on the FwPM.897

• by Technical Advisor Dr Saunders (a lengthy final report which canvasses EIA issues and concludes898 that there are no substantive technical reasons to refuse the current licence application having articulated899 “The case for granting the licence on the basis of no significant effects on the environment ….”)

• by Technical Advisor Dr O’Toole as to the FwPM.

• by Dr Crowe as to Birds.

543. Having specifically invoked a series of documents which could not have informed the Minister’s EIA, ALAB then in its Determination expressly itself expressed itself satisfied:

• that the 2011 EIS had been inadequate such that it had required the sEIS published and circulated it and invited submissions on it. 900

• with the adequacy of the information before it by reference to what are clearly the subject matters of EIA as identified in the EIA Directive.901

• that the proposed aquaculture activity at the Site will not have significant effects on the environment “by virtue of, inter alia, its nature size or location” – this phrase being taken directly from the EIA Directive.

544. Though it is not essential to my conclusion, I note that the minute of ALAB’s meeting of 24 June 2021 records that “Part 4 of the draft Determination outlined the steps taken by the Board to ensure a complete Environmental Impact Assessment process was followed”.

545. Having regard to the express terms of ALAB’s Determination, I reject the complaint that ALAB did not

do an EIA – both generally and as to the complaint that it regarded the Minister’s EIA as “the” EIA (though I

accept that the usage was unfortunate). Reading ALAB’s decision overall and in the context of the

897 NPWS s.47 Reply 16 November 2018.

898 §11

899 §10.2

900 ALAB Determination §§4.2 & 4.3.

ALAB Determination §4.6: “ …… the Board is satisfied that all the documents taken together identify, describe and assess in an appropriate manner, in

the light of the appeals before it, the direct and indirect effects of the proposed activity at the Site on the following factors:

4.6.1 human beings, fauna and flora; 4,6.2 soil, water, air, climate and the landscape;

4.6.3 material assets and the cultural heritage; and

4.6.4 the interaction between the factors referred to in 4.6.1- 4.6.3

and that the proposed aquaculture activity at the Site will not have significant effects on the environment, including the factors listed in 4.6.1- 4.6.4, by

virtue of, inter alia, its nature size or location. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

environmental information before it and the fact that much of had not been before the Minister, it is clear that ALAB did its own EIA. I respectfully reject the complaint in this regard.

EIA – Notice to IFI

546. IFI says ALAB did not notify IFI that ALAB was doing EIA. ALAB raises a pleading objection, which IFI concedes in admitting that this is not a ground on which relief may be granted. IFI instead calls in it aid as evidence supporting its compliant that no EIA was done or properly done. Even assuming that such evidence could appreciably support that complaint, there is a very considerable lack of reality to that challenge – as a brief consideration of the chronology will readily reveal.

547. As stated, MOWI’s application for the licences was accompanied by an EIS. MOWI’s letter to IFI of 11 January 2012 records that IFI (as a statutory consultee) had been sent the EIS in June 2011, and encloses a revised application and EIS, inviting comment to the Minister. IFI made submissions to the Minister by letter of 1 February 2012 explicitly asserting the inadequacy of the EIS for reasons given and opposing the grant of a licence by reason of asserted risk to wild salmonids. The Minister’s decision was published in Iris Oifigiúil on 18 September 2015 with a weblink to his reasons. While I have not seen a copy of that webpage, the Minister’s determination approved reasons for publication on the website which included the words “the proposed aquaculture will … have no significant ecological effects on wild fisheries, natural habitats, flora and fauna or the environment generally." While this (and indeed the Iris Oifigiúil notice and newspaper notices to like effect) could better have explicitly recorded that the Minister had done an EIA, to an expert statutory consultee like IFI, viewing it in light of its knowledge of and participation in the process to that point, it can only have conveyed that EIA had been done.

548. IFI’s appeal dated October 2015 to ALAB of the Minister’s determination to grant the Aquaculture Licence explicitly asserted “The EIS is inadequate” as a ground in itself and citing EIS content repeatedly on each of its other 6 grounds of appeal. Further, IFI as an expert statutory consultee, cannot but have known that ALAB was obliged to do its own EIA. Indeed one of its grounds of complaint is of ALAB’s alleged failure to do an EIA – an obligation of which IFI must be presumed to have been aware at all times. Dr Forde for

IFI902 makes the very point that ALAB was “clearly” obliged to do a “mandatory” EIA as the activity was quite

evidently a “seawater salmonid breeding installation”.903 ALAB issued a s.47 request to IFI on 6 October 2016. While it did not explicitly mention EIA, its subject matter – Salmonids in Dromagowlane River – could not but have reminded IFI of the basis of its assertion of 1 February 2012 of the inadequacy of the EIS.

902 Affidavit 21 October 2021 §19

903 Dr Forde cites relevant regulations.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

549. The assertion that ALAB “did not commission or conduct its own EIA” does not support any proposition that IFI was unaware, or unaware for any good reason, that ALAB had embarked on EIA. Dr

Forde’s affidavits are far from a mere verification of IFI’s grounds on which it seeks judicial review. They are

detailed and substantive. While they refer to (surprising) advice to ALAB by Dr Saunders that EIA was not required,904 Dr Forde asserts that it was mandatory.

550. It is true that Dr Saunders’ Interim Report of December 2016 to ALAB drew the surprising conclusion that EIA was not needed and that report was published at that time. However, Dr Paddy Gargan, senior research officer of IFI and a salmon expert of considerable standing,905 represented IFI at the ALAB oral hearing in February 2017. His first and primary point at the oral hearing was the inadequacy of the EIS906 and EIA907 in various respects. On that day, others had alleged the inadequacy of the Minister’s EIA as distinct from the EIS – the conclusion of which EIA was recited908 on behalf of the Coomhola Salmon and Trout Anglers Association who asserted, in effect, their common interest with IFI as “stewards of wild salmonids”. Prof McIntyre, chair of the oral hearing and ALAB member, noted SWI submissions “relevant to the quality of the EIA and to be brought to the attention of ALAB in making its decision.”909 An Taisce, by its legal advisor, articulated in extenso,910 grounds of appeal sited in the EIA Directive and EIA – not least the view that the Appeal to ALAB was an administrative review within Art. 11911 of that Directive. He too cited the Minister’s EIA912 and explicitly envisaged ALAB’s EIA.913 On its second day the oral hearing adjourned on foot of explicit submissions that ALAB’s failure to publish the RPS report of 2015 breached public participation provisions of

the EIA Directive.914 If, which I would find very surprising, IFI had not prior to the Oral Hearing itself taken the

view that ALAB was obliged to do an EIA, that prospect can only have been brought home to IFI at the oral hearing. It was adjourned to September 2017 and in the interim IFI prepared a written submission to the resumed oral hearing disputing MOWI’s sea lice dispersal study as inadequate and not scientifically robust. Mr Millane for IFI spoke to that written submission at the resumed oral hearing. In this context it is especially notable that Dr Gargan’s affidavit915 does not suggest that at any time IFI was unaware that ALAB intended EIA or that IFI was prejudiced in any way by ignorance in that regard.

551. However, the matter does not end there. The chair of the oral hearing in his report recommended and ALAB agreed that ALAB should require MOWI to submit an sEIS – which MOWI did in April 2018. ALAB notified and enclosed it to IFI by letter dated 20 November 2018 and invited IFI’s observations. A public

904 Interim technical report 31 December 2016 §7.

905 His impressive credentials are stated at §3 of his Affidavit of 27 October 2021.

906 Transcript 14/2/17 p93 et seq.

907 Transcript 14/2/17 p95.

908 Transcript 14/2/17 p32, 37, 44, 45; See also Dr Gargan’s affidavit 30 September 2022 §7.

Transcript 14/2/17 p62.

910 Transcript 14/2/17 p121 et seq.

911 Formerly 10a.

912 Transcript 15/2/17 p11.

913 Transcript 15/2/17 p13 – “For those reasons I say that there isn't sufficient data before you to carry out an EIA properly. If the Board disagrees and

considers that it has sufficient data, I would ask it to ensure that it addresses those issues in a very robust manner.”

914 Transcript 15/2/17 p65.

915 His impressive credentials are stated at §3 of his Affidavit of 27 October 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

notice to similar effect was also published. This can only have conveyed to IFI that EIA was in progress despite Dr Saunders’ view that it was unnecessary. Dr Gargan replied in December 2018 addressing the sEIS but expressing no surprise or doubt that EIA was in train or that it was disadvantaged by any belated realisation on the part of IFI in that regard.

552. Notably, neither Dr Gargan nor Dr Forde in their numerous and detailed affidavits complains of a want of notice by ALAB that it was embarking on EIA, or any ignorance of IFI in that regard, much less any prejudice in consequence. I have no doubt that the absence of such averments is both entirely proper and no accident. On the assumption of a failure by ALAB to formally notify IFI of its intention to perform EIA, I find that any such failure had no consequence adverse to IFI which not merely had but availed of ample opportunity to participate in the IFI process and which does not complain that it was in fact unaware that EIA was afoot. I reject IFI’s reliance on alleged absence of notice as insubstantial and formalistic in the sense disparaged by Sharpston AG in Boxus and on the basis of the view of McGovern J in Ó Gríanna.916 I reject it as not grounded in law and if I am mistaken in that regard I reject it in the exercise of my discretion to refuse relief.

EIA & HUMAN HEALTH – CONSUMPTION OF FARMED SALMON

553. PS CG15 pleads only that EIA was done under the wrong directive – as to which see above. However the Sweetman Grounds in their particulars plead that ALAB’s Impugned Decision breached Articles 3 and

Articles 5 to 11 in both EIA Directives – inter alia by failing to consider human health properly or at all. It is

surprising, though not fatal to the plea, that this was nowhere heralded in any Core Ground. More significantly, the particulars do not elaborate on the plea or plead any facts in support. The Sweetman Applicant John Brendan O’Keeffe complains on affidavit that ALAB considered his submissions on human health irrelevant. He does not, even briefly, substantively identify, much less substantiate, his concerns as to human health. However, his appeal and submission to ALAB are exhibited917 and these do raise the issue of alleged deleterious indirect effect of the Salmon Farm on human health by eating farmed salmon. Of course, affidavits are no substitute for pleadings (Reid918) and, on affidavit, exhibition is no substitute for averment

(Murphy v Greene919). But it is clear that the issue is not of insufficiency of consideration of the issue – it is

simpler than that. It is common case that ALAB simply excluded it as irrelevant – the EIA did not consider that alleged effect. No pleading objection appears in the agreed list of such objections and so, with some misgivings as to pleading, I will decide the issue.

916 Ó Gríanna v An Bord Pleanála [2017] IEHC 7 §20-22 – citing, inter alia, Joined Cases C-128/09, C-129/09, C-130/09, C-131/09, C-134/09 and C-135/09

Boxus et al v Région Wallonne Opinion of Sharpston AG of 19 May 2011, §79.

917 Tab 9A, Exhibit ‘NC1’ to Affidavit of Noel Carr sworn 27th September 2021.

918 Reid v An Bord Pleanála [2021] IEHC 230 (High Court (Judicial Review), Humphreys J, 12 April 2021) §11.

919 Murphy v Greene [1990] 2 IR 566, Finlay CJ: “with regard to any matter which in the absence of special procedures involves the submitting of affidavit

evidence to the court for the purpose of it exercising a jurisdiction, the facts relied upon must be actually and clearly deposed to in an affidavit and not

by reference to any other written document.” SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

554. I should not be taken as acknowledging or rejecting the real existence of any such risk to human health. But assuming such a risk and given its having been excluded as irrelevant, the question is whether it comes within the required scope of EIA of the Salmon Farm project.

555. By 1992 the TEU had provided that EU policy on the environment shall contribute to objectives including “protection of human health”.920 The 1985 and 2011 EIA Directives each once recited “concerns to protect human health, to contribute by means of a better environment to the quality of life,” and Article 3 of each listed effects on “Human beings” as amongst the environmental effects requiring EIA. The EPA’s 2002 EIA Guidelines noted the relevance to EIA of effect on human health.921 In Case C-535/18 Land Nordrhein- Westfalen Hogan AG acknowledged indirectly that effect on human health was a concern of EIA under the 2011 Directive.922

556. The recitals to the 2014 EIA Directive make this concern explicit in that they repeatedly refer to “human health” and to “the environment and human health”. Specifically, the 2014 EIA Directive recites923 that the 2011 EIA Directive had contributed to a high level of protection of the environment and human health. Article 3 of the 2014 EIA Directive substituted “population and human health” for “Human beings”. Certainly, the concern with human health is explicit and prominent in the 2014 Directive but it is not a novelty of that directive: EIA under the 2011 Directive encompassed effect on human health. I do not see that applying the 2014 Directive would have made a difference in that regard.

557. ALAB cites Kilkenny Cheese.924 It addressed the alleged failure of an EIA for a cheese factory to consider the environmental effect of the additional milk production which the factory’s operation would allegedly prompt.925 The Supreme Court held that the upstream consequences of the supply chain of a new factory were too remote to require assessment in EIA and could not be considered effects of the project within Article 3.1 of the EIA Directive. Hogan J explicitly rejected an “open-ended” reading of Art.3(1)(a) of the EIA Directive as it requires assessment of the indirect effects of a project. He did so in favour of general approach that confines assessment to direct or indirect “effects which the development itself has on the environment” – which would not generally include the environmental impacts of the inputs or outputs of a factory. In the present case, the outputs are farmed salmon.

920 Treaty on European Union, signed at Maastricht on 7 February 1992 Art 130r – later Art 174 TEU and later again Art 191 of the Treaty On The

Functioning Of The European Union.

921 §§2.4.2, 3.2.5 & Glossary “Risk Assessment”.

922 §§25 & 49.

923 Recital 1 - “” identifies recitals introduced in 2014.

924 An Taisce v An Bord Pleanála et al, including Kilkenny Cheese Limited [2022] 1 I.L.R.M. 281.

925 That proposition was disputed as to fact.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

558. In Kilkenny Cheese, an Taisce cited An Taisce Edenderry926 in which EIA of a power plant was held to require assessment of the effects of extracting the peat to fuel it. Hogan J considered An Taisce Edenderry to have been a special case and that beyond such special cases “a range of difficulties open up”. Before I turn to them, I note that Hogan J cited Kemper,927 in which Allen J rejected the view that EIA of a wastewater treatment plant was deficient in failing to assess the impact on the environment of the eventual use of biosolids and other materials, which the plant would produce, as fertilizer on lands not part of the development site as those lands could not yet be identified. The primary difficulty Hogan J saw was that the suggested open-ended interpretation of Art. 3(1) of the EIA Directive would render it impossible to place any a priori limit on the range of indirect effects which would have to be assessed in EIA. He cited Finch928 as an example in which EIA of a project to expand a crude oil well was held not to require assessment of the effect of greenhouse gas emissions by reason of use of the oil produced, once refined. Hogan J cited with approval the rejection by Holgate J of the view that indirect effects included

“likely environmental effects more remote than direct effects (whether in time or location), but not so remote they cannot be attributed to the development at all, having regard to the purpose, nature, and any end product of the development, including the environmental impacts liable to result from the use and exploitation of the end product.”929

Holgate J considered that

“the fact that the environmental effects of consuming an end product will flow “inevitably” from the use of a raw material in making that product does not provide a legal test for deciding whether they can properly be treated as effects “of the development” on the site where the raw material will be produced for the purposes of exercising planning or land use control over that development.”

Hogan J cites in particular Holgate J’s view that “The EIA cannot be required to include effects which go beyond the effects of the project or development.”930 For purposes of the present case, I would add reference to the conclusion reached by Holgate J on this issue:

“The upshot is that the case law confirms that EIA must address the environmental effects, both direct and indirect, of the development for which planning permission is sought, ….. but there is no requirement to assess matters which are not environmental effects of the development or project. In my judgment the scope of that obligation does not include the environmental effects of consumers using (in locations which are unknown and unrelated to the development site) an end product which will be made in a separate facility from materials to be supplied from the development being assessed. I therefore conclude that, in the circumstances of this case, the assessment of GHG emissions from the future combustion of refined oil products said to emanate from the development

926 An Taisce v An Bord Pleanála [2015] IEHC 633.

927 Kemper v An Bord Pleanála [2020] IEHC 601.

928 R. (Finch) v Surrey County Council [2020] EWHC 3566 (Admin).

929 R. (Finch) v Surrey County Council [2020] EWHC 3566 (Admin) §98 et seq.

930 R. (Finch) v Surrey County Council [2020] EWHC 3566 (Admin) §98 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

site was, as a matter of law, incapable of falling within the scope of the EIA required by the 2017 Regulations for the planning application.

559. Given its approval by Hogan J, Holgate J’s decision represents the law here. However I should note that the EWCA931 upheld Holgate J as to the result but not his reasoning that the downstream products were excluded from EIA “as a matter of law”. The EWCA held that the decision-making authority needed to consider the necessary degree of connection required between the development and its putative effects if

EIA of those effects was to be required. The existence and nature of “indirect”, “secondary” or “cumulative”

effects would always depend on the particular facts and circumstances of the development under consideration. Whether there was a sufficient causal connection was a question of fact and reasonable evaluative judgment for the decision-making authority – which judgment was reviewable only for irrationality.932

560. Hogan J in Kilkenny Cheese noted that the Scottish Court of Session933 had approved the views of Holgate J. Hogan J observed:

“The upshot of Holgate J.'s analysis of these cases is that they reinforce the view that, first, an EIA must address the environmental effects, both direct and indirect, of the project or development for which planning permission is sought - there is no requirement to assess matters which are not environmental effects of the development or project; and second, that an effect of a project or development is one that is “concerned with the use of land for development and the effects of that use.”

“For my part, save for one possible caveat, I cannot but agree with these conclusions. It seems to me that if art.3(1) is given a remorselessly literal and open-ended interpretation there is no principled basis by which the limits of any EIAR assessment could confidently be ascertained. On this view, for example, the significant environmental effects resulting from the consumption or use of the end product would - or, at least, might - also have to be assessed. Would this mean, for example, that carbon emissions resulting from the use of articulated lorries to transport the cheese produced by the new factory to their various destinations in continental Europe would also have to be assessed? If - as seems not unlikely - large quantities of plastic were generated for the purposes of wrapping the cheese produced by the proposed factory at issue in the present case, would the environmental effects of this activity also have to be identified and assessed? If this were so, then this might also entail, for example, an environmental assessment of both the circumstances in which the plastic came to be generated in the first place and how it ultimately came to be disposed of following

931 Court of Appeal of England and Wales; R. (Finch) v Surrey County Council [2022] All ER (D) 93 (Feb) [2022] EWCA Civ 187.

932 On the Wednesbury standard applicable in England and Wales as elucidated in R (Law Society) v The Lord Chancellor [2018] EWHC 2094 (Admin);

[2019] 1 WLR 1649.

933 Greenpeace Limited v Advocate General [2021] CSIH 53 – the consumption of refined oil and gas by an end user need not be assessed in EIA as a direct

or indirect significant effect of the exploitation of the Vorlich oil field. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

consumption in the second place. These are just representative examples of potential indirect environmental effects in this wider, open-ended sense, examples of which could easily be multiplied.”

561. Significantly, Hogan J analyses the text of the EIA Directive as strongly suggesting that “the information to be supplied934 must be firmly tethered to the project itself, so that the indirect significant effects to be assessed must be intrinsic to the construction and operation of the project.” Further, he considered that where “proof of causality such as would satisfy the requirements of the EIA in respect of “direct and or indirect significant environmental effects” remains entirely elusive, contingent and speculative … its very elusiveness means that it is incapable of measurement or assessment and, hence, cannot be the sort of significant indirect environment effect which art.3(1) of the Directive must be taken necessarily to contemplate.” These observations seem to me to tend to exclude from the scope of EIA of the Shot Head Salmon Farm the alleged detrimental effects on human health of the consumption of farmed salmon.

562. In my view and in light in particular of Kilkenny Cheese, the alleged detrimental effects on human health of the consumption of farmed salmon, if any, are not indirect effects of the Shot Head Salmon Farm project as the concept of indirect effect is applied in EIA. They are not effects “firmly tethered to the project

itself” or “intrinsic to the construction and operation of the project.” Nor are they effects “concerned with the

use of land for development and the effects of that use.” The proper context in which any such risk should be regulated is that of food safety. I reject the challenge to the Impugned Decisions in that regard.

EIA – DESCRIPTION AND ASSESSMENT OF PROJECT, NUMBER OF CAGES

563. The EIA Directive 2011 requires that:

• projects likely to have significant effects on the environment by virtue, inter alia, of their “nature, size or

location” must be require development consent and EIA.935

• an EIS/EIAR describe the project to be assessed as to the “physical characteristics of the whole project”

and “its land use requirements”.936

• The public must be given a project description as soon as that information can reasonably be provided.937

564. The Impugned Aquaculture licence is for 18 cages.

934 By the developer for purposes of EIA.

935 Art. 2(1).

936 Annex IV §1(a).

937 Art 5 & Art 6(2)(a).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

565. The Minister’s pre-printed standard “Aquaculture and Foreshore Licence Application Form” completed by MOWI in June 2011 requires confirmation whether an EIS has been submitted and, immediately thereafter938 states “Drawing of the structures to be used and/or the layout of the farm OBLIGATORY”939 – to both of which MOWI answered “Yes”. MOWI’s 2011 Application was for 12 cages plus 2 more for temporary use to separate stock during harvesting (“12+2”) occupying 5.88 hectares of foreshore plus anchors and moorings, occupying an overall 19.20 hectares, being 45% of the proposed licensed foreshore area of 42.5 hectares. It and the EIS940 include drawings of:

• the “general layout of components”941 showing 12 cages, including proposed generalised cage and farm layout and specification diagrams.942

• the layout of those 12 cages within the proposed Foreshore Licence area in the bay and relative to Shot Head to the north west and Mehal Head to the north east.

566. The Minister’s EIA and draft Aquaculture Licence943 of September 2015 envisaged 12 cages only, in 2 rows of 6, and stipulates that, apart from a feed barge, “No other floating structures may be moored for extended periods at the site.” It does not provide for the 2 additional temporary cages for which licences had been sought. The draft Aquaculture Licence944 required the keeping of records as to each Cages – including “Position – by coordinates or grid position”. However, Schedule 2 of the draft Aquaculture Licence, the “approved plans and drawings” – contains only a poor quality drawing (at least as I have it) which I have not seen elsewhere. For all that, it clearly depicts, in section, 12 cages.

567. MOWI’s appeal to ALAB in October 2015 argued against limiting cage numbers and for flexibility to adopt evolving best practice in this regard and as to variation of cage layout and position. By this time, it said, best practice to produce the same quantum of salmon required 16 cages plus 2 temporary cages (“16+2”) and 18 was desirable.945 It did not explicitly nominate an increased cage footprint but, at least in general terms, that is readily to be inferred. The RPS Report 2015 was based on 12 geo-located cages.946

568. However the Saunders Interim Report of December 2016 said as to the seabed impacts of the requested 4 additional cages that, as the maximum allowable biomass of 2,800 tonnes will be maintained across the additional cages, he accepted MOWI’s view that the change will only result in more diffuse waste outputs, with lower concentrations distributed over a slightly greater footprint but well within the licence area and so he considered the issue “of negligible concern”.

938 §1.E(6).

capitals in original.

940 EIS Figures 63 and 64.

941 EIS §3.3.2.

942 One with and one without navigation marks.

943 Schedule 4.

944 Schedule 5.

945 Appeal p3. It has to be said that the precise proposal in the appeal was somewhat unclear – It was really an argument for flexibility as to cage numbers.

946 RPS Report 2015, Figure 4.2: Proposed pen locations, Shot Head, Bantry Bay.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

569. MOWI’s NIS of July 2020 stated:

“Note that the number of pens on the site may now vary to a maximum of 18, in order to assist the production process. However, production tonnage and the number of fish held on the site as a whole will not increase as a result of the increase in pen numbers. Experience in the years since the publication of the EIS in 2011 has demonstrated to the applicant that increased pen numbers provides a more sustainable way of maintaining the organic status of the stock on the site. This is now achieved by stocking each grow out pen with a fixed number of smolt, which will not be graded or moved, as was the case in the past, except for any standard treatments required, which involve the use of wellboat tanks, prior to harvest. This approach reduces stress on the stock and its consequences, including mortality and interruption of growth. Up to two of the pens installed on the site will be reserved for fish husbandry and harvesting procedures.”

“…. the limits of the grid frame which moors and supports the floating pens, will be no greater than 630m x 140m (if a maximum of 18, 126m circumference pens are deployed). This will cover a sea surface area of 8.82ha. The grid frame is moored, via grid buoys to the seabed by mooring ropes and seabed anchors, to a depth of about 35m. These extend the seabed area partially or fully obstructed by the installation to some … 24.9ha, some 60% of the licensed area applied for; see Figure 4.1.”

Unfortunately, Figure 4.1 shows the original 12-pen arrangement but I do not think that fatal as the text is clear.

570. The Sweetman Applicants’ written submissions947 complain that:

• it was only in September 2000, on seeing the NIS of July 2020, that the public “other than the parties to the 2015 appeal” learnt of the proposed 18-cage development and increase of the cage footprint from 5.88ha to 8.82ha and were “invited to participate in the EIA process”.

• the environmental impacts from the 18-cage proposal would materially impact a significantly wider seabed area of 8.82 hectares.

• The Non-Technical Summary of the EIS was not amended to reflect the changed project description.

571. MOWI reply948 that this issue was never pleaded. The Sweetman Applicants rely in turn on their plea of “Failure to prepare a Non-Technical Summary”. MOWI have the better of this dispute. Transmuting a clearly factually incorrect and bald assertion that no NTS was prepared of the EIS to an assertion that one was in fact prepared but was not later amended as to the description of the project and thereby that public participation was unlawfully degraded is impermissible under pleading rules in judicial review. I therefore reject the argument.

947 §25.

948 MOWI Submissions in the Sweetman Proceedings §11.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

572. Further while, as far as I can see, the Sweetman Applicants are correct that it was only in September 2000, on seeing the NIS of July 2020, that the public learnt of the increase of the cage footprint from 5.88ha to 8.82ha. But though they and the public had opportunity to do so, no-one made submissions to ALAB in this regard. That being so I do not see that the Sweetman Applicants can complain in these proceedings, whether on their own account or on the basis of a jus tertii, of late notification on the point. And, given that the requirement of description of the nature size and location of the project is relevant in terms of any significant environmental effect, it bears observing that there is no evidence to any effect that the change made a difference environmentally.

573. Environmentally, the main posited significance of the increased number of cages and cage footprint is that waste deposition on the seabed beneath the cages will degrade a larger area of benthos. In this regard, ALAB in its determination and at a point at which it clearly was considering an 18-cage licence,949

• accepted that there will be a localised adverse impact on the benthic community beneath the cages.

• noted that the benthic community is unremarkable and locally common and there are no concerns for rare or vulnerable species ALAB generally adopted, inter alia, the Saunders final report.950 Accordingly, the foregoing findings can be read with Dr Saunders’ advice that the deposition of settleable discharges will degrade the seabed communities, including shellfish, immediately beneath the cages (most the sessile epifauna will be lost), but given their unremarkable nature, this will be a small loss that will have no significant impact on the benthic ecology of Bantry Bay as a whole.951 He remained of the view that the issue was “of negligible concern”.952

574. Remembering that EIA does not deem environmental effect acceptable or unacceptable – that is for the authority competent in development consent – here ALAB – it seems to me that, generally, ALAB was entitled in law to accept Dr Saunders’ advice. In light of his view that the issue is of negligible concern, and remembering that the obligation to describe the project in the EIS/EIAR is with its environmental effects in view, I cannot say that ALAB’s EIA should be invalidated for inadequate identification of the project by reason of the increase in cage numbers or their footprint. I refuse relief in this respect.

575. However, the issue of effect on benthos will require further consideration in this judgment for Water Framework Directive purposes in due course.

949 §6.1.8, §6.5.1.

950 ALAB Determination §§7.2 & 7.4.

951 Saunders Final Report 8 December 2021, §6.1 Site Suitability. Also §§6.5.4, 6.8 9.6 (p80).

952 Saunders Final Report 8 December 2021, p104.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

EIA – AQUACULTURE LICENCE – CUMULATIVE IMPACT

576. IFI CG4 and 9 plead that ALAB’s EIA was inadequate in failing to assess cumulative dangers to marine life and/or give reasons for its conclusions in that regard. In substance, this plea relates to the alleged cumulative risk posed by sea lice. In particulars, IFI pleads that its appeal to ALAB had:

• alleged that MOWI has not assessed the cumulative effect of all existing and proposed Salmon Farm production in the bay.

• alleged that effective sea lice control would not be possible during harvesting.

• argued that, for lice control, any licence should be conditional on a synchronous stocking strategy, and a production cycle/regime synchronised with those at other salmon farms in the bay.

577. IFI pleads that ALAB granted the licence despite Dr Saunders, in his final report, having advised that cumulative impacts had not been considered. As will be seen, that is an incomplete account of Dr Saunders’ Final Report.

EIA – Cumulative Impact – Dr Saunders’ Final Report 8 December 2021

578. Dr Saunders in his Final Report,

• notes objections, including by SWI953 and IFI,954 that cumulative effects of multiple aquaculture installations in Bantry Bay had not been assessed. o I observe that it is clear that Dr Saunders appreciated the necessity of assessment of cumulative effects.

• records that “the cumulative impacts of the Shot Head site have not been specifically955 considered in the EIA956 or EIS. The addition of a farm site in Bantry Bay would be expected to increase the total number of lice present in Bantry Bay.”957

• records, under the heading “Cumulative Impacts” that “The cumulative impacts per se958 were not directly959 addressed in the EIS, nor subsequently specifically assessed in the EIA.960 Some key cumulative impacts have, however, been evaluated in EIS Section 4.6 and subsequently by further hydrographic

Dr Saunders’ Final Report 8 December 2021 p13.

954 Dr Saunders’ Final Report 8 December 2021 p17.

955 Emphasis added.

956 i.e. the Minister’s EIA.

957 Dr Saunders’ Final Report 8 December 2021 p64.

958 Emphasis in original – underlining added.

959 Emphasis added.

960 i.e. The Minister’s EIA.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

modelling (RPS, 2015). These include consideration of the most significant potential cumulative impacts to Bantry Bay, which relate to sea lice management, together with nutrient and pesticide discharges.”961

• concludes, under the heading “Cumulative Impacts” that “whilst not addressed specifically or directly in the EIA, the most significant potential cumulative impacts have since been addressed by the best available hydrological modelling. The results indicate that the Shot Head site, when considered within the context of all aquaculture operations within Bantry Bay will not contribute to a significant in-combination environmental impact.” 962

o I observe that, taken together, these passages acknowledge formal deficiency of the EIA and EIS as to assessment of cumulative effects but assert substantive sufficiency of the information to hand as to such effects. Notably, the RPS report, explicitly investigates “in-combination effects” of possible interactions of the Shot Head salmon farm with others in Bantry Bay and identifies those other fish farms.963 It seems to me that here, the views of Sharpston AG in Boxus964 and McGovern J in Ó Gríanna #2965 apply. What matters is not formalism or excessive pedantry and the EIA Directive need not be construed and applied “in the most onerous manner possible”. What matters is the substance of effective EIA.

• opines that “a cumulative intensification of any disease spread to wild stocks in Bantry Bay is not expected.

… No significant risk of increased fish diseases within the bay is expected.”966

• Notes that there may be a cumulative impact regarding lice loading in the bay, the details of which are discussed elsewhere in his report.967

• States that o the sufficiency of hydrographical distance between rivers, existing farm sites, and the proposed Shot Head site to prevent lice cross-infection and o the fact that that lice numbers have not triggered treatment in Bantry Bay since 2008, suggest that, while complacency is to be avoided, lice are “a manageable risk to both farmed and wild salmon”.

961 Dr Saunders’ Final Report 8 December 2021 §9.11 p95.

Dr Saunders’ Final Report 8 December 2021 §9.11 p96.

963 RPS Report 2015 §5.1.3.

964 Joined Cases C-128/09, C-129/09, C-130/09, C-131/09, C-134/09 and Joined Cases C-128/09, C-129/09, C-130/09, C-131/09, C-134/09 and C-135/09

Boxus et al v Région Wallonne Opinion of Sharpston AG of 19 May 2011.

965 Ó Gríanna v An Bord Pleanála (No. 2) [2017] IEHC 7. See also Kelly, Eoin v An Bord Pleanála [2019] IEHC 84 and Fitzpatrick v An Bord Pleanála [2017]

IEHC 585.

966 Dr Saunders’ Final Report 8 December 2021 p74.

967 Dr Saunders’ Final Report 8 December 2021 p74.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

EIA – Cumulative Impact – ALAB’s Determination

579. ALAB’s determination records that

• Appellants had raised issue of cumulative impact.968

• Dr Saunders had addressed cumulative impact as to sea lice.969

• ALAB considered that “the Site is hydrologically isolated from adjacent main rivers and other fish farms and will therefore present an overall low sea lice infestation and pollution risk”.970

• ALAB had considered cumulative impacts and determined, when considered within the context of all finfish aquaculture operations within Bantry Bay, the addition of the Site will not contribute to a significant cumulative ecological impact relating to sea lice burden. It considered the monitoring plan971 and the Pest Management Plan sufficient to reduce any risk to a reasonable, non-significant level.972

• ALAB had considered the issue of overall ecological cumulative impacts and determined, when considered in the context of all aquaculture operations in Bantry Bay, that the addition of the Site will not contribute to a significant cumulative ecological impact.973

• As to Water Framework Directive issues ALAB considered and accepted the Water Modelling Report results indicating that “the Site, when considered within the context of all aquaculture operations within Bantry Bay will not contribute to a significant cumulative environmental impact.”974

• As to AA issues, ALAB had considered issue of “in-combination” effects975 and concluded that:

“the proposed fish farm is not likely to have any deleterious effect, either individually, or in combination with other plans or projects, on the qualifying features of any of the designated sites within or related to Bantry Bay as a result of habitat use and there is no potential for significant effects and it is not likely to have a significant effect on such designated sites, either individually or in combination with other sites, plans or projects.”

“the proposed activity at the Site has no potential for significant effects and it is not likely to have any significant deleterious effect, either individually, or in combination with other plans or projects, on SCI species or conservation objectives for the connected SPA and SAC sites concerned and as such,

ALAB Determination §3.13.

969 ALAB Determination §6.5.11.

970 ALAB Determination §6.1.2.

971 pursuant to Monitoring Protocol No.3 for Offshore Finfish Farms – Sea Lice Monitoring and Control, compliance with which is a licence condition.

972 ALAB Determination §6.5.11.

973 ALAB Determination §6.5.11.

974 ALAB Determination §6.6.4.

975 ALAB Determination §§2.41, 5.1, 5.3, 5.4, 5.5, 5.6.3, 5.6.6, 5.6.8. 5.7, 7.4.3.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

will not adversely affect the integrity of the connected SPA sites concerned either individually or in combination with other plans or projects”.

EIA – Cumulative Impact – General Conclusion

580. While I will address some specific issues below, including those as to sea lice, in my view it cannot be said in any general sense that ALAB failed to consider cumulative effect in EIA.

LICE – INTRODUCTION & RELATION TO WILD SALMON & FWPM

581. The issues as to adequacy of EIA relate largely to sea lice976 and their effect directly on wild salmon and indirectly on the FwPM. I hope it will aid comprehension of what follows to pause here to explain a little of the environmental significance of sea lice. Sea lice infestation of salmonids is relevant in this case for both its alleged:

• direct effect on wild salmon populations (the primary real concern of angling and like interests).

• indirect effect on FwPM recruitment in the Dromagowlane/Trafrask river system.

582. In an early and essential stage of the FwPM life-cycle, its glochidia larvae977 encyst themselves in the gills of juvenile salmonid hosts – migratory wild salmon and sea-trout and/or non-migratory brown trout. This enables both their physical development and, by the movement of the fish, their dispersal in the river. Dr Good of the NPWS says978 – it is not disputed – that

• the encysted larvae feed on fish mucus or blood, grow on the fish and fall off about 4 weeks later to continue growing in gravels on the riverbed.

• while growing on the riverbed for about 5 years FwPMs are particularly vulnerable. If there is sediment cover or phosphate increase which uses up the oxygen in the gravels, “they very likely die

• the “main conservation issue” with the FwPM is “phosphate and siltation”. That said, his evidence was that, as to phosphate and siltation risk, the Dromagowlane/Trafrask river system were not known to be badly affected.

583. But FwPM reproduction and life cycle is also and clearly in part dependent on the population size and wellbeing of host salmonids in the river and, the objectors suggest, diminution in the number of

976 There are various species of sea louse. The EIS, Vol. 1 §5.1 states that the marine louse, Caligus elongatus parasitises many marine fish species including

salmon. The salmon louse, Lepeophtheirus salmonis is more euryhaline in habit and is a parasite specific to salmonids in brackish to fully marine

conditions. L. salmonis is the more problematic of the two species for both wild and farmed salmonids. “euryhaline” means able to tolerate a wide range

of salinity. For the greater part the papers do not distinguish these two species and neither will I.

977 “glochidia”.

978 Dr Jervis Good, NPWS Oral hearing 14 February 2017 transcript p144 - 147.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

potential salmonid hosts in a river (for example by lice-induced mortality), diminishes opportunities for FwPM recruitment. Dr Good of NPWS cited authority979 (as to the FwPM generally rather than specific to the Dromagowlane/Trafrask river system) that:

• the long term survival of the FwPM depends ultimately upon host availability.

• sea-trout are more suitable hosts than resident trout.980

• salmon farms pose a main threat to wild fish hosts to FwPM in Scotland.981

• a reduced fish population could have an immediate effect on the viability of a FwPM population.

• in response to an EU Commission complaint of loss of host fish in three FwPM habitats, Jackson et al982 had found overwhelming evidence that declines were caused by sedimentation and eutrophication983 and no evidence of declines due to changes in salmonid populations. Dr Good observed that this evidence related to specific habitats – not the Dromagowlane/Trafrask river system – and cited the need to consider cumulative effects.

584. Sea lice affect migratory984 salmonids – salmon and sea-trout. The young fish migrate from rivers to

the sea in the spring – their “susceptible” period for lice infestation in the brackish to marine waters in which

they “smoltify” – undergo the physiological changes required for the move from fresh to salt water. In the natural order of things, wild salmon returning to their natal rivers to breed bear egg-bearing sea lice which produce larvae which, in greater or lesser degree, infect the next generation of smolts about to go to sea.985 Lice parasitism induces both morbidity and mortality in infected salmon. Lice generally do not affect brown trout, which spend their entire life-cycle in freshwater. So, as a general proposition, if the FwPM glochidia larvae in the Trafrask River primarily use brown trout as hosts, sea lice present a far lesser indirect threat to FwPMs than if those larvae primarily use salmon or sea trout as hosts. The corollary also applies.

585. Wild salmon populations in Ireland have fallen dramatically over many years.986 It seems that mortality at sea for salmon in their first migration is high and only about 5% of smolt return to their natal rivers to breed. What is disputed is why. Many causes are suggested.987 Sea lice are one. In farmed salmon sea lice infestation has significant fish-health and economic consequences. The concentration of large numbers of salmon in cages may amplify any lice infestation (from the wild or from other salmon farms). Anglers and others are convinced that salmon farms breed lice in large numbers which, by dispersal of sea

979 'The Ecology of the Freshwater Pearl Mussel', English Nature series 'Conserving Natura 2000 Rivers', 2003; Irish Wildlife Manual #8 “Conservation

Management of the Freshwater Pearl Mussel” Moorkens 1999;Osterling and Soderberg, 'Sea Trout habitat fragmentation affects threatened

freshwater pearl mussel”, Biological Conservation Vol 186, 2015.

980 Brown trout are resident trout.

981 Cosgrove et al, Journal of Conchology Volume 41 (2014) “Population size, structure and distribution of an unexploited freshwater pearl mussel

population in Scotland”.

982 Jackson et al. Irish Fisheries Bulletin #43 (2013) 'Report on sea lice epidemiology and management in Ireland with particular reference to potential

interaction with wild salmon and freshwater pearl mussel populations'.

983 i.e. phosphates.

984 “anadromous”. Dr Saunders’ Final Report of 8 December 2000 records that sea louse infection of wild populations largely occurs in spring when salmon

and trout smolts migrate to sea. In contrast to wild salmon, which migrate to the open ocean, wild sea trout largely remain residents of sea lochs and

estuaries and are consequently more exposed to sea lice infection.

985 The mechanisms are complex and not entirely understood. See sEIS 2018 §2.3.3.

986 E.g. sEIS p47.

987 One view on these issues is provided in MOWI’s sEIS p52 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

lice larvae988 from salmon farms, result in greatly increased lice infestation of wild salmon in the same waterbody, as compared to natural infestation levels. MOWI in effect agree that, uncontrolled, salmon farms may act as excellent lice breeding grounds989 but it disputes that from there sea lice infect wild salmon. The quantum of wild salmonid mortality due to lice, the contribution of lice from salmon farms to that mortality and the efficacy of lice control in salmon farms are all very highly contentious issues – not least as between reputable experts.990

586. Copepodid larvae are the primary infestive stage of the sea louse life-cycle.991 Their dispersal in coastal waters is “largely a matter of chance and hydrography, as … copepodids drift, in diminishing

densities,992 from their birth-site.”993 As stated, many issues relating to sea lice infestation and its effects are

very controversial and disputed – including, but not limited to, highly divergent views as to:

• the efficacy of lice control and treatment in salmon in farms.

• larval dispersal patterns in concentrations at which and to locations at which the endemic risk of lice infestation of wild salmon may or may not be appreciably increased.

• salmon mortality rates due to lice infestation.

• whether salmonid mortality rates and population decline affect FwPM recruitment.994

• whether and to what extent, in the case of the Dromagowlane/ Trafrask river system, juvenile migratory salmon and/or sea trout, as opposed to juvenile brown trout, play a significant role as hosts to FwPM larvae such that a sea lice induced diminution in the population of juvenile migratory salmonids would damage FwPM recruitment.

I emphasise that while I record these contentions as they are not disputes of law I cannot resolve them in these proceedings.

988 First as nauplius larvae, then as copepodid larvae.

989 Bass/Shannon PowerPoint of September 2017 orally presented to the oral hearing and entitled “Numerical Modelling of the dispersion of wastes,

medication and salmon lice … from the Proposed MHI Shot Head site in Bantry Bay”. P21 – “stocking densities on fish farms can provide the critical host

densities and fixed location required for successful infestation by very low densities of Copepodids, drifting in the plankton.”

For example, Dr Jackson of the Marine Institute (S.47 Reply 15 February 2019) asserts that while lice induced mortality can be significant in outwardly

migrating smolts it causes only about 1% of mortality in wild salmon post-smolting and is a minor and irregular component of marine salmon mortality.

He considers it to be unlikely to be a significant factor influencing the conservation status of salmon stocks. He asserts that he found “no correlation

between the presence of aquaculture and the performance of adjacent wild salmon stocks”. In notable contrast, Dr Gargan of IFI (retired) cites “a very

large body of literature available in Ireland, Scotland and Norway on the impact of sea lice from salmon farms on both salmon and sea trout stocks” to

what he considers a contrary effect. He states, (Affidavit 30 September 2022) that “Statistical models suggested that returns were >50% lower in years

following high lice levels on nearby salmon farms during the smolt out-migration indicating a very strong causal link between sea louse infestation, and

wild salmon mortality.”

991 Preadults and adults are mobile on their host fish and if detached can swim for short periods providing the possibility of infecting other fish.

992 i.e. diminishing with distance from salmon farms.

993 Supplemental EIS p16.

994 For example, Dr Jackson of the Marine Institute (S.47 Reply 15 February 2019) asserts, though citing only a personal communication with the NPWS,

that there is no evidence that changes in salmonid populations have contributed to the current unfavourable status of the freshwater peal mussel in

Ireland and that NPWS studies provide overwhelming evidence that declines were caused by sedimentation and eutrophication of habitats.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

IFI CG4 & PARTICULARS THEREOF

587. As recorded earlier, IFI pleads ALAB’s irrational reliance on factual and scientific conclusions without proper factual or scientific basis in the face of contrary evidence, which ALAB ignored, and including:

• A sea lice dispersal model erroneously premised on the neutral buoyancy of sea lice.

• Its conclusion of no negative impact on wild salmonids, ignoring the consolidated academic writings to the contrary – for ignoring which it failed to give adequate reasons.

• Failure to require single or synchronous bay management, including synchronous whole bay fallowing given evidence of likely cumulative environmental effect with other salmon farms.

• Failure to adequately consider Dr Saunders’ advice that the Shot Head farm will increase average lice numbers in the Bay and risks to wild salmonids accordingly, including those in the Dromagoulane/Trafrask Rivers, with consequent risks to the FwPM in those rivers.

• Its conclusion that most FwPM larva hosts would be brown trout – which was speculative.

588. I generally observe that these complaints are largely as to the merits of the judgments made by ALAB on the evidential and like materials before it. That is why the plea is framed in terms of irrationality – as to their substance/merits it is only for irrationality that such judgments can be impugned. Without getting into issues of the standard of review for irrationality,995 it is clear on all views of such issues that the threshold to show irrationality requires “something overwhelming” (O'Keefe and Keegan996) – the Keegan test of “fundamental variance from reason and common sense” represents a very high bar to certiorari which is not surmounted even by a judicial view that an impugned decision was “clearly wrong” on its merits (Holohan997 and Jennings998). The threshold “is extremely high and is almost never met in practice” (St. Audoen’s999). Importantly, while all relevant evidence for and against a particular administrative decision must be considered, and a decisionmaker must engage (including in its reasons) with all submissions of substance, preferring evidential and like materials supportive of the grant of a licence does not imply that evidence to contrary effect has been ignored. Humphreys J was not laying down new law when he cited “…. the hallowed applicants’ fallacy of confusing lack of narrative discussion with lack of consideration. A

decision-maker normally considers what is before her even if not expressly referred to.” – Killegland.1000 It is

also, at least generally, perfectly rational to consider, from a standpoint of expertise, that some materials (such as academic papers) might better have been included in an application but yet that, overall, sufficient information is to hand allow of a decision and to justify the grant of a licence.

995 See for example, discussion in Jennings & O’Connor v An Bord Pleanála & Colbeam 2023 IEHC 14 §15 et seq.

996 Finlay CJ in O'Keeffe v An Bord Pleanála [1993] 1 I.R. 39 citing Henchy J in The State (Keegan) v Stardust Compensation Tribunal [1986] I.R. 642 citing

in turn citing Lord Greene M.R. in Associated Provincial Picture Houses Limited v Wednesbury Corporation [1948] 1 K.B. 223.

997 Holohan v An Bord Pleanála [2017] IEHC 268.

998 Jennings & O’Connor v An Bord Pleanála & Colbeam 2023 IEHC 14.

999 The Board of Management of St. Audoen’s National School v An Bord Pleanála [2021] IEHC 453. 1000 Killegland Estates Limited v. Meath County Council [2022] IEHC 393 §207. Upheld [2023] IESC 39.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

LICE – SBM, CLAMS – SITE ALTERNATION, SYNCHRONOUS STOCKING AND HYDROLOGICAL ISOLATION

589. While not always so presented in the case, it seems to me that the issues of sea lice and Single Bay Management (“SBM”) are in practical terms inseparable. The latter is the posited means of combatting the former.

The Licensed Production Cycle, SBM & CLAMS

590. As proposed by MOWI, salmon production takes 22 months from transfer of smolts1001 to the farm in October/November to end of harvesting. Harvesting starts in March of the 2nd year – at about 17 months from transfer to the farm – and is completed in the 5 months to August.1002

591. Generally and unsurprisingly, it seems to be in aquaculture producers’ commercial interests to achieve annual, as opposed to bi-annual, harvesting. At its simplest, annual harvesting in a 24-month production cycle,1003 while separating generations of salmon as a lice-control measure,1004 requires two similarly-sized farms producing in alternate years. MOWI proposes and intends production asynchronously with its other salmon farm at Roancarraig/Aghabeg in Bantry Bay1005 to enable annual harvest alternately from each site.

592. MOWI’s licence application is clear that it intends Single Bay Site Alternation and not Synchronous Stocking.1006 In fact asynchronous production with the Roancarrig/Aghabeg site can fairly be described, from MOWI’s point of view as the raison d’etre of the Shot Head salmon farm, to achieve annual as opposed to biannual production. Dr Saunders clearly and correctly so understood the MOWI application.1007 ALAB did not impose synchronous production. In my view, the Aquaculture Licence must be interpreted in those lights. Given the terms of the MOWI application and Dr Saunders’ advice, had ALAB intended to impose synchronous stocking and production at Shot Head, it would have had to say so clearly in its Determination and explicitly impose a condition in the Aquaculture Licence accordingly. It did not.

1001 As stated earlier, in the wild smolt are young salmon moving from freshwater to seawater and undergoing physiological changes accordingly. In

MOWI’s farming system a site elsewhere is used rear smolt, from transfer from freshwater to seawater to an intermediate weight, for up to a year, when

they are transferred to a salmon farm. 1002 EIS Vol 1 §3.1. 1003 Assuming 2 months fallowing. 1004 i.e. each farm containing only one generation of salmon. 1005 It lies northeast of Bere Island, south west of Adrigole Bay and west of the proposed Shot Head salmon farm. 1006 EIS 2011 Volume 1 §4.5. & §10. 1007 Technical Advisor’s Final report p64.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Fallowing & the Fallowing Protocol 2000

593. At the end of each harvest, each site is to be fallowed to try to get rid of residual sea lice and larvae. The DAFM Fallowing Protocol 20001008 identifies fallowing as technique for the control of disease and parasite problems (including sea lice infestation).1009 Fallowing a site between harvesting one generation and stocking the site with the next separates successive generations of fish farmed on a particular site, preventing the transmission of diseases and parasites between generations. Fallowing assists in eradicating sea lice by depriving them of hosts in the farms. For fallowing to be effective, it must be of sufficient duration and either farm sites must be sufficiently separate to prevent lateral transfer of infective material between sites or “adjacent sites” must fallow at the same time. So, as a general proposition and depending on local circumstances there can be merit, from the point of view of lice control, in synchronising production as between fish farms in a bay to ensure that they all fallow at the same time. That tends to reduce recolonisation of recently-fallowed farms by lice from farms not fallowed at the same time. As licensed and as their harvests end in alternate years, it follows that the Shot Head and Roancarraig/Aghabeg farms will be fallowed in alternate years – not synchronously.1010 So the effectiveness of fallowing will depend on whether they are “adjacent”. The concept of “adjacent sites” is not elucidated in the Fallowing Protocol 2000. However, as I hope to demonstrate below, adjacency is related to the concept of hydrological isolation of sites, such that sites with a single tidal excursion of each other are considered adjacent. It seems clear that by that criterion, the Shot Head site is not adjacent to any other salmon farms.

594. The DAFM Fallowing Protocol 20001011 states “Where there is more than one finfish farm in a particular bay, fallowing shall be pursued in the context of the Single Bay Management/CLAMS process. The Department1012 … reserves the right to prescribe specific fallowing requirements in particular cases if necessary.” It is not clear, at least on the information to which my attention has been drawn, that there is a legal basis on which DAFM asserts a right to “prescribe specific fallowing requirements” – other than as deriving from aquaculture licence conditions.

595. However Schedule 5 of the Impugned Aquaculture Licence stipulates:

• “Adhere to the sea lice monitoring and control protocols as set out in the Strategy for Improved Pest Control on Irish Salmon Farms (2008)”.1013

1008 Protocol for Fallowing at Offshore Finfish Farms 11 May 2000. 1009 The DAFM Lice Strategy 2008 describes fallowing as follows:

“Fallowing is a tool used to control the level of sea lice, benthic conditions and the spread of fish disease. To be effective it is dependent on a satisfactory

length of time for fallowing and appropriate geographical separation between sites and/or synchronous fallowing of adjacent sites. The Protocol on

Fallowing essentially establishes the principle of fallowing and best practice in fallowing. All finfish farms subject to the Protocol are obliged to undertake

appropriate fallowing for the control of disease and parasite problems (including sea lice). Where there is more than one finfish farm in a particular bay

the protocol requires licensees to pursue fallowing in the context of the Single Bay Management process. The Protocol specifies a minimum period of 30

continuous days for fallowing an individual site, …”

1010 In its ordinary and in its dictionary meaning, synchronising means “the fact of happening at the same time, or the act of making things happen at the

same time”. SYNCHRONIZATION | English meaning - Cambridge Dictionary. 1011 Protocol for Fallowing at Offshore Finfish Farms 11 May 2000 §3, p2. 1012 Of the Marine and Natural Resources.

1013 Note that this Condition does not require compliance with the Strategy more generally – as opposed to compliance with the Protocols it cites. This is

unsurprising as it is a strategy document not suited to direct implementation in specific salmon farms. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• “Comply with such protocols, including in relation to monitoring, auditing and any aspect of managing an aquaculture site, as may be published by the Minister”.1014 Other conditions of the Impugned Aquaculture Licence require compliance with:

• §6.5 the Fallowing Protocol 2000.1015

• §7.3 the Sea Lice Protocol #3 2000.1016

• §12.1 “all laws and protocols applicable to aquaculture”.

• §12.5 the Audit of Operations Protocol 2000.1017

596. As no issue was raised in that regard, I need not consider here the efficacy of a licence condition requiring compliance with protocols yet to be published at the time the licence is granted as it might be considered to give the Minister carte blanche, in effect, to amend the licence. However, SBM and Coordinated Local Aquaculture Management Systems (“CLAMS”) are clearly voluntary processes which the Minister cannot impose. It follows that, while the Fallowing Protocol 2000 states that “fallowing shall be pursued” via SBM and CLAMS, the following sentence must be interpreted as entitling DAFM to “prescribe specific fallowing requirements” compulsorily and outside SBM and CLAMS. As to the Shot Head site, Condition 6.5 of the Impugned Licence provides the legal basis for his doing so. Non-compliance with any such Departmental prescription is breach of condition of the Aquaculture Licence.

Meaning of Synchronicity

597. For the avoidance of doubt,1018 I should say that I understand “synchronicity” (and cognate words) in the present context to mean doing the same thing at the same time. It may imply co-ordination of action but is more specific than mere co-ordination – which could consist, for example, in doing different things at different times in accordance with a plan. It follows that I understand “asynchronicity” (and cognate words) in the present context to mean at least the possibility of doing the same thing at different times.

SBM & SBMPs

598. Alternate-year asynchronous production as between Shot Head farm and MOWI’s Roancarraig/ Aghabeg farm raises issues for reconciliation of commercial interests and production models on the one hand with SBM on the other. SBM raises the issue of the intended relationship of the production cycle of the proposed Shot Head farm to that of all other salmon farms in Bantry Bay.1019

1014

All protocols are explicitly “subject to revision from time to time” but, as far as the evidence discloses, and despite the DAFM Sea Lice Strategy

2008, the Fallowing Protocol, the Sea Lice Protocol and the Audit of Operations Protocol do not appear to have been reviewed since published in 2000.

1015 Protocol for Fallowing at Offshore Finfish Farms 11 May, 2000. 1016 Monitoring Protocol No. 3 for Offshore Finfish Farms- Sea Lice Monitoring and Control – 11 May 2000.

1017 Monitoring Protocol No. 4 for Offshore Finfish Farms – Audit of Operations. It prescribes integrated assessment of finfish farm operations, inter alia

to allow the DAFM to monitor compliance with the licence. 1018 I appreciate there may be none. 1019 It lies northeast of Bere island, south west of Adrigole May and west of the proposed Shot Head salmon farm.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

599. Broadly, the ideas of SBM and Single Bay Management Plans (“SBMP”) are easily understood. They aim at disease control and specifically at sea lice control. They are the means of carrying certain DAFM protocols into effect – including Sea Lice Protocol #3 2000 and the Fallowing Protocol 2000.1020 Importantly, SBM is voluntary. That SBM was described in MOWI’s EIS in 2011 as an “aspiration adopted in Ireland some years ago”1021 and by Dr Saunders as “a non-statutory aspiration in CLAMS”1022 may be telling given the DAFM Lice Strategy 20081023 describes SBM as having been progressively introduced from as long ago as 1991. There is no SBMP in Bantry Bay. SBM requires an SBMP to be adopted by agreement between all salmonid farmers in the bay in question to co-ordinate some or all of the following elements:

• Stocking/Production cycles.

• Sea lice monitoring and treatment.

• Fallowing of farm sites.

600. However, it is difficult to pin down what exactly (if indeed anything) is essential to an SBMP. That is not, per se, a criticism of SBM. But it may bear on questions of certainty and enforceability of licence conditions adopting a concept based on agreement. There is dispute whether in all circumstances and/or to what extent SBM requires that production cycles be not merely co-ordinated but synchronised and whether all the elements described above must be included in an SBMP. For example, MOWI assert that SBM does not necessarily require synchronous stocking and fallowing.1024 What exactly is essential to an SBMP is complicated further by MOWI’s and ALAB’s plea that the Shot Head farm will be “hydrologically isolated” from other farms within the bay such that they do not require synchronicity of one kind or degree or another.

601. SBM, SBMP and CLAMS are, it seems, nowhere clearly and authoritatively defined.1025 They are best described, at least in the materials to which I properly have access, in the DAFM Lice Strategy 2008, as involving all fish farms in an area co-operating to develop an Integrated Pest Management Plan. For present purposes, “pest” means sea lice. The DAFM Lice Strategy 2008 was a response to then-recent problems with sea lice – a very resilient pest which had shown itself able to rapidly develop resistance to the limited range of veterinary medicines available to treat it. As a strategy, it is not a document intended to be directly enforceable. It envisaged intensified efforts “to revitalise the single bay management approach and make it central to national policy for sea lice management”. This does not suggest that, as of 2008, SBM was in reality already vitalised or central in practice. It did contain recommendations and an action plan with an objective to bring progressively tougher actions to bear on infestation to ensure the highest possible level of

1020 Protocol for Fallowing at Offshore Finfish Farms 11 May 2000. 1021 EIS 2011, Vol 1 §3.2.2. 1022 Dr Saunders Final Report 8 December 2021 §9.11, p95.

1023 A slightly inaccurate sobriquet, but useful for present purposes. It refers to “A strategy for improved pest control on Irish salmon farms, May 2008”.

1024 Affidavit of Catherine McManus 23 January 2023.

1025 Dr Saunders Final Report refers to an undated BIM Co-ordinated Local Aquaculture Management Systems (C.L.A.M.S.) Explanatory Handbook. But it

is not exhibited. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

compliance. I am not aware of what came of these recommendations and the action plan. But the DAFM Lice Strategy 2008, did not include revision of the Sea Lice Protocol #3 2000, which is cited in, and was appended to the Strategy and remains current.1026 Dr Jackson, Fisheries Inspector of the Marine Institute and a coauthor of the Minister’s EIA, testified at the oral hearing that the DAFM Lice Strategy 2008 was followed by “a steady and sustained improvement in sea lice control” as at 2011 and also that wild salmon mortality due to sea lice was about 1%.1027 Those assertions were contentious but the contention is not for me to resolve. For present purposes, the importance of the DAFM Lice Strategy 2008 lies in its description of SBM.

602. The DAFM Lice Strategy 2008 identifies as “principal components”1028 and “crucial elements”1029 in the success of an SBMP as: separation of generations; annual fallowing of sites; strategic application of chemotheraputants; good fish health management; close co-operation between farms. However this is a strategy and it unsurprisingly suggests a flexible suites of responses which “need to be addressed” and in

which detail is lacking.1030 The Lice Strategy 2008 identifies “7 basic principles of best practice” 1031 as widely

agreed by all interested parties:

• 1) Complete separation of generations (sites to be one tidal excursion apart). o I understand this to mean that if sites are a tidal excursion apart it is SBM-compliant for them to raise different generations of salmon – which is what MOWI propose as between Shot Head and Roancarraig/Aghabeg).

• 2) Each site to be fallowed annually, or at end of a production cycle, for one month (30 days) before restocking.1032 All sites within one tidal excursion to be fallowed synchronously. (The strategy had earlier said that effective fallowing depends, inter alia, on “appropriate geographical separation between sites and/or1033 synchronous fallowing of adjacent sites”.)

• 3) Annual synchronous "winter" lice treatment for all adjacent sites (one tidal excursion). o I understand this to mean that:

▪ To be “adjacent”, sites must be within a single tidal excursion of each other.

▪ If sites are not adjacent – not within a single tidal excursion of each other – SBM does not require synchronous lice treatment.

1026 See below.

1027 Citing ICES WKCULEF Report 2016: Report of the Workshop to address the NASCO request for advice on possible effects of salmonid aquaculture on

wild Atlantic salmon populations in the North Atlantic. The “Key Findings of the Workshop included that “The survival of Atlantic salmon during their

marine phase has fallen in re-cent decades. This downturn in survival is evident over a broad geographical area and is associated with large-scale

oceanographic changes. Viewed against current marine mortality rates commonly at or above 95%, the ‘additional’ mortality attributable to sea lice has

been estimated at around 1%. In some studies, the impact of sea lice has also been estimated as losses of returning adult salmon to rivers. These

estimates indicate marked variability, with losses in individual experiments ranging from 0.6% to 39%. These results suggest that sea lice induced

mortality has an impact on Atlantic salmon returns, which may influence the achievement of conservation requirements for affected stocks.”

1028 p14. 1029 p17. 1030 §4.5. 1031 §4.1 1032 ALAB imposed 2 months. See ALAB determination §6.5.2 & Aquaculture Licence Schedule 4. 1033 Emphasis added.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• 4) Planned rotation of sea lice treatments over the production cycle & adjacent sites to use the same product rotation. o See my comment above.

• 5) Treatment triggers: in spring, 0.5 egg-bearing lice1034 per fish; rest of year, 2.0 egg-bearing lice.

• 6) All above to be set out as part of formal signed SBM Agreement.

• 7) Where there is a persistent problem with sea lice control there is a need for an incremental series of actions up to and including mandatory treatments and sanctions where these are not effectively implemented.

603. Also informative in understanding SBM is Sea Lice Protocol #3”, §5 of which requires lice treatment and control via SBM. It identifies “principal components” in terms generally consistent with the description of SBM: Separation of generations; Annual fallowing of sites;1035 Targeted treatment regimes, including synchronous treatments; Agreed husbandry practices. It describes 14 DAFM inspections1036 annually – twicemonthly in March, April and May. It identifies treatment triggers as integral to a treatment regime. Treatment triggers in the critical spring period1037 are 0.3 to 0.5 egg-bearing lice per fish. (The protocol does not identify which of 0.3 and 0.5 is to apply in what circumstances and 0.3 and 0.5 differ by 67% – but presumably a farm operating a 0.5 trigger can assert compliance). Also “high numbers” of mobile lice trigger treatment even absent egg-bearing lice – “high” is not quantified. Other than in spring, 2.0 egg-bearing lice per fish triggers treatment.

604. CLAMS is something of a sideshow in the case but repeatedly appears in the papers nonetheless. If only to dispel confusion it requires brief consideration here. ALAB’s Determination noted1038 that there is no CLAMS in Bantry Bay and that CLAMS (like SBM) is voluntary and is explicitly removed from the licensing and regulatory process and is not intended for use in the consideration of individual site licences. Dr O’Toole for ALAB has deposed, in both the IFI and SWI cases, that CLAMS is a voluntary agreement between all producers in a bay on best practice as to a number of issues, many relating to fish health, and including SBM. The DAFM Lice Strategy 2008 describes CLAMS as a non-statutory process driven by aquaculture producers, BIM and the Marine Institute, working within the framework of national policy and designed to facilitate the development of plans for individual bays incorporating and extending the concept of SBM.1039 It is envisaged as a framework, separate to the licensing process, for addressing issues that affect, or are affected by, aquaculture activities and for streamlining resolution of these situations. Beyond these general descriptions, and noting that it encompasses but is broader than SBM, I have failed to find a clear description of CLAMS in

1034 a.k.a. Ovigerous female lice.

1035 It is difficult to see how annual fallowing is reconcilable with two-year production cycles unless the salmon in each cycle are moved off site.

1036 Pursuant to a National Sea Lice-Monitoring Plan. 1037 When wild smolts are leaving the river for the sea. 1038 Appeal determination §6.8.7. 1039 It will also be integrated with Coastal Zone Management policy and County Development Plans.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

the papers. In my view, and not least as there is no CLAMS in Bantry Bay, for the purposes of this case the concept of CLAMS does not contribute to the determination of issues to which SBM is relevant.

Single Bay Site Alternation & Synchronous Stocking

605. The 2011 EIS1040 describes two production scenarios: Single Bay Site Alternation1041 and Synchronous Stocking.1042

• Single Bay Site Alternation in this case, and as licensed by ALAB, consists in asynchronous stocking of the Shot Head and Roancarraig/Aghabeg farms to each produce a harvest in every second year.

I observe that Site Alternation is commercially preferable but has potential disadvantages as it o puts more than one generation of fish in a bay at all times.

o means the entire bay cannot be fallowed simultaneously.

o renders synchronous lice treatment problematic – at least as to limitations on use of EmBZ if the latter, as under the Impugned Licence, is prohibited later in the production cycle.

These disadvantages, amongst other variables such as distance between sites and water currents, tend to increase risk of cross-infestation of lice as between farm sites.

• Synchronous Stocking requires stocking, harvesting and fallowing of synchronised sites in a bay at the same time. The 2011 EIS describes Synchronous Stocking as “more in line with Single Bay Management” as compared to Single Bay Site Alternation.

I observe that Synchronous Stocking has the significant lice control advantages that only one generation of fish is present at any time and the synchronised sites are fallowed at the same time and lice treatments can be synchronised. These features tend to break lice infection cycles – tending to prevent lice transfer between the synchronised sites and between generations of salmon.

Its disadvantage, it seems, is that it is commercially less attractive unless annual harvesting can be achieved by the producer’s alternating with a similar sized site in another bay. It also requires close co-operation between all producers in a bay, which may be difficult to achieve. Further, maximum standing stocks in the synchronised sites cumulatively reach almost double those for alternately stocked sites,1043 thus producing higher maximum combined discharges towards the end of the production cycle.1044

1040 §3.2. Production scenarios for Bantry Bay. 1041 §3.2.1. 1042 §3.2.2. Whole Bay Rotation, by which entire bays can be fallowed for extended periods, is also mentioned. 1043 EIS Figure 61.

1044 Presumably mean discharges are the same as between both scenarios, but the range in Synchronous Stocking is double that in Single Bay Site

Alternation. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

606. These issues were raised in the appeals to ALAB.1045 SWI argued that synchronous stocking is necessary to optimise the lice eradication effects of fallowing, but that the EIS and EIA were not clear that synchronised management is in place in Bantry Bay. FoIE1046 asserted that there was no evidence of collaboration between existing fish farm operators (including MOWI) and that this compromises sea lice control. FoIE and IFI argued that MOWI’s proposed asynchronous stocking and fallowing regime does not follow the best practice recommendation of the DAFM Lice Strategy 2008, agreed SBMPs and the principle of synchronous production. IFI asserts that licence conditions should require synchronous stocking.

Tidal Excursion, Hydrological Isolation and Adjacency of Salmon Farms

607. It will have been seen that, by the DAFM Lice Strategy 2008, SBM requires synchronicity between

farms within a single “tidal excursion” in a bay and that such sites are considered “adjacent”. By necessary

implication, the need for synchronicity does not, at least necessarily, arise as between farms separated by more than one tidal excursion – which are considered non-adjacent. Unfortunately, I have found in the papers no definition or description of a tidal excursion. No doubt it is a term understood by experts, including ALAB. But, as it is clearly very relevant to the need, or lack of it, for synchronicity between farms, its opacity is regrettable in a public environmental licensing process. And, as stated, MOWI and ALAB plead that Dr Saunders1047 and ALAB1048 concluded that “the site is hydrologically isolated from adjacent main rivers and other fish farms and will therefore present low sea lice infestation and pollution risk.”

608. The 2011 EIS says1049 that

• SBM is incorporated into CLAMS, “where these have been introduced. In both cases, their objective is to separate salmon farm sites into groups which lie within overlapping tidal excursions from those which lie in separate tidal excursions. Bantry Bay is the Single Bay Area containing the proposed Shot Head site, the MHI Roancarrig site and two sites operated by Fastnet Irish Seafood (see Section 4.6. and Figure 79).

• CLAMS has yet to be established in Bantry Bay.”

The source of the designation of Bantry Bay as a “Single Bay Area” is not identified – nor has it been since. In any event, if I understand the foregoing, it records that Bantry Bay is a “Single Bay Area” because all of the sites identified are within a single tidal excursion. If correct, that implies that SBM requires synchronicity between all sites in the bay.

1045 What follows is taken from the summary in the Saunders final report of December 2020

1046 Friends of the Irish Environment – a notice party to but not an active participant in the SWI and IFI proceedings.

1047 Final Report §§6.1, 6.8 & 10.2. 1048 Determination §6.1.2. 1049 EIS Vol 1 May 2011 p220.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

609. Though it does not deploy the precise phrase “tidal excursion”, the 2015 RPS Report considers “excursion” in various respects. It refers to examination of the excursion of released particles over 96 hour periods and shows, inter alia,1050

• the results for the particles which travelled furthest from the Site under the influence of the tide alone, that is at high water and low water during spring tide and

• that particles do not travel a great distance before being carried back on the returning tide.

My lay impression from the figures in their report is that RPS consider the Shot Head site to be hydrologically isolated, in terms of tidal excursion, from the other salmon farm sites and is borne out by later documents. In other words, in substance the 2015 RPS Report disagrees with the 2011 EIS that, for SBM purposes, Bantry Bay should be a “Single Bay Area”. It does not consider Bantry Bay to function as a Single Bay Area.

610. In his Interim Report of December 2016, Dr Saunders cites the 2011 EIS for the proposition that “Bantry Bay is a Single Bay Area” for SBM purposes.1051 However, while “an elevated loading of ambient lice in the Bay associated with the increased number of fish may be expected”, he also cites the RPS Report as recording that “on the basis of modern modelling techniques the site is hydrologically isolated from adjacent main rivers and other fish farms and will therefore present low sea lice infestation” and there is “adequate hydrological distance between sites” “to prevent lice cross-infection” such that more rigorous SBM (i.e. synchronicity) “may not be necessary at present” and the RPS Report “suggests that the lice issue constitutes a manageable risk to both farmed and wild salmon within the Bantry Bay catchment.” However, he notes the limitations of modelling and is concerned at the “lack of recognition” that the significant increase in salmon stock in Bantry Bay when Shot Head is operational can only increase lice numbers and associated fish health risks. In his view, this strengthens the case for considering a CLAMS/SBM strategy, which would include “a synchronous entire-bay production regime”. Yet he states that “The (asynchronous) production strategy proposed by MHI complies with SBM and is suitable for Bantry Bay”.1052

611. While, as I say, a clear definition of “tidal excursion” was eminently desirable, and while the materials available could have been clearer, nonetheless it seems from the foregoing taken together that:

• Whether one describes it in terms of hydrological isolation or tidal excursion, there is, in the view of MOWI’s experts and Dr Saunders what I may call a marine distance between salmon farms in the same bay beyond which SBM may not require synchronisation of production cycle, generations, fallowing and lice treatment.

1050 Figure 5.35. 1051 Saunders Interim report December 2016 p54. 1052 pp58, 59, 91, 92, 95

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• In that sense, the Shot Head site is hydrologically isolated such that, for SBM purposes and, at least on some expert views, the need for synchronicity with other sites is as yet absent – though it is recognised that events may yet require synchronicity.

Irrational Failure to Require Single or Synchronous Bay Management – Decision

612. While different views could have been, indeed are, taken as to whether synchronous single bay management in Banty Bay is required, I cannot see that a decision that it is not required is irrational in the sense required by law to justify certiorari. One need only cite the finding of hydrological isolation of the Shot Head Site. So the challenges as to that issue – notably IFI CG4 – fail.

LICE – DR SAUNDERS’ FINAL REPORT 8 DECEMBER 2020 – INCLUDING AS TO SBM SYNCHRONICITY

613. Generally, Dr Saunders’ Final Report is nuanced on the lice issue.1053 He observed that the potential impact of sea lice originating from salmon farms and infecting wild salmonids remains highly controversial. Despite scientific studies, the detailed mechanisms of sea lice interactions as between salmon farms and wild salmonids are far from clear, due to limitations in the ability to apply robust scientific method to the problem and the potentially confounding effects of unrelated mortality. That said, Dr Saunders gave a detailed account of the mechanisms of infection, of risk mitigation methods, of salmonid vulnerability in Bantry Bay, an assessment of the evidence for the impact of salmon farms on wild populations via sea lice and of the risk of lice transfer to wild salmonids in Bantry Bay.

614. Dr Saunders observed that the wider body of (conflicting) scientific literature as to dispersal of lice from salmon farms may not have been fully considered in MOWI’s assessment of the potential impact of the Shot Head site on lice infestation of wild salmon in Bantry Bay.1054 However he emphasised that, whilst studies are informative, local hydrographical conditions are key to understanding the risks specific to a particular water body. The hydrological conditions of Bantry Bay will differ from those of other water bodies and are characterised by relatively low residual currents.1055 He considered it particularly important that sea lice levels are not currently a significant problem in fish production in southwest Ireland, including Bantry Bay and, since 2008, lice in Bantry Bay salmon farms have been consistently below trigger levels1056 requiring

1053

Technical Advisor’s Final Report 8 December 2020 §9.1 Increased threat to wild salmon and sea trout from sea lice, p66.

1054 Page 64 of Technical Advisor’s Final Report.

1055 I understand tidal residual current to represent the net tidal movement of waters over a full cycle of flood and ebb tides. In Bantry Bay the residual

current runs anti-clockwise such that at the Site it runs out towards the sea. Low residual currents would tend to limit the net movement, (i.e. dispersal

distance) of lice larvae during their infestive lifespan.

1056 He observes that the Marine Institute Annual Sea Lice reports for 2015 to 2020, record that ovigerous lice levels breached the lower 0.3 threshold

value (but not the statutory 0.5 sensitive period Trigger Level) only once in 2015 with no recorded lice problems since.

since. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

few treatments. Though they had earlier been problematic, Dr Saunders considered that this information suggests that, overall lice are a low-level and manageable risk to farmed and wild salmon in Bantry Bay. 1057

615. Dr Saunders acknowledged the reassurance of the RPS water quality modelling and hydrological surveys that there is sufficient hydrographical distance between rivers, existing farm sites, and the proposed Shot Head site to prevent lice cross-infection. But he cautioned that modelling is predictive and based on a necessarily limited set of empirical variables (albeit at the “worst case” end of the spectrum) which may or may not fully represent the full range of hydrological conditions present within the study area.1058 He took the view that

• the proximity of the proposed Salmon Farm to the Trafrask estuary might arguably constitute an enhanced sea lice risk to the viability of the river’s salmonids.

• the “key issue” for ALAB is whether MOWI’s Pest Management Plan is sufficient to mitigate any such impact.1059

616. Dr Saunders considered “Cumulative impacts and production strategy” in some detail – including the respective benefits and detriments of synchronous and asynchronous production.1060 In his “Technical Advisor’s evaluation”, Dr Saunders noted that MOWI’s reasoning is “To achieve annual harvesting by coordinated asynchronous stocking between Shot Head and Roancarrig sites” and that MOWI had “chosen to pursue the asynchronous methodology”.1061

617. Dr Saunders’ Final Report states that:

• the sufficiency of hydrographical distance between rivers, existing farm sites, and the proposed Shot Head site to prevent lice cross-infection – “effects on lice burden are mitigated by an adequate hydrological distance between sites.” – and

• historic low lice levels in Bantry Bay – lice numbers have not triggered treatment in Bantry Bay since 2008 - together suggest that, while complacency is to be avoided and the incidence of future lice issues or disease outbreaks is difficult to predict, lice are “a manageable risk to both farmed and wild salmon”, and “a single bay production strategy is currently not a necessity in Bantry Bay. At this time, however, we do not believe that the establishment of an additional farm site at Shot Head requires the imposition of synchronised wholebay production as a license condition.”1062

1057 pp 66 & 104 – Underlining in originals. 1058 P 66. 1059 p 102.

1060 i.e. in terms of synchronicity of production cycles with other salmon farms in the bay and their compatibility with sea lice management techniques of

Single Bay Management and CLAMS (Coordinated Local Management Schemes which group salmon farms into those within overlapping tidal excursions.

On this basis, Bantry Bay is a ‘Single Bay Area’.) 1061 Pp 65 & 66. 1062 For the foregoing, see Technical Advisor’s Final report pp 66, 95 & 96.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

618. Dr Saunders’ Final Report also states that:

“This production cycle will be alternated asynchronously, offset by one year with the Roncarraig1063 site, …”

“We conclude that the production strategy proposed by MHI1064 complies with SBM and is suitable for Bantry Bay.”

“Consequently, we do not consider it necessary to impose restrictions on production strategy for the Shot Head site.” 1065

This last comment can only relate to MOWI’s intended asynchronous production as between its own Shot Head and Roancarraig/Aghabeg sites. It appears both in that part of Dr Saunders’ report consisting of his evaluation1066 and that part containing his recommendation – “Case for Licence Approval”.1067

619. Dr Saunders’ “Draft Determination Refusal /or Grant”, concludes that:

“The possibility of farm transfer of sea lice to migrating salmonids in the Dromagowlane/ Trafrask river system and the subsequent impacts to freshwater pearl mussel, however, remains a matter of some conjecture, since there is presently insufficient scientific data to make incontrovertible conclusions on current fish or mussel population status, or indeed the relationship between current salmon populations and freshwater pearl mussel recruitment. The history of only a low-level lice incidence on the present farmed fish population together with the strict imposition of a comprehensive Integrated Pest Management Plan taken together would, however, suggest that the risk can be adequately managed, particularly if a provision for additional supporting data can be incorporated into the Plan. On taking into account all of the above1068 we can offer no substantive technical reasons to refuse the current licence application.” 1069

However, the issue of the relationship between migrating salmonids and FwPM recruitment in the Dromagowlane/ Trafrask river system was later addressed by Dr O’Toole to a conclusion that the FwPM hosts in that river system were not migrating salmonids but non-migratory brown trout which are not affected by sea lice. So what remains of Dr Saunders’ opinion is the conclusion that “the risk can be

1063 Sic. 1064 i.e. MOWI.

1065 Dr Saunders’ Final Report 8 December 2021 §9, evaluation, §9.11.4 Production and farm management strategies. pp94, 95 & 106

1066 Dr Saunders’ Final Report 8 December 2021 §9, evaluation, §9.11.4 Production and farm management strategies. pp94, 95.

1067 §10.2 p106. 1068 Which includes issues other than sea lice. 1069 §11 at pp107 & 108.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

adequately managed” and “we can offer no substantive technical reasons to refuse the current licence application.”1070

620. Dr Saunders’ observation that “Future conditions in Bantry Bay may therefore benefit from the establishment of non-statutory CLAMS and the adoption of synchronised bay production and lice treatments” is clearly a general and prudent observation. But, reading his report as a whole, it is equally clearly not intended to suggest that CLAMS or synchronicity are required in any licence ALAB may grant.

621. Yet, for all the foregoing content, Dr Saunders also states as part of his “Technical Advisor’s Evaluation”,1071 that the lice risk strengthens the case for CLAMS and SBM including a synchronous entirebay production regime1072 that “asynchronous bay production is not in line with best practice advocated in CLAMS”1073 and that: “At a minimum, all sites managed by the applicant should be operated synchronously.”1074 This view is informed by Dr Saunders’ expression of concern at what he considers a lack of recognition of the expected significant increase in fish stock in Bantry Bay due to the Shot Head Salmon Farm (a 43.75% increase in production to 4,000 tonnes), which, he says, can only be expected to increase average lice numbers in the Bay.

622. Reading Dr Saunders’ report as a whole, as it must be read, “the current licence application” is clearly for production co-ordinated but asynchronous with production on MOWI’s Roancarraig/Aghabeg sites and it concludes that granting a licence does not require “the imposition of synchronised whole-bay production as a license condition.” There can be no doubt that Dr Saunders’ reasoning and conclusion that “At a minimum, all sites managed by the applicant should be operated synchronously” represents a regrettable and unexplained internal conflict in his report. Nonetheless, in my view, reading the report as a whole, in light of the MOWI proposal and the challenges to it, the general thrust of the report, the explicit advice at §9.11 is to the effect that any licence should not impose synchronous stocking/production either in all of Bantry Bay or as between MOWI’s Shot Head and Roancarraig/Aghabeg sites. In light of the clear premise of MOWI’s application, of Shot Head production asynchronous with MOWI’s Roancarraig/Aghabeg site, it is clear that advice at §9.11 informs his conclusion that he “can offer no substantive technical reasons to refuse the current licence application.”1075 In short, in my view Dr Saunders approves the prospect of Shot Head production asynchronous with MOWI’s Roancarraig/Aghabeg sites and that approval encompasses all risks posed by sea lice.

1070 p108. 1071 §9.1 at p59 et seq.

1072 p66 Dr Saunders notes that for other criteria – discharges such as nutrients – synchronised production represents a worst case scenario which MOWI

has modelled. 1073 p65. 1074 Dr Saunders’ Final Report 8 December 2021 p66 §9.1. 1075 Dr Saunders’ Final Report 8 December 2021 p108 §11.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

LICE – PROTOCOLS & SBM – LICENCE CONDITIONS VALIDITY

Lice – Duration of Fallowing – Interpretation of the Aquaculture Licence – Decision

623. MOWI has confected a movable feast as to fallowing.1076 The 2011 EIS said the Site would be fallowed for “at least” 2 months1077 or 2 to 4 months,1078 before restocking with smolts. MOWI’s 2016 Integrated Pest Management Plan1079 attempted to reduce this to at least 1 month. Even after the Impugned Determination, Ms McManus of MOWI intimated fallowing of “up to” 2 months.1080 Each of these represents a very different account by MOWI of MOWI’s intentions as to duration of fallowing. Indeed, “up to” could, admittedly shorn of context, mean no fallowing at all. All this prompts the question, what exactly does MOWI understand its fallowing obligations to be? More importantly, what duration of fallowing does the licence impose?

624. Unfortunately, ALAB confused the issue further. By §2.2 and Schedule 4 of the Aquaculture Licence, it imposed 2 months fallowing – providing of the 24-month production cycle: “the final 2 months being a fallowing period prior to re-stocking”. In notable contrast, §6.5 of the Aquaculture Licence required MOWI to “ensure the fallowing of each licensed area for at least 30 continuous days” and applies the DAFM Fallowing Protocol.1081 §3 of that protocol confuses the matter further by requiring annual fallowing “for a minimum of

30 continuous days”. That implies fallowing for at least continuous 30 days in the middle of a 2-year production cycle – there is no suggestion in the papers that such a regime is envisaged. And, as has been seen, that protocol requires that fallowing “shall be pursued”1082 in the context of the SBM/CLAMS process – i.e. voluntarily.

625. Overall, the lack of clarity is unimpressive as to exactly how much fallowing the Licence requires of MOWI and when it is to occur. As between Schedule 4 and §6.5 of the Licence, it seems to me that Schedule 4, as a bespoke term specific to the proposed development (even more obvious from the Minister’s draft licence on which it is clearly based) outweighs §6.5, which is a standard term. And, more importantly, interpreting the licence in the context of ALAB’s determination, which is the considered and effective legal act authorising the issuing of the license, §6.5.2 of the determination specifically inserts the relevant clause of Schedule 4 stipulating a 24-month production period and a 2-month fallowing period.

1076 Fallowing to break lice cycles is an element of sea lice risk reduction. 1077 EIS Vol 1 §3.1, p143. 1078 EIS Vol 1 §3.2, p148. 1079 See below. 1080 Affidavit of Catherine McManus sworn 6 July 2022. 1081 Protocol for Fallowing at Offshore Finfish Farms 11 May 2000. 1082 Emphases added.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

626. So, despite the regrettable confusion, I hold, as a matter of interpretation of the Aquaculture Licence, that a minimum 2 months’ fallowing is required at the end of each 22-month production cycle.

Lice – Aquaculture Licence Conditions, SWI CG8 & Content of Sea Lice Protocol #3

627. SWI CG8 pleads that the Impugned Decision is invalid for “impossibility” of compliance as ALAB both

• authorised asynchronous stocking (and hence asynchronous fallowing) of the Shot Salmon Farm and other fish farms in Bantry Bay, and

• required compliance with Sea Lice Protocol #3 which requires SBM and synchronous fallowing.

SWI says this self-contradiction renders compliance with the licence impossible. Whatever about that, it is certainly illustrative both of the perils of incorporating documents in conditions by reference without considering their content in detail and of the possibility of self-contradiction in a licence. In essence the plea is that the licence requires both asynchronous stocking and synchronous fallowing – which is impossible. It does not seem to me that asynchronous stocking and synchronous fallowing are possible, so the question is whether SWI are correct is asserting that the licence requires both.

628. As I have said, it is apparent overall, including but not limited to the finding of hydrological isolation, that ALAB did not intend to impose synchronous stocking and production at Shot Head. Nonetheless Condition 7.3 of the Impugned Aquaculture Licence (“Condition 7.3 (Sea Lice)”) requires that MOWI “shall comply with the most up to date detailed specifications of Monitoring Protocol No. 3 for Offshore Finfish Farms - Sea Lice Monitoring and Control, as may be revised from time to time, for sea lice monitoring and control in the licensed area”.

629. It is convenient here to set out the primarily relevant content of Sea Lice Protocol #3:

“5. Synchronous Sea Lice Treatment and Control in Bays All fish farms operating in a particular bay will be required to undertake appropriate synchronous sea lice treatment and control strategies through the Single Bay Management/CLAMS process. The Department of Marine and Natural Resources or its agent reserves the right to devise appropriate strategies for synchronous action by fish farms in any bay.”1083

630. As observed above as to a similar reservation in the Fallowing Protocol 2000, given the voluntary nature of SBM/CLAMS and insofar as §5 of Sea Lice Protocol #3 reserves a Departmental “right” to devise strategies for synchronous action by fish farms this passage must be understood as:

1083 Emphases added.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• reserving a Departmental “right” to mandatorily impose strategies for synchronous action outside of SBM/CLAMS.

• deriving its legal force from condition §7.3 of the Impugned Aquaculture Licence which requires compliance with Sea Lice Protocol #3 2000.

631. So, by Condition 7.3, MOWI must “undertake appropriate synchronous sea lice treatment and

control” with all fish farms operating in Bantry Bay. The word “appropriate” is a generally significant qualifier

of any proposition in which it appears. It prompts some questions. What is “appropriate”? By what criteria is appropriateness determined? What do those criteria exclude? As to what is “appropriate” it seems to me that §5 of the Sea Lice Protocol #3 of May 2000 must be considered in light of the best practice principles set out in the DAFM Lice Strategy 2008 as cited above. As I have observed, those principles require synchronicity only between adjacent fish farms – those within a single tidal excursion of each other. I reach this conclusion with hesitation and some regret, as I am far from happy that this meaning of §5 of the Sea Lice Protocol #3 of May 2000 was transparent to the public. Indeed, the 2008 Strategy much post-dates that protocol. However, it does seem to me that on reading the two documents together, it is clear that SBM does not

require that all farms in a bay practice synchronicity – it requires only that “adjacent” farms do so – those

farms within the same tidal excursion.

632. The question whether, for this purpose, the Shot Head Salmon Farm is hydrologically isolated is clearly one on which there was expert evidence of RPS and expert advice from Dr Saunders before ALAB as an expert tribunal and ALAB was entitled to so decide. From the conclusion that the Salmon Farm is hydrologically isolated flows the answer, in the negative, to the question whether SBM requires synchronicity of the Shot Head salmon farm with any other farm. It is clear that this is the conclusion ALAB reached in granting a licence the premise of the application for which was a production cycle of single bay site alternation, and not synchronicity, with the Roancarrig/Aghabeg salmon farm. This all seems to me a matter eminently for expert decision by ALAB and not a matter for decision by the Court.

Validity of Condition 7.3 & the SBMP Condition – Enforceability – Initial Observations

633. Condition 7.3 also, and importantly, requires that any strategies be effected “through the Single Bay Management/CLAMS process”. And they are voluntary processes. Their voluntary nature is also reflected in a “Schedule 5” condition of the Impugned Aquaculture Licence1084 as to SBM (the “SBMP Condition”) that MOWI shall,

“comply with any bay wide single bay management plan or code of practise for Bantry Bay developed in agreement with any relevant State body.”

1084 Schedule 5, Additional Conditions.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

634. It will be immediately seen that, the SBMP Condition requires compliance with an SBMP only if there is one. Which is to say, if MOWI and all other aquaculture licensees in the Bay have agreed one. In context, it is clear that an envisaged counterparty to any such agreement with the State is MOWI itself. To put it simply the SBMP Condition requires MOWI to comply only with any future agreement with which it agrees to comply. By simply not agreeing to any proposed SBMP, MOWI has it in its power to nullify any effect of the SBMP Condition.

635. The DAFM Sea Lice Protocol #3 and the DAFM Fallowing Protocol contemplate their being carried into effect via SBMPs. But, as stated, SBMPs require agreement of all producers in a bay – including MOWI. And even if the participation of producers only in an hydrologically isolated part of the Bay sufficed to create an effective SBMP (as to which I express no view) to bind MOWI, MOWI would have to agree to it. So compliance with those protocols seems dependant on such agreement – including by MOWI. That this process both is voluntary on the part of the licensee and depends on agreement with private third parties (who may be competitors) results in incongruity in the protocols’ being carried into effect by licences containing mandatory conditions. Ultimately, given the voluntary nature of SBM/CLAMS, I have difficulty seeing that, for example, Condition 7.3, Condition 6.51085 or the SBMP Condition impose enforceable legal obligations. One might argue that they take effect in the sense that if MOWI were to subscribe to an SBMP its agreement would be enforceable as a licence condition. But that is not very reassuring. I return to this issue below.

636. I should add that there is no suggestion that these conditions are akin to “Boland conditions”1086 leaving matters to further agreement. Nor could such an argument have been made: whether and in what terms SBM will be effected is not a matter of detail – as is required of a Boland condition.

Lice – Is there an SBMP for Bantry Bay?

637. As stated above, the 2011 EIS described SBM as an “aspiration”.1087 It says, somewhat ambiguously1088 that SBM “which was integrated in due course into .. CLAMS .. in a number of Irish bays and loughs (excluding Bantry) was first introduced by the .. Department .. in the early 1990's.” It also refers to SBMPs, as incorporated into CLAMS “where these have been introduced” and states that “CLAMS has yet to

1085 §6.5 of the Aquaculture Licence requires MOWI to “ensure the fallowing of each licensed area for at least 30 continuous days” and applies the DAFM

Fallowing Protocol.

1086 For recent accounts of Boland Conditions see Krikke v Barranafaddock Sustainable Electricity [2021] IECA 217, [2022] IESC 41, [2023] 1 I.L.R.M. 81 and

Shadowmill v ABP & Lilacstone[2023] IEHC 157. 1087 EIS 2011, Vol 1 §3.2.2. 1088 p21 footnote 3.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

be established in Bantry Bay”. 1089 All that is ambiguous in that it is unclear whether the words “excluding Bantry” apply to the existence of SBM or merely to its integration into CLAMS. The Minister’s EIA records that since 1997, SBMPs “have been in place in all salmon farming areas in Ireland, including Bantry Bay.” This appears to contradict the EIS and factual basis for this assertion is not apparent.

638. On my inquiry at trial1090 I was told by Counsel in submissions that there is in fact an SBMP for Bantry Bay and it is agreed annually with the Marine Institute. I do not doubt counsel – who was answering my question. But no SBMP for Bantry Bay has been exhibited or deposed to or cited in the application papers. Nor is there evidence that any SBMP was before or considered by ALAB. This was explained on the basis that as Shot Head Salmon Farm does not yet exist there is no SBMP which relates to it. I find this surprising as, if there is such an annually agreed SBMP for Bantry Bay,

• MOWI is an existing operator in Bantry Bay and is presumably party to the SBMP.

• it would surely shed light on what is thought by both salmon farmers and the Minister to be the proper SBM regime for Bantry Bay.

• even though needing revision to accommodate the Shot Head Salmon Farm, presumably it would form the starting point for an SBMP incorporating the Shot Head Salmon Farm or at very least inform the terms of such an agreement.

• one would have expected it to be submitted to ALAB.

639. To add to the confusion there is exhibited a document produced for MOWI and submitted to ALAB, entitled “Shot Head Integrated Pest Management Plan/Single Bay Management Plan”, which I describe further below. It was produced by MOWI in October 2016 and submitted to ALAB. However, as the description above of my interaction with counsel makes apparent, it is not the Bantry Bay SBMP or an agreed SBMP at all. Rather, it stated MOWI’s unilateral intentions as to lice risk management – as between its sites only.

640. Ultimately, all I can say is that at present there is no evidence of any existing SBMP which could bind MOWI other than by MOWI’s voluntarily subscribing to it.

Lice – RPS Report 2015 – Dispersion modelling

641. The RPS Report 2015 set out1091 to inform an assessment of sea lice risk posed by the Shot Head farm by modelling, in the form of diagrammatic plumes, dispersal into the bay of sea lice larvae from the farm. It described its model as specific to and valid only for Bantry Bay. Of some note, it considered and

1089 p220. 1090 Transcript Day 13 p24 et seq 1091 RPS Report 2015 §5.7 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

differentiated similar modelling in Norwegian fjords1092 and considered a 2009 paper by Amundrud and Murray1093 modelling lice dispersion in a Scottish sea loch. It considered both currentand tidal-induced and wind-induced dispersion. 1094 It concluded that “under all scenarios tested, the concentrations of infestive copepodid stages throughout the Bay were found sufficiently small that augmentation of natural sources of infestation are considered unlikely to occur as a result of the presence of MHI’s proposed salmon farming operations.”1095

642. Controversially, RPS assumed neutral buoyancy of lice larvae and it seems that assumption was important to the conclusion that lice do not congregate (as opposed to disperse) under onshore wind conditions.1096 This neutral buoyancy assumption was challenged as to its effect on the EIA. I will return to it in due course.

643. The RPS base simulations assumed various elements of a worst-case scenario, including the trigger levels at which lice treatment would be required. It notes1097

• government-set1098 lice treatment trigger levels of 0.5 ovigerous1099 lice per fish on bi-weekly monitoring in spring1100 and, otherwise, 2 ovigerous lice per fish on monthly monitoring.1101

• that MOWI in practice uses a lower more precautionary lice treatment trigger level of 0.3 ovigerous lice per fish. The RPS base dispersal simulations assumed 1 ovigerous louse per fish, which would allow use of a simple multiplier to estimate lice larva dispersal at other infestation rates. However, its various tables and figures illustrate plume envelopes assuming 11102 and 0.31103 ovigerous lice per fish only. None assume a licence imposing only the government-set treatment triggers of 0.5 or 2 ovigerous lice per fish – which is what transpired. While trigger levels were discussed at trial, no issue as to discrepancies in that regard was pleaded. Accordingly I will not pursue any questions in this regard.

1092 RPS Report 2015 §5.7.1

1093 Amundrud, T. L. & Murray, A. G. 2009 Modelling sea lice dispersion under varying environmental forcing in a Scottish sea loch. J. Fish Dis. 32, 27–44.

(doi:10.1111/ j.1365-2761.2008.00980.x) 1094 p77. 1095 RPS Report 2015 §6. 1096 pp66 & 80. 1097 p66.

1098 Sea Lice Protocol #3 of 2000: “4. Trigger Levels for Treatment. The setting of appropriate treatment triggers is an integral part of implementing a

targeted treatment regime. Treatment triggers during the spring period are set close to zero in the range of from 0.3 to 0.5 egg bearing females per fish

and are also informed by the numbers of mobile lice on the fish. Where numbers of mobile lice are high, treatments are triggered even in the absence of

egg bearing females. Outside of the critical spring period, a level of 2.0 egg bearing lice acts as a trigger for treatments.”

The DAFM Lice Strategy 2008 records a slightly different spring trigger level at 0.5 egg-bearing lice. 1099 Egg-bearing. 1100 March to May inclusive, when wild salmonids migrate. 1101 It appears that these are means. 1102 Table 5.8 and Figure 5.41 et seq. 1103 Table 5.9 Figure 5.49 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

644. RPS opined that under all scenarios tested, concentrations of infestive larvae throughout the Bay were sufficiently small that augmentation by the Shot Head Salmon Farm of natural sources of infestation is unlikely.

Lice – MOWI’s IPM/SBMP – October 2016 & s.47 Reply 2 November 2016

645. ALAB’s s.47 Notice of 6 October 2016 to MOWI,

• noted that the EIS and RPS report indicate that use of EmBz – the most effective lice treatment – will be severely restricted to avoid breach of the statutory EQS.1104

• therefore seeks details of MOWI’s alternative treatment strategy in the event of a lice infestation event beyond the 7-month post smolt transfer period, “when Emamectin Benzoate may not be used”.1105

646. The DAFM Lice Strategy 2008 describes SBM as involving fish farms co-operating to develop an Integrated Pest Management Plan.1106 MOWI’s s.47 reply enclosed an “Integrated Pest Management/ Single Bay Management Plan” (“MOWI IPM/SBMP 2016”) dated 26 October 2016. As I have said, it is not an agreed SBMP proper – it is, rather a unilateral proposal by MOWI. Also, it must be read in the context that, as MOWI itself observed at trial1107 “The licence doesn't condition the Integrated Pest Management Plan” such that

“There may not be an obligation in the licence.” to carry it into effect. MOWI’s position at trial was that while

it was in MOWI’s interest – for commercial and perhaps other reasons – to effect the MOWI IPM/SBMP 2016, it was under no legal obligation to do so.

647. Its content includes the following:

• It describes SBM as in place since 1997 “in all salmon farming areas”, as “fundamental to the rational management of the salmon farming industry” as and as having proved “very effective in enhancing the efficacy of lice control”. o While this is almost exactly the same text as is cited in the Minister’s EIA in this regard and suggests that an SBMP exists as to Bantry Bay, it notably excludes the EIA’s words, after “ … all salmon farming areas” the words “including Bantry Bay”. Thus the MOWI IPM/SBMP 2016 implies, but did not expressly state, that SBM was in place in Bantry Bay. I am not aware nor, as far as I can see, was ALAB

1104 Environmental Quality Standard as to pollutant emissions to a water body. Set pursuant to the Water Framework Directive Regime.

1105

For the words in parentheses see ALAB Opposition §118. The RPS Report (§5.6.1) modelled EmBZ levels in the water body against Environmental

Quality Standard set for EmBZ by the European Communities (Control of Dangerous Substances in Aquaculture) Regulations 2008 (S.I. No. 466 of 2008)

and calculated that “the maximum standing stock that could be treated at the Shot Head site is 440 tonnes, which the site reaches at month 7 post smolt

transfer, in May of the first growth year. It should be noted that this would enable strategic treatment with Slice prior to the first sea winter to protect

the fish into the "Susceptible Period" under the mandatory lice inspection and treatment protocols, when wild salmon and sea trout smolt are migrating.

Alternatively, treatment of 440 tonners of stock at month 7 would protect the farm stock for 120 days until about September, should this be required.”

1106 For present purposes, “pest” means sea lice. 1107 Transcript Day 15 p15.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

aware when SBM was put in place there and in what terms or who the parties were or are and, as I have said, it was not exhibited.

• It set outs how MOWI intends to control sea lice in the Shot Head site “which incorporates existing salmon farm sites in Bantry Bay”. While this is grammatically unclear, the accompanying map illustrates MOWI’s Shot Head grower site, Roancarrig/Ahabeg grower sites, and another MOWI-proposed harvest site further west at Waterfall in Berehaven. It does not illustrate any sites not in MOWI control. o I observe that SMBPs clearly must relate to all salmon farms in a bay. MOWI’s limitation of its MOWI IPM/SBMP 2016 might be explicable if MOWI controlled all salmon farms in the bay in 2016 but it does not say so or exclude the existence of salmon farms in the bay other than those it controls. The 2011 EIS identified “two producers in Bantry Bay”.1108 MOWI is one and it showed two Fastnet Irish Seafood1109 licensed salmon farms on the other side of Bantry Bay at Gearhies, across the Bay and south east of the proposed Shot Head farm.1110 And the previous year RPS had included those sites in its modelling.1111 o MOWI indicated at trial that it expected to purchase those farms at Gearhies but it does not seem to me that this consideration can affect the validity of the impugned Aquaculture Licence. o Accordingly, the MOWI SBMP 2016 does not make it apparent that it proposes an SBMP as that term is generally understood as an agreement between all producers in a Bay.

• If required, EmBz will be applied at latest in April of the first year at a Maximum Treatable Site Biomass of

396 tonnes. Other medicinal and non-medicinal (cleaner fish1112) lice treatments will also be used.

• Where two sites are stocked in the Bay, treatments will be carried out on both during the same time period and with the same chemical class. o I confess that I am unclear how asynchronous production as between the Roancarraig/Aghabeg and Shot Head sites and the limitation of EmBz use to at latest April of the first year at a Maximum Biomass of 396 tonnes can allow for synchronous EmBz treatment – but perhaps that is explicable by the absence of a similar limitation in the Roancarraig/Aghabeg licences. In any event and as recorded below, Dr Saunders was likewise dubious.

• The fallowing plan for MOWI sites is set out in a diagram and is based on “fallowing for a minimum of one month per cycle” (As I have held above, the Licence requires 2 months fallowing at the end of each production cycle). The diagram is difficult to follow, not least as only one Shot Head growth cycle is shown.

However, it does seem clearly enough to assume alternation of production as between MOWI’s Roancarraig/Aghabeg and Shot Head sites, as opposed to their synchronous production.

1108 §3.2.2, p149. 1109 a.k.a. Murphy’s Irish Seafood. 1110 Figures 7 & 79. 1111 See for example §2 and Figure 2.1 and many figures marked “Shot head/Fastnet dominant”.

1112 Cleaner fish display a natural behaviour in the wild of removing parasites and dead skin from other fish species.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Lice Treatment Trigger Levels – Voluntary?

648. The position as to lice treatment trigger levels requires some elucidation. MOWI deposes and stresses that it has, and I accept that in practice it has, applied in its Roancarraig/Aghabeg farm, the stricter trigger levels of 1.0 ovigerous lice per fish generally and 0.3 ovigerous lice per fish in the spring for which RPS produced its tables and figures in 2015. These are stricter as compared to the Sea Lice Protocol #3 triggers of

2.0 ovigerous lice per fish generally and 0.5 ovigerous lice per fish in the spring.1113 However, by its IPM/SBMP 20161114 MOWI proffers, not its stricter practice, but the less strict Sea Lice Protocol #3 triggers. Further, at trial MOWI agreed that its stricter practice was “not required by the licence” and that they are free to apply only the Sea Lice Protocol #3 triggers if they prefer.1115

649. The applicable regulatory process is characterised by a licensing requirement imposing binding obligations and the imposition of binding conditions in such licences. In law, it is not characterised by voluntarism – though, doubtless, voluntary practice can be significant outside the regulatory process and can supplement it towards similar ends. In no way do I seek to discourage or discount voluntary practice in any general sense. However, on an issue of such significance to salmon farming regulation as lice treatment triggers (the importance of which is established by the Sea Lice Protocol #3 in particular and more generally by the great weight of evidence, protocols and the like1116 before ALAB as to the importance of lice control) it does not appear to me that appreciable weight in the formal environmental assessment in the mandatory regulatory process can properly be attributed to assurances of voluntary stricter practice. That is generally so but must be especially so when the licensing applicant, MOWI, positively proffers in the process – in this case by its IPM/SBMP 2016 – a less strict commitment than to its historic voluntary practice. That is not to suggest that the proffered Sea Lice Protocol #3 lice treatment triggers are inadequate. It is to say that MOWI cannot expect, in a mandatory regulatory process, that the formal environmental assessment of the project will be appreciably predicated on continued future application of a stricter but voluntary practice.

650. That is not to suggest however that the formal environmental assessment of the project need disregard the fact, if it be a fact, that MOWI’s historic voluntary practice of applying lower treatment triggers at its Roancarraig/Aghabeg farm has resulted in a low number of treatment instances.1117 The point here is not that the formal environmental assessment of the Shot Head project will be appreciably predicated on continued future application of a stricter but voluntary practice. It is that a low number of treatment instances resulting from the historical application of lower treatment triggers at the Roancarraig/Aghabeg farm may shed light on the underlying scale or absence to date of any serious lice problems in Bantry Bay.

1113 Sea Lice Protocol #3 §4. It contemplates a trigger “in the range of 0.3 to 0.5” ovigerous lice per fish in the spring – but clearly, to the extent this

imposes any obligation a trigger of 0.5 meets it. 1114 p4. 1115 Transcript Day 15 p15.

1116 For example, the Fallowing Protocol 2000, the Monitoring Protocol #4 2000, and the DAFM Pest Control Strategy 2008.

1117 As to which see below as to MOWI’s sEIS of April 2018.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

ALAB Conclusion on the MOWI IPM/SBMP 2016 and is MOWI Bound to Implement it?

651. As the minute of ALAB’s meeting of 22 November 2016 records, Dr Saunders advised that the MOWI IPM/SBMP 2016 proposal of treatments on two sites during the same time period with the same chemical class cannot, in his view, be applied where the Shot Head stock is in its second winter. (I infer that is because EmBZ can be applied at Shot Head only in the first winter) and so the proposal needed clarification. “Subject to that however his view was that the Plan was a robust one.” And as there is no lice problem in Bantry Bay at present and given site separation, he considered that “the issue can be dealt with”.

652. At this meeting, ALAB “formed the view” that the issue was resolved subject to the clarification required and to inclusion of a relevant condition in the licence should ALAB “decide to grant a licence

following consideration of all issues”. ALAB “agreed to consider this aspect further as the appeal progresses”.

653. As I have noted above MOWI itself observed at trial:1118 “The licence doesn't condition the Integrated Pest Management Plan”. That is correct in point of fact. Clearly MOWI, on its interpretation of the Licence, does not regard itself as bound by its own IPM/SBMP 2016 which it submitted to ALAB. Equally clearly, ALAB had regard to the IPM/SBMP 2016 and regarded it as significant to its view of 22 November 2016 that, subject to some clarification, the lice treatment issue was resolved. Nonetheless, ALAB specifically directed that Prof. McIntyre at the oral hearing was to inquire into the robustness of MOWI’s IPM/SBMP 2016 – though rather than express a view he advised an sEIS. Also, ALAB’s Determination of 29 June 2021 to grant the Impugned Licence identifies it, sub nom “Pest Management Plan”, as having been considered1119 and repeatedly expressed the view as to sea lice risk that,

“……. the DAFM Monitoring Protocol No.3 for Offshore Finfish Farms - Sea Lice Monitoring and Control, with which the Applicant is required to comply as a standard Term and Condition of the licence ….. along with the Pest Management Plan, is deemed to be sufficient and reduces any risk to a reasonable, non-significant level …”1120

654. Given ALAB’s repeated reliance, in its Determination of 29 June 2021 to grant the Impugned Licence, on the IPM/SBMP 2016 as significant in reducing sea lice risk to “a reasonable, non-significant level”, it is frankly disquieting to note that MOWI does not consider itself bound by the Licence to implement the IPM/SBMP 2016. Whether that disquiet properly proceeds from MOWI’s attitude, an incorrect interpretation by MOWI of the Licence understood in light of the Determination, or a correct interpretation by MOWI based on an omission by ALAB to condition implementation of the IPM/SBMP 2016, I need not decide. What

1118 Transcript Day 15 p15. 1119 ALAB Determination 29 June 2021 §2.8. 1120 ALAB Determination 29 June 2021 §§4.4, 6.1.2, 6.1.12, 6.5.4. 6.5.5 & 6.5.11.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

can certainly be said is that if ALAB intended that MOWI be bound to implement the IPM/SBMP 2016, clarity in the Licence in that regard was at very least desirable.

Lice – Saunders Interim report 31 December 2016 on Lice and FwPM

655. MOWI’s s.47 reply of 2 November 2016 also cites various authorities and reasons for its assertion that the Shot Head salmon farm will not pose a sea lice risk to the host salmonids of the FwPM.

656. Dr Saunders’ interim report of 31 December 2016 is detailed as to risk to the FwPM in the Trafrask river system via sea lice risk to its salmonid larva hosts. His conclusion is that, while he accepts that the RPS hydrological modelling indicates a negligible chemical contamination impact, and that MOWI’s Integrated Pest Management Plan will maintain lice levels at manageable levels, he was not fully convinced that the Salmon Farm presents no lice infection risk to the salmonids of the river and a residual concern remained of the risk to these salmonids and the implications for their interdependency with the FwPM. He cautioned that it would be prudent to assume that the Trafrask River FwPMs are at least partially dependent on returning Atlantic salmon. Also, some brown trout may migrate to sea and interbreed with the sea trout. The proximity of the Salmon Farm “to the Trafrask embayment entrance might arguably constitute an enhanced sea lice risk, in which a significant infestation event may substantially affect the viability of the river’s salmonid population. The key issue for the Board to consider is therefore whether the .. Integrated Pest Management Plan is sufficient to mitigate any future fish farm-derived impact on salmonids” in the Trafrask River.

Lice – Marine Institute/Jackson Response 9 February 2017

657. The Sweetman Applicants1121 observe that the Marine Institute paper of 9 February 2017, in response to the Saunders Interim Report, was not in response to a s.47 request. The implication seems to be that it was inadmissible in the process. The Marine Institute paper seems to have been submitted in anticipation that Dr Jackson would not be present at the oral hearing and in lieu of his testimony. However, as it transpired, he did attend and so I do not see that anything of consequence turns on the Sweetman Applicants’ observation. The Marine Institute paper, by Dr Jackson, addresses three issues as arising out of the Saunders Interim Report:

• 1. Risks to salmonids in the Trafrask River – as to which Dr Jackson states that while lice-induced mortality on outwardly migrating smolts can be significant, the risks of lice-induced mortality in post-smolt wild salmon have been quantified by Marine Institute research at in the order of 1% and so as unlikely to

1121 Carr Affidavit sworn 27 September 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

be significant to the conservation status of salmon stocks and so there is no correlation between the presence of aquaculture and the performance of adjacent wild salmon stocks.1122 o I note that this observation by Dr Jackson does not address the risk that the acknowledged “significant” lice-induced mortality of outwardly migrating smolts may imply significantly fewer postsmolt wild salmonids to return in due course to their natal Trafrask river. One might suggest that what does not get out to sea cannot come back from the sea. But that seems to me to be a matter for ALAB, not for me.

• 2. Associated impact on the FwPM – as to which Dr Jackson states: o The only mechanism through which the Salmon Farm could impact on the FwPM is by loss of host salmonids for its larvae. o There is no evidence that changes in salmonid populations have contributed to the current unfavourable status of the FwPM in Ireland. The overwhelming evidence, including from NPWS studies, is that that FwPM declines are caused by sedimentation and eutrophication of FwPM habitats. o In the Dromagowlane/Trafrask Rivers it is the non-migratory brown trout which act as FwPM larva hosts and brown trout do not interact with marine salmon farms (or, the implication clearly is, lice from those farms). o There is therefore no mechanism for impacts on the FwPM.

• 3. The robustness of MOWI IPM/SBMP 2016 – as to which Dr Jackson states: o The National Sea Lice Monitoring and Control Programme has driven a strong downward trend and steady and sustained improvement in sea lice control. o Sea lice control on existing farms in the southwest since 2008 has been and is excellent, with numbers well below the national average. o MWI’s plan is appropriate, workable and fully meets the conditions of the SBM process.

Lice – Oral Hearing 2017 & Report Thereon

658. The Notices to Appellants in December 2016 of ALAB’s intention to hold an oral hearing specifically advised that its purpose was to seek to establish clarity on:

• 1. The nature of and the risks to salmonids in the Dromagowlane/Trafrask River.1123

• 2. Associated impact on the FwPM.

• 3. The robustness of the MOWI IPM/SBMP 2016. It was abundantly clear, therefore, that the oral hearing would address and the Appellants would be heard, on those specific issues.

1122 Jackson et al., 2013. 1123 Sic.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

659. Lice issues generally loomed large at the oral hearing. Numerous reports, submissions and oral testimonies, from all sides of the debate and none were adduced. SWI, IFI (Dr Paddy Gargan), the Sweetman Applicants, the Marine Institute (Dr Jackson) and the NPWS (Dr Jervis Good) contributed. Doctors Bass, Gargan, Jackson and Good are experts in the field – albeit they do not all agree. The expert credentials of the competing experts were very impressive. It is impossible to give here a full account of what was a vigorous controversy and it is unnecessary as I am not entitled to resolve it. However a very brief impression will assist. The following its taken from the transcript and Professor McIntyre’s report on the Oral Hearing.

660. The robustness of the MOWI IPM/SBMP 2016 was extensively canvassed. Dr Jackson for DAFM, of the Marine Institute, gave evidence along lines of his report of 9 February 2017 and added that

• NPWS/Department of Environment1124 studies provide overwhelming evidence that FwPM population declines have been caused by sedimentation and eutrophication of FwPM habitats. 26 of 27 populations were found to be in an unfavourable conservation status despite the presence of juvenile salmon in all 26 catchments and juvenile trout in 25 with glochidial attachment to fish detected in 12.

• this evidence contradicts concerns that changes in salmonid populations have contributed to the current unfavourable FwPM status.

661. Dr Bass for MOWI also suggested that indigenous brown trout populations in these rivers might play a key role in hosting reproducing FwPMs. Dr Bass and RPS1125 also summarised the RPS Report of 2015.1126 As here relevant, they asserted that,

• in part due to poor transmission rates in the slow current regime, lice infestation is naturally and consistently low in Bantry Bay and medication is rarely required.

• lice treatment trigger levels have never been breached at the Roancarrig/Ahabeg sites since MOWI took them over in 2008.

• the RPS lice dispersal modelling was numerical and done on a worst-case basis.1127

• modelling showed that, to stay within the EmBZ EQS, peak permissible treatment biomass was 440 tonnes.1128 That would allow strategic EmBZ treatment in the first “susceptible spring period”, during which wild smolt migrate.

• In specifically Bantry Bay conditions, given the dilutions modelled from open water to estuaries, dispersed farm-origin copepodid sea lice larvae will always be present in numbers insufficient to augment natural

1124

Dept. of Environment, Heritage and Local Government, The Freshwater Pearl Mussel: Sub-Basin Management Plans – SEA Scoping Document (DEHLG,

2009). 1125 Per Dr Naomi Shannon.

1126 via a PowerPoint entitled “Numerical Modelling of the dispersion of wastes, medication and salmon lice … from the Proposed MHI Shot Head site in

Bantry Bay”.

1127 e.g. maximum fish numbers, larval lice released over the flood tide only, constant wind forcing from the south west into the Bay and natural predation

not factored in. 1128 The greater the biomass of salmon requiring treatment the greater the amount of EmBZ required.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

sea lice infestation pressure on wild salmonid smolts in their natural infestation zones in river mouths. The chances of farm-origin larvae encountering host salmon are far too low to augment natural infestation.1129

662. In marked contrast, Dr Gargan for IFI opined that the proposed production cycle, with peak biomass achieved in February / March of every second year and harvesting from March for 6 months, runs counter to all SBM principles. Fully-grown fish are unlikely to get sea lice treatment before harvesting, at the very time that wild salmon and sea trout smolts are heading to sea, thereby increasing the risk of cross-infestation. Dr Gargan recommended best practice SBM of all fish farming in the Bay, with all sites stocked together in March to ensure that they are not due to be harvested when wild salmon are heading to sea. Dr Gargan opined that there should be synchronised stocking, harvesting and fallowing in Bantry Bay.1130

663. SWI disputed as “entirely speculative” suggestions that the loss of migratory salmonids hosts to the FwPM might be mitigated by the benefit of brown trout carrying the larvae instead. SWI said “It isn't based on an adequate assessment of the role of salmonids themselves or on the role of brown trout in the upward dispersal of freshwater pearl mussel.”1131

664. Dr Jervis Good, of the NPWS1132 opined that:

• Smaller rivers such as the Dromagowlane/Trafrask may have a particularly significant role to play in FwPM conservation.

• The principal conservation issue for the FwPM in those rivers is phosphate contamination of waters and siltation.

• There is little evidence that salmon production has impacted the FwPM locally.

• Reduction in wild salmonids would have a detrimental impact as the FwPM’s long-term survival depends on a population of wild salmonids. He suggested that migratory trout might be the more important host, due to their better health and dynamism.

• Detrimental effects of sea lice on wild salmon populations cannot be ruled out and should be dealt with in the EIS.

665. It is clear from the foregoing that the expert scientific controversies as to the indirect risk of lice to the FwPM were fully ventilated at the oral hearing. Not merely that but Prof. McIntyre in his report concluded that the Minister’s EIA had neglected the risk of sea lice infestation of wild salmonids in the

1129 pp 21 & 22. Dr Bass asserts that in spring smolts concentrate in the brackish water of river mouths acclimatising to salt water where they are naturally

prone to infection with sea lice from wild salmon returning to spawn. He says that critical host densities of wild fish only occur in river estuaries pre-

migration dispersal and not in open waters where wild fish are far more dispersed.

1130 Transcript 14 February 2017 p98. The Oral Hearing report does not use the phrase “synchronous stocking” of Dr Gargan’s evidence, but that is what

it amounts to. 1131 Oral Hearing Transcript Day 21 pp49 & 50. 1132 National Park & Wildlife Service.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Dromagowlane/Trafrask Rivers and the indirect consequences of any resulting decline in wild salmonid populations thereof for inter alia, the FwPM. He suggested that

• ALAB require an sEIS addressing, inter alia, “The risk of sea-lice infestation of wild salmonids migrating from/to the Dromagowlane and Trafrask Rivers, and any resulting implications for local freshwater pearl mussel populations”.

• were a licence granted, the sea lice issue and the protection of salmonid smolts might be addressed by conditions changing the timing of the production cycle or controlling stocking densities1133 at the time of wild salmonid migration.

Lice – sEIS 2018 & Responses thereto

666. On Prof McIntyre’s recommendation and at ALAB’s request, MOWI’s sEIS of April 20181134 investigated, inter alia, the fate of farm-origin sea lice copepodid larvae via the RPS hydrographic and dispersion modelling study of 2015. The sEIS asserts that treatment triggers in the spring are “0.3 to 0.5” egg-bearing lice per fish.1135 Also “high numbers” of mobile lice per fish trigger treatment even absent egg-

bearing lice (“high” is not quantified). Outside of the spring, 2.0 egg-bearing lice per fish triggers treatment.

The sEIS asserts1136 that egg-bearing lice numbers in Bantry Bay salmon farms never exceeded a mean average of 1 per fish at any time in the 9 years of Marine Institute inspections from 2008 to 2016 and never approached 0.5 per fish in the critical spring period. It asserts that MOWI’s own more frequent inspections at its Roancarraig / Ahabeg site, using treatment triggers of 0.3 per fish in the spring and 1 per fish at other times, required treatment only 6 times from 2008 to 2016 – and only 4 EmBz treatments. The Bass affidavit of 7 July 2022 says the 0.3 trigger was reached only 6 times between 2008 and 2017. I observe that this implies that all treatments were only in the spring periods in which outgoing smolts are vulnerable. In any event and based on this history, the RPS 2015 modelling was based primarily on a treatment trigger of 1 eggbearing louse per fish and a smaller number of simulations were based on a trigger of 0.3.

667. For reasons set out at length, the sEIS asserts that there is no indication that farm-origin lice have impacted on the 5 National Salmon Rivers around Bantry Bay in the 40-year history of salmon farming in the bay – 4 of which rivers have positive conservation status for salmon and are open for angling.1137 (The Dromagowlane/Trafrask is not a National Salmon River).

1133 For example, by early harvesting. 1134 Supplemental EIS. 1135 I have considered above the effect of the Licence on this issue of lice treatment trigger levels. 1136 p25-26 & Figure 2.8. 1137 E.g. Supplemental EIS p56.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

668. In essence, the sEIS repeats the Bass/RPS presentation1138 to the oral hearing and explains it in more depth. The sEIS says that worst case modelling demonstrates that it is not possible for farm-origin infestive sea lice copepodid larvae to ever reach:

• sufficient concentration in coastal waters to threaten infestation of wild salmonids.

• any estuary, including the closest – that of the Dromagowlane / Trafrask river. The sEIS says that even immediately around the fish farm, the chance of farm origin sea lice copepodid larvae infecting wild salmonids will be below 0.08%.

669. The sEIS also considers the issue of risk to the FwPM in the Dromagowlane/Trafrask river. Inter alia, it cites at some length the 2008 Ross FwPM Study1139 of, inter alia, the Dromagowlane/Trafrask river system. It records that, of the 14 Cork and Kerry rivers assessed, Ross considered the Dromagowlane/Trafrask to be one of the four most significant FwPM populations identified and that the population may be of national significance. Ross recommended Stage 2 and 3 surveys but this has not happened despite national concern for FwPM status. The sEIS describes Ross as making clear that the Dromagowlane/Trafrask FwPM populations may be at severe risk of failing due to lack of recruitment and likely to continue to age without recruitment to the point of extinction, if the steps recommended for FwPM populations within SACs are not applied. Indeed, the sEIS suggests that the river be designated an SAC as a matter of urgency in order that local FwPM can be protected. It certainly cannot be said that the sEIS ignored Ross or downplayed the general vulnerability of the FwPM population in the Dromagowlane/Trafrask river system.

670. The sEIS explains the role of juvenile salmonids as FwPM larvae hosts. It cites academic authority that brown trout are said to be the main host species in Ireland and states that rivers carry varied population ratios of brown trout, sea trout and salmon. It notes that their relative importance for FwPM is not fully clear and it may be that differences in their reproductive behaviour affect mussel recruitment. Therefore, measures to protect FwPM must also include the monitoring and assessment of host fish status. It states that “Just as FwPM are neglected in the Trafrask, because it is not a National Salmon River, there has been almost no assessment of the salmonid populations in the Trafrask to date ..” though all have been found present and it maps and tabulates the IFI electrofishing survey of May 2017.1140 It observes that juvenile brown and sea trout are indistinguishable. It comments:

“The most material outcome of the survey is that the density of fish found seems to be low, for both salmon and trout, in addition to which salmon were only found in the lower reaches of the river.

Although this is not unusual in small rivers in mountainous areas, it may indicate that brown trout are better dispersed in the system to take on the role of Glochidial vectors. Fish densities of both species appear somewhat lower than the stated requirement for glochidial hosting of 0.2-0.3 fish per m2 of river. This suggests more than anything that more electrofishing is required …”

1138 PowerPoint – “Numerical Modelling of the dispersion of wastes, medication and salmon lice … from the Proposed MHI Shot Head site in Bantry Bay”.

1139 Rapid Assessment FwPM Survey, Dr Eugene Ross, 2008 – report to NPWS. 1140 Electrofishing stuns fish temporarily so they can be identified and counted.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

671. The sEIS cites an SEA of 2009, in substance cited by Dr Jackson at the Oral Hearing and the subject of a paper by him,1141 as part of “overwhelming evidence … that sedimentation and eutrophication of juvenile and adult FwPM habitats is the primary cause of FwPM declines” rather than lack of host salmonids. That paper concluded that the complaint of impact of salmon farms on the FwPM “had no basis”.

672. The sEIS concludes that, as the RPS model shows that no farm-origin augmentation of wild salmon lice infestation levels is anticipated, there is also a zero risk that anadromous salmonids will be reduced in numbers in their freshwater phase, to impact on the availability of hosts for FwPM Glochidia larvae. The risks for the Trafrask FwPM lie within their freshwater environment.

Publication of sEIS & Responses by Objectors, NPWS, Marine Institute, IFI

673. Importantly though unsurprisingly, ALAB published the sEIS and invited responses. IFI, SWI and the Sweetman Applicants all responded. All disputed the sEIS conclusions but none addressed the question whether the brown trout is the primary FwPM larva hosts in the Dromagowlane/Trafrask river, nor did they refer to Dr Good’s evidence at the oral hearing as to the role of different salmonids as FwPM larva hosts.

674. NPWS advised as to the sEIS1142 that,

• risk to the FwPM in the Dromagowlane/Trafrask river should be considered in the EIA.

• it had no opinion if the sEIS was correct on this issue.

• if ALAB accepts that there is zero risk that anadromous salmonids will be reduced in numbers in their freshwater phase, as a result of the presence of the Shot Head site, “then adverse indirect effects on freshwater pearl mussel in the Trafrask river system are not likely via declines in their hosts.”

It seems reasonable to consider that, at least generally, the evidence of Dr Good of the NPWS as to the FwPM is to be taken with this s.47 reply of the NPWS.

675. The Sweetman Applicants1143 assert that NPWS here assumes that migratory salmon may act as FwPM larva hosts. That is correct – but only in the limited sense that the assumption is made for the sake of the argument whether, if there is zero risk that anadromous salmonids will be reduced in numbers in their freshwater phase, any risk to the FwPM would subsist. On that assumption, the NPWS answer is no. The NPWS assumption is not evidence that migratory salmon will act as FwPM larva hosts nor does it tend to

1141 Report on Sea Lice Epidemiology and Management in Ireland with Particular Reference to Potential Interactions with Wild Salmon ….. and Freshwater

Pearl Mussel ….., Irish Fisheries Bulletin #43 2013, Marine Institute, Aquaculture Section. 1142 s.47 Reply 16 November 2018. 1143 Affidavit of Noel Carr 27 September 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

undermine evidence that brown trout are the primary FwPM larva hosts in the Dromagowlane/Trafrask river.

676. The IFI (Dr Gargan)1144 set out at some length its reasons, and academic support,1145 for disputing as incorrect the sEIS contention that there is no potential for farm origin lice to infect wild salmonids – effectively no sea lice risk – including that:

• any factor, such as lice, which reduces the number of salmonids in a river system will reduce the number of FwPM larva hosts.

• statistical models of a 26-year record suggest that salmon returns to spawn were over 50% lower in years following high lice levels on nearby salmon farms during the smolt out-migration.

• Dr Gargan’s published research in 2016, finding a significant positive relationship between the number of salmon lice in the local salmon farm and the number of lice found on sea trout collected at the same time in local rivers.

• Dr Gargan’s published research in 2017 finding a high risk of sea trout mortality dues to salmon lice at many sites in Bantry Bay in the early 1990s when salmon farms were operating there.

• The sEIS consideration1146 of the lice dispersal model’s assumption of neutral buoyancy of lice larvae in the specific conditions of Bantry Bay while acknowledging lice larva phototaxis,1147 misapplies the authority it cites,1148 is false, totally inappropriate and compromises the model outputs as to lice larva dispersal in the bay.1149

677. The Marine Institute (Dr Jackson)1150 replied that it concurred with the NPWS “that a decline in salmonid hosts for pearl mussel larvae is unlikely to occur particularly as the salmonid hosts in the Dromagowlane/Trafrask River are predominantly non-migratory brown trout.” I agree with the Sweetman Applicants that this appreciably overstated the NPWS view and, indeed, erroneously suggested that the NPWS had rested its view on the proposition that the hosts were “predominantly non-migratory brown trout.” In fact, that proposition appears to have emanated from the Marine Institute itself. However, it had been made, and disputed, at the oral hearing so little appears to turn on the point in terms of fair procedures.

678. In any event, ALAB issued a s.47 request to the Marine Institute inviting evidence to support its assertion that the salmonid hosts in the Dromagowlane/Trafrask system are predominantly non-migratory

1144 s.47 Reply 18 December 2018 1145 citing 16 academic references. 1146 sEIS p65. 1147 Movement in response to light.

1148 Amundrud and Murray (2009), Modelling sea lice dispersion under varying environmental forcing in a Scottish sea loch. Journal of Fish Diseases 32,

27-44.

1149 Essentially echoing this submission IFI’s Head of Operation, Dr Forde deposes in these proceedings that the sEIS assertion of “effectively a zero direct

risk of lice infestation of wild salmonids … is entirely at odds with the evidence presented by the scientific studies in the Bantry Bay Rivers in the 1990s”.

1150 S.47 Reply 15 February 2019 enclosing Jackson report 6 February 2019.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

brown trout and evidence of salmon in that river system. The Marine Institute replied1151 that the river system is not listed among the 261 fishery systems designated as holding salmon and/or sea trout. The presence of brown trout has been confirmed both by direct observations of Marine Institute personnel and in reports from relevant competent authorities. So, according to the Marine Institute, “the presence of the brown trout populations and the absence of any evidence for breeding populations of migratory salmon or trout lead to a conclusion that it is the brown trout and in particular the 0+ and 1+ year classes which act as hosts for” the FwPM larvae.

MERC AA Report, September 2020

679. The MERC AA Report was not concerned with the sea lice issue. However, it did recommend SBM for the Bantry Bay aquaculture sector to manage overall impacts from aquaculture and ensure that development and production is managed and co-ordinated in order to mitigate against adverse effects.

Lice Larva Buoyancy in EIA

680. SWI

• disputes ALAB’s conclusion that sea lice risk can be managed

• asserts a dearth of evidence that sea lice risk can be managed,

• disputes that ALAB had any proper basis to conclude that, advances in sea lice treatment reduce the overall impact of sea lice at all.

681. SWI1152 i. says that 25 years’ experience has yielded no proven reliable control to prevent the spread of sea lice from salmon cages to the wild.

ii. cites academic papers asserting that infestation of wild sea trout by sea lice from salmon farms can cause trout mortality and drive changes in population structure and that significantly fewer adult salmon return to their spawning grounds in bays where aquaculture is carried on than where it is not.

iii. states that while there is no certainty as to the cause, lice are a prime suspect.

iv. is critical of the DAFM Lice Strategy 2008 on which, it says, MOWI and ALAB rely, as mentioning “possible negative interactions with wild migratory salmonid populations” but failing to examine what that impact might be.

v. says that MOWI’s IPM/SBMP 2016 provides for sea lice treatment only when fortnightly or monthly Marine Institute inspection finds sea lice numbers exceeding targets. But by the time that is

1151 S.47 Reply 14 May 2019. 1152 Affidavit of John Murphy 21 October 2022.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

discovered, two to three weeks after infestation has begun, millions of sea lice larvae will have escaped to the bay. SWI says this occurs everywhere there are salmon farms. SWI says that as the Shot Head salmon farm will be initially stocked with 850,000 smolts this will allow up to 425,000 or 1,700,000 ovigerous lice to concentrate in the small area of the farm before treatment even commences and that their infestive larvae will number in the millions.

vi. says that as treatments are applied to the caged salmon only, they will not protect wild juvenile salmonids outside the farm.

vii. says that, accordingly, the only protection lies entirely in the assumption that sea lice will not be carried in to any river mouth, which in turn depends on the hydrological modelling.

682. So, while SWI accepts that lice treatment efficacy and the effects of sea lice infestation is a factual issue not for determination by the Court, it says ALAB would have had to determine that efficacy issue if it had recognised that MOWI’s sea lice buoyancy assumptions invalidate the hydrological reports as to lice dispersal.

683. SWI (SWI CG6) and IFI (IFI CG4) assert fundamental error in the RPS 2015 Report and the sEIS1153 in assuming neutral buoyancy of sea lice copepodid larvae in seawater in their water modelling. This was canvassed in some detail and at length at the oral hearing. SWI say1154 that the whole point of their case in this regard is that MOWI’s hydrographic and dispersal modelling is unreliable as the larvae are positively buoyant. IFI’s position as to sea lice larva dispersal, articulated by Dr Gargan, was set out in its written statement to the oral hearing and in its reasoned but trenchant response1155 to the sEIS and is reiterated by Dr Gargan on affidavit in these proceedings. Put simply, SWI and IFI say that sea lice larvae are positively, not neutrally buoyant: they actively swim towards and concentrate at and near the surface of the waters in which they are situate.1156 IFI says ALAB completely misinterpret an academic paper of Amundrud & Murray1157 the lice dispersal modelling described in which, IFI says, explicitly assumed that the larvae are retained in the surface layer of the waters – i.e. are positively buoyant. Positive buoyancy and resultant location of the larvae in the surface layer, they say, has two important effects: it will influence dispersal,

• where, as is typically the case, currents are not uniform across the water column and

• as dispersal is more affected by prevailing winds than the sEIS modelled1158 and ALAB assumed. So, it is said, lice larva dispersal is, to at least a significant degree, independent of the hydrography of Bantry Bay.

The Bay is longitudinally aligned with the prevailing south westerly wind, which, it is said, “clearly could have very serious consequences for the accumulation of sea lice at the head of Trafrask Bay”.1159 So, there

1153 p28. 1154 Affidavit of John Murphy 21 October 2022. 1155 IFI (Gargan) to ALAB 19 December 2018 in response to MOWI’s sEIS.

1156 IFI say they can avoid freshwater layers, move towards host fish, away from predators and are attracted to light near the surface during the day and

sink away from the surface during the night.

1157 Amundrud and Murray (2009) ‘Modelling sea lice dispersion under varying environmental forcing in a Scottish sea loch’.

1158 See generally Supplemental EIS p43 et seq. 1159 Gargan, Affidavit 30 September 2022.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

is “a potential mechanism by which high numbers of … infective Copepodids can be carried into close contact with their out-migrating hosts”. 1160

684. IFI cites its participation in the EU-funded Lice Track study1161 developing a dispersal model based on existing validated modelling tools which do consider the active vertical behaviour of sea lice larvae and the participating scientists’ agreement that assuming neutral buoyancy compromises the assessment. IFI says that “treatment of sea lice larvae as neutrally buoyant is totally inappropriate” and compromises the accuracy of the sea lice dispersal simulations such that the conclusion of no sea lice risk to wild salmonids was incorrect. (Dr O’Toole is incorrect in asserting1162 that no-one in the appeal indicated how treating sea lice as neutrally buoyant compromised the hydrographic modelling.)

685. I note the conclusions of Amundrud & Murray that their paper, enables “general conclusions applicable to any fjordic-type sea loch”, “highlights the importance of understanding the influence of varying environmental conditions, such as local winds” and that local winds were the “dominant influence” on dispersal patterns. They note that while their results suggest that dispersal patterns are reasonably represented by their assumed two-dimensional surface model, more work is needed to test and verify this as the construction of their model highlighted lack of knowledge about the depth distribution of larvae. However, I emphasise that I do not have the expertise to conduct and do not purport to conduct an analysis of or draw conclusions from a scientifically complex paper.

686. MOWI’s sEIS responds at length1163 to this IFI position – asserting many virtues of the RPS model, asserting the particularly oceanic, unstratified, nature of the waters of outer Bantry Bay and its other hydrographic, wind-influenced and tidal flushing characteristics (as contrasted to Norwegian fjords). These include vertical current speeds 3 to 5 times copepodid larvae swimming speeds – hence disrupting phototaxis.1164 The sEIS addresses the behaviour of copepodid larvae and concludes that, in the specific case of Bantry Bay, assumption of neutral buoyancy is reasonable – explicitly “in contradiction of the view expressed by IFI”. The sEIS also concludes that, in particular in view of the historical maintenance of low lice levels on farm sites and the naturally low lice infestation potential of Bantry Bay open waters as a whole, there is effectively no lice risk projected from the proposed Shot Head site, to wild salmonids at any location, either in the open waters of Bantry Bay or in the immediate vicinity of the Trafrask River or any other estuary in the bay.

1160 IFI (Gargan) to ALAB 19 December 2018 in response to MOWI’s sEIS. 1161 with Norwegian and Scottish scientists. 1162 O’Toole Affidavit sworn 23 January 2023 §35. 1163 p57 et seq. 1164 Movement towards or away from light.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

687. ALAB’s Dr O’Toole1165 disagrees with SWI and IFI on buoyancy and disputes the alleged misinterpretation of Amundrud & Murray. That paper says as to “Distribution with Depth” that “In the absence of active motion, copepodid lice tend to sink … such that their near-surface location is actively maintained by a phototactic response1166 …” and assumes that "sea lice are retained in the surface layer". Dr O’Toole says that this amounts to neutral buoyancy – though she does not address whether phototaxis abates at night and the lice larvae sink in consequence. Dr O’Toole returns to the issue1167 stating that

• Amundrud & Murray found that assuming positive buoyancy and so tracking dispersal at the surface only makes no difference relevant to sea lice management – they disperse in the same direction/spatial pattern no matter at what depth they are modelled.

• any ability of sea lice larvae to move independently is overwhelmed by the currents and conditions in Bantry Bay.

• accordingly, sea lice larvae in Bantry Bay conditions are effectively neutrally buoyant.

688. ALAB accepted the conclusion of the sEIS that the Site will “present an overall low sea lice infestation” on the basis of the RPS water modelling.1168

Lice Larva Buoyancy – Conclusion

689. It is important to remember that, as explained above, EIA is essentially procedural – as long as it is lawfully performed and informs the development consent decision there is no rule of EIA that residual significant adverse environmental effect requires refusal of permission.1169

690. In that context it does seem to me that, though it is no doubt perfectly possible to legitimately disagree with both as to their merits as exercises in EIA and as they relate to lice larva buoyancy, as Dr Gargan does from a standpoint of considerable expertise, Dr Saunders’ final report and ALAB’s determination were thorough and comprehensive. The dispute between the experts as to lice larva buoyancy and its consequences is not one the Court can or should resolve – any more than I could resolve those disputes which SWI accepts I cannot resolve. Even were I convinced that ALAB resolved that dispute incorrectly on the merits, I could not interfere as to those merits save for irrationality. ALAB, not the courts, is the body empowered by law to resolve such disputes. So, I reject the challenge to its decision as to the lice larva dispersal modelling assumption of neutral buoyancy.

1165 Dr Ciar O’Toole – Affidavit sworn 29 June 2022. 1166 Movement towards or away from light. 1167 Dr Ciar O’Toole – Affidavit sworn 23 January 2023 §35 et seq.

1168 Determination §6.1.2 – also §7.2 generally adopting the contents of the sEIS and the water modelling report.

1169 See above and my citation of Case C-261/18 European Commission v Ireland, Opinion of Pitruzzella AG of 13 June 2019 and Fitzpatrick v An Bord

Pleanála & Apple [2019] 3 IR 617. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Lice Dispersal Modelling Generally

691. More generally as to lice dispersal modelling, IFI in their submission light on the following in Dr Saunders’ Final Report as allegedly demonstrating a lack of reasonable basis for ALAB’s conclusions in EIA as to the risk posed by lice:

“We would also, however, emphasise that although the best available modelling tools have been utilised for this analysis, it is important to recognise that this provides a predicted scenario based on a necessarily limited set of empirical variables (albeit at the “worst case” end of the spectrum) which may or may not fully represent the full range of hydrological conditions present within the study area”.1170

692. This isolation of text from its context in an 8-page section of the Dr Saunders’ Final Report and in particular from the preceding sentence which reads that “The information to hand suggests that the lice issue constitutes a manageable risk to both farmed and wild salmon …” does Dr Saunders a disservice. In so observing I am conscious that in the following passage he also expresses concern also expresses concerns as to lice.1171 However the point here is, first that modelling was not the only basis for the conclusion as to lice. It was certainly important and while not a panacea I cannot see how in EIA generally (the WFD may raise different issues) use of the best modelling software can be criticised. Remembering that EIA is not about elimination of all significant risk but is about evaluation and acceptability of residual risk, it is entirely unsurprising – indeed eminently desirable – that EIA should articulate uncertainties. I have addressed above the significance of uncertainty in EIA – it is a proper feature of, not at all necessarily a flaw in, EIA.

Lice & Risk to Salmonids – ALAB’s Determination, 29 June 2021.

693. ALAB, in its Determination, recorded in its EIA1172 and in its evaluation for purposes of s.61 of the 1997 Act1173 its consideration of the information on the lice issue, including that provided in the RPS hydrological modelling, and that provided by the Marine Institute and IFI, the Sea Lice Protocol 2000 and the Technical Advisor’s Final Report and concluded, notwithstanding the differing views,1174 that:

• the Site is hydrologically isolated from adjacent main rivers and other fish farms.

• The combination of o a licence condition requiring compliance with the Sea Lice Protocol #3 2000 and o MOWI’s Pest Management Plan, would reduce any risk due to lice to a reasonable, non-significant level.

1170 §9.1 Increased threat to wild salmon and sea trout from sea lice, p66. 1171 I deal consider elsewhere in this judgment the question of inconsistencies ion Dr Saunders’ report. 1172 ALAB Determination §4.2 et seq. 1173 ALAB Determination §6.1.2. 1174 ALAB Determination §4.2.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• the Site will present an overall low sea lice infestation and pollution risk.

• any risk to wild salmon from sea lice infestation is unlikely to significantly affect the conservation status of

salmon in Bantry Bay.

• there were no significant risks of infestation of wild salmonids migrating from/to the Dromagowlane/ Trafrask River by sea lice from the proposed activity at the Site.

ALAB also cited the O’Toole site visit to the Dromagowlane/Trafrask River to which I will later refer.

694. “Notwithstanding” these findings of no significant risk, ALAB determined1175 that the Licence should include, in Schedule 5, an additional condition requiring MOWI to comply with any SBMP or code of practice for Bantry Bay developed in agreement with any relevant State body (i.e. the SBMP Condition).

Lice & Risk to Salmonids – Selective Evidence – Decision

695. IFI submit that

“ .. significant reliance was also placed by ALAB on evidence suggesting that the risk of contamination by farmed salmon to wild salmonids was very low. In this regard, the Applicant is of the view that only very few and selective scientific publications were referred to in the assessment of potential risk of the sea lice impact on wild salmonids and that the great majority of recent publications on the subject were not taken account of and for no apparent reason. This matter appears to be acknowledged by the Technical Advisor (at page 63 of the Final Report) where it is stated: “We acknowledge that the wider body of scientific literature may not have been fully considered when undertaking the assessment of the potential impact of the Shot Head site on wild salmon populations in Bantry Bay”.

696. I recite this passage of the IFI submission in extenso as it contains the seeds of its own destruction. Notably also as to a complaint of selectivity, it cites Dr Saunders shorn of context. Dr Saunders’ note that “the wider body of scientific literature may not have been fully considered” appears to be a reference to the Minister’s EIA of 2015 in its reliance on a 2013 study by Jackson. However events had clearly moved on by the time of Dr Saunders’ report of December 2020. He is clearly himself aware of the wider literature, first in that he is equipped to point out the Minister’s failure to fully consider it. Also, Dr Saunders himself describes the wider literature as including doubt cast on Jackson’s study by Krkosek et al., and Jackson’s response. Dr Saunders states that “Numerous other studies have, however, been conducted by the international scientific community, assessing multiple locations where salmon farms have been operating. Several literature reviews have examined the evidence relating to the risks that farmed sea lice pose to wild salmonid population.” He

1175 ALAB Determination §6.1.2.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

cites three, including, notably, one by Dr Gargan of IFI – who opposes the current project on grounds of lice risk to wild salmonids. It is to Dr Saunders’ credit that he acknowledges the omission, exercises expert judgment as to its significance and remedies it. Notably also as to a complaint of selectivity, the cited passage as to omission of full consideration of the literature omits the following sentence: “We reiterate, however, that local hydrographical conditions are key to understanding the specific risks to any particular water body.” Overall, I cannot read this passage as undermining the ultimate conclusion of Dr Saunders’ report: “On taking into account all of the above we can offer no substantive technical reasons to refuse the current licence application.”1176

697. Another answer to IFI’s complaint is that it does not amount to a complaint of irrationality. Expert

decision-makers must consider all relevant materials, but selection from those materials of some as more or less useful and important to an EIA is precisely what expert decision-makers do. That is what making a

decision based on judgment requires. Whatever the developing views of the law of irrationality, it is at very least illustrative to note the O’Keeffe requirement that the applicant to quash a decision must establish “that the decision-making authority had before it no relevant material which would support its decision.” That standard clearly reflects the entitlement of a decision-maker to decide to rely on some materials rather than other materials. The very accusation of selectivity implies that that there was such material before ALAB – that there was, as IFI appear to accept, “evidence suggesting that the risk of contamination by farmed salmon to wild salmonids was very low”. The challenge on this ground must fail.

698. In this context I note also that ALAB’s Determination1177 records that Ml and IFI provided (conflicting) information to the Board on the sea lice issue – listed at §§2.9, 2.13.6, 2.13.8, 2.29.2, 2.30 and 2.31 of the Determination. ALAB considered this information and the findings of Dr Saunders’ Final Report. Despite the differing views, ALAB was satisfied that o risks to wild salmon from sea lice infestation is unlikely to be a significant factor influencing conservation status of salmon stocks in Bantry Bay.

o there were no significant risks of sea lice infestation of wild salmonids migrating from/to the Dromagowlane and Trafrask Rivers.

Again, this all seems to me a matter eminently for decision by ALAB rather than by the Court.

1176 p108. 1177 §4.1

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

LICE & SBM – IMPOSSIBILITY OF COMPLIANCE WITH LICENCE AS TO SBM

SWI CG8

699. As noted earlier, SWI CG8 pleads that “The Impugned Decision is invalid for impossibility as ALAB authorised asynchronous stocking of fish farms in Bantry Bay but required compliance with DAFM Monitoring Protocol No. 3 For Offshore Finfish Farms – Sea Lice Monitoring and Control, which requires a single bay management system and synchronous fallowing.”

700. Compliance with Sea Lice Protocol #3 is required by Condition 7.3 of the Licence, so the plea clearly amounts to an assertion that Condition 7.3 is impossible of compliance. SWI’s particulars also plead ALAB’s imposition of the Schedule 5, SBMP Condition – which, as we have seen, requires compliance with any agreed SBMP.

701. The pleaded particulars of SWI CG8 include the following:1178

• SBM/CLAMS is a process whereby a single bay is managed as a whole to minimise disease, including sea lice infections.

• Lice Protocol #3 requires that “All fish farms operating in a particular bay will be required to undertake appropriate synchronous sea lice treatment and control strategies through” SBM/CLAMS. In short, it is pleaded that Lice Protocol #3 requires synchronous sea lice treatment and control.

• MOWI proposed asynchronous stocking of the proposed Shot Head farm and its Aghabeg/Roancarrig farm.

• The appellants opposed this as, o SBM/CLAMS is not established in Bantry Bay. o asynchronous stocking is incompatible with SBM. o worthwhile SBM would require synchronised fallowing.

• Dr Saunders’ Final Report says that asynchronous stocking is incompatible with SBM (which requires synchronous fallowing and synchronous lice treatments). o It is convenient to note here that this is an incomplete plea of the content of Dr Saunders’ report as to these matters. I have above, ▪ teased out internal inconsistencies in Dr Saunders’ report in this regard.

▪ noted his view that MOWI’s proposed asynchronous production strategy “complies with SBM”.

▪ noted his view that asynchronous production is not in line with CLAMS best practice and that at a minimum, all MOWI sites “should be operated synchronously.”

▪ concluded that,

1178 Which I have edited and rearranged without altering meaning.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

 read as a whole and despite its inconsistencies, Dr Saunders’ report approves the prospect of asynchronous production by the Shot Head and Roancarraig/Aghabeg sites.

 his approval encompasses all risks posed by sea lice.

• By Condition 7.3, MOWI must comply with Lice Protocol #3. MOWI cannot comply unless an SBMP is in place, which it is not. So Condition 7.3 is invalid as incapable of being complied with. o (This is in substance the impossibility plea.)

• As ALAB’s Determination1179 to grant the licence is predicated on compliance with Lice Protocol #3 reducing “any risk to a reasonable, non-significant level”, the invalidity of Condition 7.3 necessarily entails the invalidity of the Impugned Decision. o (To put it another way Condition 7.3 is both invalid and incapable of severance from the Aquaculture Licence.)

• No other rationale is apparent from ALAB’s decision, and the reasons for the decision are in that respect inadequate and in breach of s.40(8) of the 1997 Act.1180

702. This ground of challenge should have been more clearly pleaded and particularised. Not least, on occasion, SWI is unclear as to which of Condition 7.3 (Sea Lice) and the SBMP Condition are in issue. I should add that the other parties also refer to “the condition” on occasions without being clear which condition is in issue. Nonetheless, on a fair and reasonable reading of the plea, the meaning is discernible. That meaning asserts impossibility, by reason of ALAB’s authorisation of asynchronous production (and hence asynchronous lice treatment and fallowing), of compliance with Condition 7.3 which, SWI says, requires synchronous production, lice treatment and fallowing. However, as the case ran, SWI CG8 was advanced more as an argument that Condition 7.3 was unenforceable and/or uncertain.

703. It is not specifically pleaded that the SBMP Condition was invalid as requiring the agreement of other producers or as aspirational only. Nor is any plea raised by any applicant as to doubt or discrepancies regarding lice treatment trigger levels.

SWI Submissions

704. SWI’s submissions include the following:

• ALAB erred in law in imposing a condition that requires compliance with Sea Lice Protocol #3 which requires SBM; compliance with the condition is impossible.

1179 Of 29 June 2021. Part 4: EIA §4.4. 1180 s.40(8) requires that a determination state the main reasons and considerations on which it is based.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Sea Lice Protocol #3 requires, o synchronous sea lice treatment across all sites in a bay, regardless of operator. o synchronous sea lice treatment achieved through SBM – in particular an SBMP to coordinate different operators.

• Compliance with the SBMP Condition is impossible as SBM is not established in Bantry Bay.

• There is at present no way to enforce synchronous treatment as required by Sea Lice Protocol #3.

• Whether, as ALAB found,1181 the site is “hydrologically isolated” and presents a low infestation risk is irrelevant to the requirement of compliance with a licence condition.

• ALAB’s rejection of SBM and synchronized production is irrelevant to the issue. The issue is synchronised lice treatment.

• MOWI’s submission deems the relevant part of Sea Lice Protocol #3 to be effectively inapplicable for now whereas ALAB’s Determination requires compliance with Sea Lice Protocol #3 – including synchronous lice treatment.

• Condition 7.3 requires the Licensee to do something it has no means to do: comply with a Sea Lice Protocol #3, which requires all operators in the bay to synchronise lice treatments. Such a condition is neither enforceable, nor reasonable.

• A condition that cannot be complied with is invalid. So Condition 7.3 and the reasons for it are invalid.

Because ALAB relied on the reason as a reason for its determination, the Licence is invalid.

705. At trial, SWI submitted that Condition 7.3 in substance requires SBM as the means of effecting Sea Lice Protocol #3, and

• as SBM requires the voluntary agreement and participation of other aquaculture licensees, Condition 7.3 is impossible to enforce.

o Note that this argument is of impossibility of enforcement – as opposed to the pleaded argument of impossibility of compliance.

• paradoxically, the SBMP Condition actually takes from the requirement which ALAB sought by the SBMP Condition to reinforce – as it requires MOWI to comply with an SBMP only if one is in place for Bantry Bay and such an SBMP can be put in place only if MOWI agrees one.

706. SWI, as to “the impossible conditions point”, insisted1182 on a “distinction between a condition that cannot be complied with and a condition that cannot be enforced” and submitted, of the condition requiring SBM, that it “cannot be enforced” – in breach of the tests set in the analogous planning law cases of Ashbourne Holdings1183 and Killiney & Ballybrack.1184 I observe that those tests are that:

1181 ALAB Determination §6.1.2. 1182 Transcript day 6 p66. 1183 Ashbourne Holdings v Bord Pleanála [2003] 2 IR 114 §123. 1184 Killiney and Ballybrack Development Association v Minister for Local Government, [1978] I.L.R.M. 78, §80.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• conditions must be necessary, relevant to planning,1185 relevant to the permitted development,1186 enforceable, precise and reasonable.

• the reason for a condition must be fairly and reasonably capable of justifying the condition. For example, “if the reason given is the attainment of an objective, and compliance with the condition could not possibly attain that objective, the condition will be held bad because it was given for an unreasonable reason.”

707. However, in substance and of these various tests, SWI relied only on that as to enforceability of a condition. As stated, its written submissions argue that Condition 7.3 requires MOWI,

“… to do something it has no means to do: comply with a Protocol which requires all operators in the bay to synchronise lice treatments. Such a condition is neither enforceable, nor reasonable. The Condition and the reason for it are therefore invalid and …(so)… the Licence is also invalid.”

It will be seen that this submission conflates three, arguably overlapping, concepts: impossibility of compliance, impossibility of enforcement and irrationality.

708. The gravamen of SWI’s invoking a distinction between conditions impossible of compliance and those impossible of enforcement was not entirely clear. But taking it, as seems to me, most favourably to SWI, the gravamen seems to have been one of identifying both phenomena as species of impossibility so as, arguably, to bring unenforceability of Condition 7.3 within SWI’s pleadings. But, as has been seen, SWI CG8 is clearly pleaded in terms of impossibility of compliance and not in terms of impossibility of enforcement.

ALAB & MOWI Opposition

709. ALAB’s Opposition appears to have misunderstood the essential element of SWI CG8. I have some sympathy with them in this regard. Its pleas are directed primarily at the allegation of impossibility of compliance with the SBMP Condition as opposed to Condition 7.3. I allow it some latitude in this regard. It replies that,

• The SBMP Condition is contingent and forward looking: if and when a “bay wide single pain management plan or code of practice for Bantry Bay” is “developed” in agreement with “any relevant State body” then the licensee must comply with it.

• ALAB’s determination1187 records that ALAB was satisfied that,

1185 As applied here, relevant to aquaculture. 1186 As applied here, Shot Head salmon farm. 1187 §6.1.2

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o “.. the site is hydrologically isolated from adjacent main rivers and other fish farms and will therefore present an overall low sea lice infestation and pollution risk”.

o ALAB deemed the necessary compliance with Sea Lice Protocol #3 and the Pest Management Plan “sufficient and reduces any risk to a reasonable, non-significant level.”

• Dr Saunders Final Report1188 states that o The advantages of SBM are outlined in the EIS1189 and in this report.1190 o Single bay production is a non-statutory aspiration in CLAMS, whereby all sites in the bay have a synchronous production schedule, allowing fully coordinated whole-bay lice treatments and fallow periods. o Extensive hydrographic modelling1191 and historical low lice numbers in the bay indicate that there is adequate hydrological distance between sites, which would support the view that a single bay production strategy is currently not necessary in Bantry Bay. o Should these conditions change in Bantry Bay, single bay production may be considered. Or it might become necessary in the event of uncontrollable lice infestations. o At this time, however, we do not believe that the establishment of an additional farm site at Shot Head requires the imposition of synchronised whole-bay production as a licence condition.

• ALAB was entitled to impose “this condition” on the Site but was not entitled to impose such a condition on other aquaculture sites.1192

• All MOWI sites in Bantry Bay would be operated in accordance with MOWI’s Pest Management Plan/SBMP o “on a synchronous basis in terms of sea lice treatments”.1193 o “under the single bay management practises”.1194

▪ I recall here that MOWI’s position at trial was that the Aquaculture Licence does not require compliance with its Pest Management Plan – the MOWI IPM/SBMP 2016 – as it imposes no condition to that effect.

• ALAB’s reasons and considerations in imposing the SBMP Condition are adequate for the purposes of s.40(8) of the 1997 Act.

710. MOWI appears to have more accurately understood SWI’s plea in that its Opposition denies that the Aquaculture Licence is invalid because it requires MOWI to comply with Sea Lice Protocol #3. As I have

1188 §9.11.4. 1189 §3.2.2., p149. 1190 §9.1 Saunders’ Final Report. 1191 RPS, 2015. 1192 §6.8.7 of ALAB’s Determination. 1193 Opposition §62. 1194 Opposition §93.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

noted, the condition requiring compliance with Sea Lice Protocol #3 is Condition 7.3. Nonetheless, the concentration of MOWI’s pleading is on the SBMP Condition and I will allow I similar latitude. It essentially repeats ALAB’s pleas and adds that:

• the SBMP Condition is not unworkable.

• MOWI’s obligation to comply with “this aspect” of Sea Lice Protocol #3 “will only crystallise” when SBM is

introduced in Bantry Bay, which has not yet occurred.

ALAB & MOWI Submissions

711. ALAB note that SWI’s submission confirms that “the issue is synchronous treatment”. But “Synchronous Sea Lice Treatment and Control in Bays” (§5) is only one of six aspects of Lice Protocol #3. The others include,

• §1 “Monitoring Regime Required”. Lice Protocol #3 requires ongoing monitoring for sea lice and remedial action if they are found.1195

• §4 “Trigger Levels for Treatment”. Lice Protocol #3 requires “[t]he setting of appropriate treatment triggers” as “an integral part of implementing a targeted treatment regime”.

• §6 “Sampling Strategy”. Lice Protocol #3 sets out a “sampling strategy methodology” to “[p]rovide a robust and reliable objective measure of lice numbers on farmed fish”.

ALAB observes that all those other aspects of Lice Protocol #3 could be effected.

712. ALAB emphasises that, by Lice Protocol #3 §5, synchronous lice treatment is to be undertaken via the SBM/CLAMS “process”. That envisages a “process” to produce “a voluntary grouping and agreement between producers at individual bay level for” SBM.1196 ALAB’s Determination recognised that SBM was not in place but might be developed and required in addition to Condition 7.3, an “additional” condition being the “future-looking” SBMP Condition requiring compliance with “any” future SBMP.

713. ALAB says that the additional SBMP Condition does not undermine the basis for ALAB’s conclusion that the Site will “present an overall low sea lice infestation”. ALAB’s starting point was that the site is “hydrologically isolated from adjacent main rivers and other fish farms”. Compliance with the “Pest Management Plan” and Sea Lice Protocol #3 were, without the additional condition, considered sufficient to “reduce any risk to a reasonable, non-significant level.” In that context, ALAB had regard to the requirement

1195 “All finfish farms are obliged to monitor for sea lice on an ongoing basis and to take remedial action. This involves the inspection and sampling of each

year class of fish at all fish farm sites fourteen times per annum, twice per month during March, April and May and monthly for the remainder of the year

except December-January. Only one inspection is carried out during this period.” 1196 §15 Ciar O’Toole’s affidavit for ALAB sworn 23 January 2023.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

to comply with the “Pest Management Plan” and Sea Lice Protocol #3, being aware that the aspect of Sea Lice Protocol #3 relating to “Synchronous Sea Lice Treatment and Control in Bays” required an SBMP which was not in place but might be developed.

714. In other words, ALAB say that, as a matter of interpretation of its Determination, it recognised that there was no SBMP in place and, as the synchronised lice treatment requirement of Lice Protocol #3 §5 is dependent on SBM, so its requirement of compliance with Sea Lice Protocol #3 must be read as not requiring synchronised lice treatment.

715. In support, ALAB cite its pleaded reliance on Dr Saunders Final Report1197 – part of which I repeat in summary:

• Synchronous whole-bay production allows coordinated whole-bay lice treatments.

• Single bay production might become necessary in the event of uncontrollable lice infestations.

• Single bay production is currently not necessary in Bantry Bay.

716. ALAB assert that if MOWI proves unable to comply with any SBM which may be developed, that does not invalidate the Condition or the Licence.

• (I observe that this misses SWI’s point – which is that Condition 7.3, in requiring compliance with Sea Lice

Protocol #3, is invalid for impossibility because Sea Lice Protocol #3 requires, now not in the future, synchronous lice treatment in accordance with a non-existent SBMP.)

717. ALAB cites O'Sullivan1198 and People Over Wind1199 for the proposition that a developer’s inability to comply with a planning condition does not invalidate the permission – rather it means that the developer can’t develop on foot of the permission.

718. MOWI’s written submissions argue that:

• SWI’s case appears to be confined to the argument that MOWI cannot comply with Sea Lice Protocol #3 absent an SBMP so Condition 7.3 cannot be complied with and so is invalid.

• SWI is incorrect in alleging, that absent an SBMP in Bantry Bay, synchronous treatment cannot work.

MOWI cites the affidavit of Catherine McManus.1200 She asserts that:

1197 §9.11.4. 1198 O'Sullivan & ors v An Bord Pleanála & ors [2017] IEHC 716, Costello J, §35. 1199 People Over Wind Environmental Action Alliance Ireland [2015] IECA 272, Hogan J, §45. 1200 sworn for MOWI 23 January 2023.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o SBM and SBMPs require coordination and compatibility of best practices of neighbouring farms as to stocking, fallowing and lice treatment but not necessarily synchronicity of such practices.

o there is “no requirement for the Shot Head fish farm to be fallowed synchronously with the other sites in Bantry Bay as it is hydrologically isolated from adjacent rivers and other fish farms.”

I have to say I am unclear whether MOWI argues that:

• synchronous treatment can work despite asynchronous production or

• synchronous treatment is not required by the Aquaculture Licence.

719. MOWI’s written submission includes the following passage:

“Currently, there is no Single Bay Management Plan in Bantry Bay that covers the proposed fish farm at Shot Head, given that the farm has not yet been constructed. However, once the fish farm is constructed and becomes operational, the Shot Head Fish Farm will by agreement with the Department and the Marine Institute, join the Single Bay Management Plan for Bantry Bay, and comply with the requirement of the Protocol on Sea Lice.”1201

720. MOWI’s written submission cites its MOWI IPM/SBMP 2016 as outlining how it will monitor and control sea lice, co-ordinating with the other farms in the Bay in an SBMP which can only be made once Shot Head is operational – including that “Where two sites are stocked in the Bay, treatments will be carried out on both during the same period and with the same chemical class.”

721. MOWI cites Arklow Holidays1202 for the proposition that inability to comply with a licence condition does not invalidate the licence – rather would leave MOWI open to enforcement action including the possible revocation of the licence.

Lice & SBM – Impossibility of Compliance with Licence as to SBM – Discussion & Decision

722. I respectfully remind the reader that I have, in the introduction to this judgment, explained why, in the particular factual and legal matrix of this case,

1201 §99. Emphases added. 1202 See Arklow Holidays Ltd v. An Bord Pleanála [2006] IEHC 15.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• the environmental effects of the Shot Head salmon farm, as they relate to sea lice and their posited direct effect on the wild salmon and indirect effect on the FwPM, are EIA issues only – and specifically are not AA issues.

• if EIA is done lawfully and is lawfully taken into account in determining whether to grant an aquaculture licence, foreseen significant adverse effect on wild Salmon and the FwPM are not in law preclusive of the grant of grant an aquaculture licence.

• whether any such significant adverse effect is acceptable, such that an aquaculture licence should be granted, or unacceptable such that it should be refused, is a matter for the expert judgment of ALAB reviewable, as to merit, only for irrationality. As I have said, genuine, vehement, rational and/or expert disagreement with the judgment of a decision-maker does not, per se, imply irrationality of its decision. I add that it is characteristic of discretionary decision-making that more than one view may be rational.

Indeed, that is often – perhaps even typically – the case.

723. On this issue, my essential conclusion can be briefly stated. For the reasons I have attempted to explain, Condition 7.3, in requiring compliance with Lice Protocol #3 2000, does not appear to me require, as a matter of law, synchronicity between the Shot Head and Roancarraig/Aghabeg salmon farms. That is because such synchronicity is not “appropriate” as ALAB has in substance determined that the farms are not “adjacent” within the meaning of Lice Protocol #3 2000 and the Lice Strategy 2008. Rather, ALAB has determined them to be “hydrologically isolated” in that they do not lie within a single tidal excursion of each other. Accordingly, the posited impossibility of compliance with Condition 7.3, as to synchronicity, does not seem to arise and the challenge to the Licence on SWI CG08 fails.

724. Despite counsel’s adroit argument, I take the view that pleading that a licence condition is void for impossibility of compliance or, to put it another way, that the licence is internally contradictory, is a different plea from a plea that a licence condition is unenforceable. So I do not decide the issue. However, as I will quash and remit the Aquaculture Licence on other grounds, I hope ALAB may find it useful in reconsidering the matter to consider my further views as set out, obiter, below.

725. As I have noted, ALAB has cited O'Sullivan1203 and People Over Wind1204 to the effect that a planning condition with which the developer cannot comply does not invalidate the licence – it means rather that licensee can’t act on foot of the licence. MOWI’s reliance on Arklow Holidays is at base to the same effect. However it seem to me that there is a difference between,

1203 O'Sullivan v An Bord Pleanála & Keel Energy [2017] IEHC 716, Costello J §35. 1204 People Over Wind Environmental Action Alliance Ireland [2015] IECA 272, Hogan J §45.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• on the one hand, a condition with which a developer cannot, as a matter of fact – as a practical matter – comply, such that he will be unable to develop – as in O'Sullivan and People Over Wind. Such a situation is merely the working out of the requirements of a valid condition.

• on the other hand, a condition which cannot be complied with because it is internally inconsistent with the grant of the licence – with other terms of the licence. It is the latter sort of impossibility which was alleged here as invalidating the licence in that it was alleged that the licence for operation of Shot Head salmon farm asynchronously with the Roancarraig/Aghabeg salmon farms contained a condition requiring operation of Shot Head salmon farm synchronously with the Roancarraig/Aghabeg salmon farms.

726. The latter seems to me to be a form of impossibility to which different considerations apply than those at play in O'Sullivan and People Over Wind. However, as I have found that on the facts of this case Condition 7.3, in its application of the Lice Protocol #3 2000, does not require operation of Shot Head salmon farm synchronously with the Roancarraig/Aghabeg salmon farm, I need consider no further the possible legal implications of such an internal inconsistency, beyond noting that this situation seems to me more analogous with McKenna.1205 In McKenna, the District Court ordered the Commissioner of Gardaí to grant a taxi licence to Mr McKenna despite his criminal record and the Commissioner’s opposition. The Commissioner issued a licence subject to a condition that very significantly restricted the work Mr McKenna could do. The High Court held that conditions to a licence must be rational, reasonable and proportional as those criteria relate to the purposes of the statute under which the licence is granted and to the underlying purpose for which licence is granted. The condition was irrational as “so restrictive as to, in effect, set at nought the decision of the District Court.” Putting it more prosaically, a condition cannot take away with one hand what the licence in which it is found has granted with the other. However, the present is at very least a cautionary tale as to the confusion which can arise from incorporation of other documents – protocols, standards, guidelines and the like – by reference in a licence or permission unless care is taken to peruse them for consistency with the licence and to ensure that they are otherwise in terms suitable for incorporation. As noted earlier in this judgment, a similar difficulty was commented upon in Fernleigh.1206

727. Dr O’Toole somewhat ambiguously says that ALAB can’t impose a licence condition that requires the co-operation of other licensees in Bantry Bay.1207 That is certainly true in the sense that a licence can’t purport to impose obligations on other licensees. But whether a licence condition can make the licensee’s operation on foot of it conditional on such co-operation by other licensees may be a different matter which would require further argument were such a condition seen by ALAB as necessary and imposed.

1205 McKenna v Commissioner of An Garda Siochana [2023] IEHC 437. 1206 Fernleigh Residents Association v. An Bord Pleanála [2023] IEHC 525, §55.

1207 She in fact says that “It was not open to ALAB to impose a condition that relates to other licences in Bantry Bay.”

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

728. Though I do not decide the point as applicable to this case as, in my view, it was not pleaded, it is clear as a matter of general law that an unenforceable planning condition is invalid – Ashbourne Holdings1208 and Aherne.1209 By close analogy, so too must be an unenforceable aquaculture licence condition. While in some cases that is because the developer is discommoded by not knowing precisely what it must do to comply (inter alia to avoid criminal culpability), it is clear and was envisaged in Aherne that the question of enforceability must also be viewed from the perspective of a person with an interest in enforcing the condition or seeking to have it enforced. On the history and facts of this case, it can only be inferred that environmental and angling interests, perhaps even the IFI, may well take a keen and, (at least as a general proposition) legally legitimate, interest in keeping MOWI up to the mark as to its compliance with aquaculture licence conditions. If those conditions are unenforceable, that legitimate interest will be frustrated and the condition, and any environmental and other protections it purports to effect, will prove illusory. I also accept that enforceability is a requirement of reasonableness of a condition and is in a real sense a practical question: as was said in Tew:1210

“A condition that is not reasonably enforceable is not a reasonable condition for the purpose of the Newbury test.1211 …………….The enforceability of any condition has to be considered in the light of the particular facts. The problem of enforceability is a practical one, and it is undesirable to fetter the broad discretion (to impose planning conditions) …. any more than is strictly necessary in the light of the principles set out in Newbury.”

729. However, the SBMP Condition in this licence does not make operation of the salmon farm conditional on an SBMP – either as a pre-condition or a condition subsequent. In appearance it makes operation of the salmon farm conditional on an SBMP as a condition subsequent but that is illusory as the very existence of an SBMP depends not just on the agreement of an SBMP by all other salmon farmers in the Bay (or, possibly in a designated part of the Bay) – it depends on the agreement of MOWI to such an SBMP. In other words, MOWI can veto the very existence of any SBMP as applicable to it.

730. A condition with which a licensee need comply with only if it wishes to, is not a condition at all. In my view, the SBMP condition is no more than precatory – a pious hope entirely reliant for its effect on MOWI’s goodwill. I have earlier made similar remarks as to Conditions 6.5 and 7.3. A condition is characterised by compulsion of the licensee or permittee.

1208

[2003] 2 IR 114 – citing as “a reasonable common sense view of” the law, The Department of the Environment’s Development Control Advice and

Guidelines which advised inter alia that a planning condition must be enforceable. 1209 Aherne v An Bord Pleanála [2015] IEHC 606. 1210 R(Tew) v Rochdale Metropolitan Borough Council [1999] 3 PLR 74, [1999] Lexis Citation 2603.

1211 Newbury District Council v Secretary of State for the Environment [1981] A.C. 578 : “… conditions imposed must be for a planning purpose and not

for any ulterior one, and they must fairly and reasonably relate to the development permitted. Also they must not be so unreasonable that no reasonable

planning authority could have imposed them.” Newbury was cited as to the validity of planning conditions, and with apparent approval, in Ashbourne

Holdings v An Bord Pleanála [2003] 2 I.R. 114 and in Quinlan v An Bord Pleanála & Dublin City Council [2009] IEHC 228.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Summary

731. In short:

• I reject as not arising on a proper interpretation of Condition 7.3 and Lice Protocol #3 as applicable to the facts found by ALAB – notably that of hydrological isolation of the Site – the plea that Condition 7.3 in requiring SBM is impossible of compliance for internal inconsistency with the Licence as it permits production at Shot Head asynchronous with that at Aghabeg/Roancarraig.

• I reject as not pleaded, the arguments that licence conditions are impossible of enforcement.

• I have made certain obiter remarks as to the necessity that licence conditions be reasonably and practically enforceable – generally and as characterised by compulsion of the licensee.

LICE & FWPM – EVIDENCE OF DR GOOD, O’TOOLE FWPM REPORT & ALAB CONCLUSION

732. Amongst “fundamental errors” alleged by IFI against ALAB is its alleged ignoring of the evidence of Dr Good to the oral hearing in 2017 as to risks to, and the identification of host fish of, the FwPM in the Dromagoulane/Trafrask River. There is a pleading dispute on this issue but it is easier to address the issue. I have briefly recited that evidence above. I should add that the evidence of Dr Good was certainly not ignored by Prof McIntyre – whose report summarised it fairly and reflected its importance in recommending that an sEIS address the issue – which it did. I have opined that the evidence of Dr Good of NPWS to is to be taken with the NPWS s.47 reply of 16 November 2018 and I add, for that matter, all other evidence on these issues, including the sEIS and responses thereto. It is fair to say that the controversies on these issues were fully ventilated on the papers before ALAB. That it considered this evidence must be inferred from – indeed is explicit from – the ALAB minute of 10 December 2020 in which “having reviewed all information at its disposal” (which, of course, included Dr Good’s evidence and Prof McIntyre’s report of it) it acknowledged it had to be “meticulous” on a number of issues including risk to the FwPM. This prompted ALAB’s decision1212 to ask Dr O’Toole to make a site visit to the Dromagoulane/ Trafrask river system to investigate the rivers’ suitability for salmonid spawning access, particularly to the middle and upper reaches of the catchment and on the question which particular juvenile salmonids – salmon, sea trout and/or brown trout – were likely to play host to the FwPM glochidia

733. Dr O’Toole’s site visit on 5 March 2021 sought to establish whether it was more likely that the mid and upper reaches of the Dromagoulane/Trafrask river system, above a number of complex falls, were difficult for migratory salmon and sea trout to access and were used for spawning by non-migratory Brown

1212 Minutes of ALAB meetings 10 December 2020 and 12 January 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

trout, and so were therefore mainly inhabited by Brown trout juveniles. Her resulting report was dated 28 May 2021 and cited in ALAB’s determination as the “Freshwater Pearl Mussel Report”. I will cite it as the “O’Toole FwPM Report 2021”. Dr O’Toole considered it possible but improbable that migratory salmonids would access the middle and upper reaches of the catchment. Any access for spawning adults would be limited to times of high flow and limited suitable spawning habitat was found upstream. She cited Dr Good, Dr Jackson, the IFI’s electrofishing survey of 2017 and Dr Saunders’ final report of December 2021 and opined:

“I feel it is entirely reasonable to form the scientific opinion that in this small catchment, the majority of the salmonid host for FwPM would be non-migratory brown trout juveniles that are resident in that part of the catchment. As such, it is my considered opinion as the ALAB Technical Advisor that a decline in the migratory salmonid populations of the Dromagoulane/ Trafrask catchment would not negatively impact on any remaining FwPM populations present in the catchment.”

734. One might suggest that the first sentence is more edging towards rather than expressing a

conclusion and the words “as such”,1213 introducing the second sentence, could be read as fallaciously implying that because an opinion is reasonable it is necessarily held. Clarity is expected of experts – of whom clear report-writing is expected as a core skill. That said, excessive pedantry is to be avoided in the interpretation of such reports, which are not to be construed as if statutes and require a focus more on substance than on form or precise words – see Eoin Kelly1214 and cases cited therein. Excessive formalism is to be avoided and, as McGovern J. said in Ó Gríanna of EIA, “The principle of effectiveness is not a mandate for construing the Directive in the most onerous manner possible. This involves the courts being astute to ensure the objectives of the Directive are met but not in an overly pedantic way”.

735. What matters, therefore, in the O’Toole FwPM Report 2021, is her clear opinion that “a decline in the migratory salmonid populations of the Dromagoulane/Trafrask catchment1215 would not negatively impact on any remaining FwPM populations present in the catchment.”

736. The ALAB determination1216 said of the O’Toole FwPM Report 2021 that it,

“……… concluded that due to the hydrology, size and recorded location of Freshwater Pearl Mussel within the catchment, the host species for juvenile glochidia larvae of these Freshwater Pearl Mussel populations was highly likely to be juveniles of resident brown trout. This means that it is highly unlikely that the pearl mussel populations in the Dromagowlane/Trafrask River are reliant on

1213 A phrase often misused.

1214 Kelly v An Bord Pleanála & Aldi Stores [2019] IEHC 84 §99 et seq, citing Ó Gríanna v An Bord Pleanála (No. 2) [2017] IEHC 7, Dublin County Council v

Eighty Five Developments Limited (No. 2) [1993] 2 I.R. 392 and Buckley v An Bord Pleanála [2015] IEHC 572. 1215 i.e. due to sea lice infestation. 1216 §4.4.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

juveniles from migratory salmonids (salmon and sea trout) as hosts for its larval stage and that any decline in the migratory salmonid populations of the Dromagowlane/ Trafrask catchment would not negatively impact on any remaining Freshwater Pearl Mussel populations present in the catchment.

The Board considered and accepted this conclusion.”

737. The Sweetman Applicants argued that this passage – most notably the repeated phrase “highly unlikely” – overstated or exaggerated the conclusion of the O’Toole FwPM Report 2021 in a “leap worthy of a salmon”. Linguistically and as a matter of form the criticism is fair and the phrase is regrettable. However, the essential conclusion and the reason given for it are in substance the same as those expressed by Dr O’Toole. In my view to quash the impugned Aquaculture Licence on this account would flow from excessive formalism. I decline to do so.

738. I accept ALAB’s observation that the O’Toole site visit and O’Toole FwPM Report 2021 must be considered in the context of ALAB’s Determination. Immediately before its description of the O’Toole report as set out above, it listed information provided by the Marine Institute and IFI and the findings of the Technical Advisor Final Report1217 which it had considered. It concluded that:

“… notwithstanding the differing views, it was satisfied that risks to wild salmon from sea lice infestation is unlikely to be a significant factor influencing conservation status of salmon stocks in Bantry Bay. In reaching this conclusion, the Board had regard to the DAFM Monitoring Protocol No.3 for Offshore Finfish Farms - Sea Lice Monitoring and Control, with which the Applicant is required to comply as a standard Term and Condition of the licence and determined that this, along with the Pest Management Plan, is deemed to be sufficient and reduces any risk to a reasonable, non-significant level. The Board concluded there were no significant risks of sea-lice infestation of wild salmonids migrating from/to the Dromagowlane and Trafrask Rivers arising from the proposed activity at the Site. However, given the potential importance of the population of freshwater pearl mussel reported in the NPWS report submitted as part of its Oral Hearing submission …., the Board decided that this issue warranted a site visit …”

739. I have already expressed concern that the Aquaculture Licence is not expressly conditional on implementation of the Pest Management Plan or – for that matter – such amended version thereof as might to ALAB seem appropriate. That conditionality is certainly an assumption of ALAB’s determination but MOWI took the position at trial that the licence was not so conditional. However, this issue does not undermine the logic of ALAB’s being reassured by the O’Toole FwPM Report 2021 as it opines that the FwPM hosts are largely brown trout which, ALAB agrees, are unaffected by lice as, being non-migratory, they don’t come into contact with them.

1217 i.e. Dr Saunders’ report dated 8 December 2020.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

740. In short, I see no basis for a finding that ALAB ignored Dr Good’s evidence. On the contrary, it required that an sEIS address the issue precisely in light of his evidence and circulated that EIS for comment by all interested parties including the NPWS, IFI and Marine Institute. These in turn lead ALAB to commission Dr O’Toole’s site visit. These is no doubt that there was considerable and conflicting evidence on the issue of treat to the FwPM and equally no doubt that it was for ALAB, not the court, to decide upon what of that

evidence it should most truly. I reject the allegation of error on this account – whether it be, as is not entirely

clear, an allegation of ignoring relevant evidence or of irrationality.

741. There is also a pleadings despite as to an allegation that ALAB relied unduly on the evidence of Dr Jackson. Rather than tease out the pleading issue, I am content to record that the point has no legal merit.

LICE & FWPM – O’TOOLE REPORT – NON-CIRCULATION

742. IFI complain (IFI CG3) of what it calls ALAB’s failure to circulate the O’Toole FwPM Report 2021 for comment by it and other participants in the Appeal.

743. Though I have recited the sequence earlier, it is important in considering this ground to put the O’Toole FwPM Report 2021 in the context of the process which preceded it. The question of risk to FwPM was raised in appeals1218 in 2015. It was the subject of a s.47 request to MHI and the NPWS in 2016. Notably, it was an explicit agenda item as a “key issue”1219 of the oral hearing in 2017, and the subject of multiple documents presented to the oral hearing at which all interested had opportunity to make their views known and many did so. The resulting Oral Hearing Report makes clear the significant exploration of that issue at the oral hearing, including by Dr Good of NPWS. Dr Jackson of the Marine Institute specifically identified the brown trout as the “obvious host” for FwPM larvae. Any others interested had a like opportunity to address these issues. The Oral Hearing Report identified the issue as requiring further exploration in an sEIS of “The risk of sea-lice infestation of wild salmonids migrating from/to the Dromagowlane and Trafrask Rivers, and any resulting implications for local freshwater pearl mussel populations”. The resultant sEIS specifically addressed that issue. Importantly, in November 2018 that sEIS was circulated, inter alia to IFI, and published for comment by those interested.

744. The opportunity to respond to the published sEIS was yet another occasion on which interested parties and the public were explicitly afforded opportunity to be heard on the issue of indirect effect of sea lice, via salmonid mortality and consequent reduction in FwPM host populations in the Dromagowlane/

1218 Save Bantry Bay, John Brendan O’Keeffe. 1219 McIntyre Report on Oral hearing 8/11/17.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Trafrask river, on FwPM recruitment. Inter alia, the account in the sEIS of the Ross report of 2008 was made available and the interested parties and the public had been aware from the oral hearing of the view that brown trout were the primary FwPM larva hosts.

745. IFI replied by submission of 18 December 2018, inter alia articulating its view of the lice issue in terms of its effect on salmonid hosting of FwPM larvae. SWI and the Sweetman Applicants also commented to ALAB on the sEIS. Both canvassed the FwPM issue – the latter enclosing a CJEU judgment relating specifically to a dispute as to effects on the FwPM elsewhere.1220

746. In that context it must be observed that the objections that the views of the Marine Institute and Dr O’Toole were not circulated for comment lack force and the objection that views of Dr Evelyn Moorkens on these issues,1221 as expressed in affidavits sworn in the Sweetman proceedings, were not before ALAB when it made the Impugned Decision similarly lack force. Also, Dr O’Toole1222 has rejected Dr Moorkens’ criticisms of her FwPM report. In support of her view that “[t]he decline of FPM in recent years have been attributed to river water quality, largely as a consequence of discharges or contamination from domestic or agricultural sources, with increased levels of siltation and nutrient loading being of particular risk”, Dr O’Toole cites Dr Jervis Good of the NPWS in the oral hearing to the effect that, as Prof McIntyre records him, “the principal conservation issue is that of phosphate contamination of waters and siltation and that there exists little evidence to date that salmon production has impacted freshwater pearl mussels locally.”1223 Dr O’Toole adds that “phosphorous contamination and siltation arises principally from agriculture”.

747. The Final Saunders Report of 8 December 2020 addressed the possibility of risk to the FwPM in the Trafrask river system. It is important to say that this consideration derived in part from agitation of this issue by objectors1224 – some as late as 2018/2019. While he acknowledged that data on this issue was insufficient to allow “incontrovertible” conclusions, Dr Saunders considered any risk could be adequately managed and he found no substantive technical reasons to refuse the licence application.

748. ALAB in December 2020/January 2021 noted its duty to be “meticulous” on the FwPM and other issues and asked Dr O’Toole, as its technical advisor, to pay a site visit. Her visit of March 2021 and resultant O’Toole FwPM Report 2021 are described above. However, remembering that that the posited risk to the FwPM is the degradation of the population in the rivers of salmonid hosts of FwPM larvae, one may add that the O’Toole FwPM Report 2021 describes the context in which her site visit and resultant report arose. She says that “The Board has since accepted the Applicant's Pest Management Plan as being sufficient to

1220 Case C‑323/17 People Over Wind and Sweetman, Judgment of 12 April 2018 (EU:C:2018:244), §37. 1221 Affidavit of Dr Evelyn Moorkens sworn 24 September 2021. 1222 Affidavit of Dr Ciar O’Toole sworn 29 June 2022. 1223 Oral Hearing Report 8 November 2017.

1224 Save Bantry Bay 06/01/2019, Salmon Watch Ireland 05/01/2019. See for example Saunders Final Report Table at p115.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

mitigate risk of wild salmon sea lice infection, most recently at a Board meeting on the 12th January 2021. However, out of an abundance of caution, it was also decided to further investigate any possibility of a decline on salmonid populations in terms of knock on effects to FwPM in the Dromagoulane/Trafrask River catchment.”

749. As stated above, the site visit objective was to establish whether, due to difficulties in accessing the middle and upper reaches of the Dromagoulane/Trafrask River by migratory salmon and sea trout, it was more likely that those reaches were used for spawning by non-migratory trout, and inhabited by their juveniles. The site visit method was clearly observational only – and in at least appreciable part it was observation at a distance. No formal surveys (quantitative or other) were done. No samples or data were collected. Broadly described, its output was “an overall impression of conditions in the catchment and likely spawning areas from migratory salmonids” and was applied to the documentary evidence and academic papers already to hand. Indeed it was also informed by IFI’s own “semi-quantitative” electrofishing survey of May 2017. It cannot be suggested that this site visit and report introduced any new material or involved a survey of a kind which interested parties could not have readily replicated at any time in their many years involvement in the process. More narrowly described, its output was agreement with the view, previously expressed – and disputed by those who wished to dispute it – that the primary FwPM larvae hosts in the river were likely to be resident brown trout juveniles (unlikely to be affected by sea lice) rather than migratory salmonids (more likely to be affected by sea lice). It is important to observe that this issue had already been appreciably canvassed at the oral hearing in which the Appellants participated and all, including IFI, had had their chance to participate on that issue and, no doubt, to perform the same type of observational site visit as was later done by Dr O’Toole.

750. It must also be said that IFI at trial had difficulty articulating what, in substance, they had been prevented from saying to ALAB by reason of the non-circulation of the O’Toole FwPM Report 2021. Suggesting, as IFI does,1225 that had it seen the O’Toole FwPM Report 2021before ALAB’s determination and had it been afforded the opportunity to comment, it would have investigated further as Dr O’Toole’s conclusions were at variance with its own assessment amounts, given the issues had long-since been live and IFI had had ample and repeated opportunity to comment on them, and had electrofished the river at ALAB’s express request, to no more than IFI’s regret that it did not do in time the work which it now considers it ought to have done, and did only after ALAB’s Determination, by way of fuller electrofishing of the river system.

751. IFI cited a very brief report by Dr Gargan of IFI dated 24 September 2021, based on a further and more extensive electrofishing survey done in August and September. This was clearly a reaction to ALAB’s determination and, I infer, was probably done in contemplation of judicial review. It disputes Dr O’Toole’s view that migratory salmonids likely have limited access to the middle and upper reaches of the river system,

1225 Affidavit of Gregory Forde 23 September 2021 §32.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

asserts inconsistencies in the information as to FwPM locations and disagrees with Dr O’Toole’s conclusion that a decline (due to lice) of the migratory salmonid populations of the river system would not negatively impact FwPM populations.

752. Of this, it is impossible not to observe that

• IFI had noted as early as October 2016 the absence of electrofishing data and so, in its view, of specific data on salmonid densities in the river.

• The oral hearing opened in February 2017, at which time the identification of brown trout1226 juveniles as the primary host of FwPM larvae was live.

• Knowing that, IFI electrofished the river system in May 2017 specifically at ALAB’s request – i.e. for purposes of the Appeal – and reported thereon to ALAB in September 2017. It was for IFI to judge the adequacy for its purposes of its own survey effort at that time. It described the survey as a semiquantitative and inexhaustive snapshot but did not explain why a fuller survey was not done or express any anxiety to do a fuller survey. ALAB was entitled to assume that IFI, as an expert statutory authority, had done a survey adequate to the purpose of ALAB’s request.

• IFI also took the opportunity to make a written submission to the resumed oral hearing in September 2017 and addressed sea lice issues but did not express concern at inadequacy of data on salmonid FwPM larvae hosts in the river system or any anxiety to do a fuller electrofishing survey.

• The sEIS of 2018 cited that IFI Electrofishing report and IFI were circulated the sEIS for comment. Thereby IFI had, but did not take in its reply of December 2018, the opportunity to revisit any issue of inadequacy of their 2017 electrofishing survey.

• I have no reason to infer that the more complete survey done by IFI in September 2021 could not have been done in 2017 or otherwise in time to inform ALAB’s Impugned Decision.

753. I cannot but conclude that IFI had ample notice of, and opportunity to make submissions on, the issue of FwPM larvae hosting in the river system and for that purpose to conduct any degree of electrofishing or other surveys it thought appropriate, including the relatively simple observational type of survey done by Dr O’Toole and the fuller electrofishing survey which it did not do until after ALAB had made the Impugned Decision. Dr O’Toole’s site visit and the O’Toole FwPM Report 2021 accumulated no new data or other new material beyond the observational and those observations essentially sought to check and verify a view to which ALAB had already properly arrived. Dr O’Toole is an in-house, employed technical advisor to ALAB. Her visit was, to my mind analogous to a site visit by a planning inspector with a view to a report to An Bord Pleanála. While it is not impossible that, in particular cases and for particular reasons, such reports should be circulated in draft to participants in planning processes, that is very far from the normal course and non-circulation of such a draft is very rarely controversial.

1226 i.e. non-migratory trout.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

754. There is no reality here to IFI’s complaint, in effect, of breach of the rule audi alteram partem. They were in no way unjustly treated or deprived of their rights of participation in the process. It simply cannot be said that IFI lacked or had not availed of repeated opportunities to make its views on these issues known to

ALAB. “Reasonable fairness in all of the circumstances is what is required.” – Dellway.1227 The inevitable fact

is that circulation of data, reports and submissions must end at some point – a point to be determined on the application of broad principles of commonsense and fair play to a given set of circumstances – Haverty.1228 Short of the decision-maker’s decision, someone must have the last word. The “endless pingpong” must send somewhere – Evans.1229 The applicant in judicial review must be able to show that he has been deprived of actually making comment on a particular issue. In my view Redrock,1230 West Wood,1231 Southwood1232 and Wexele1233 tell against IFI here. In Wexele, Charleton J said: “The Board is not obliged to bring every fresh submission to the attention of a party to the appeal and to ask for further observations. The first principal applicable is that of utility.”

755. Posing the question posed in Wexele – whether an injustice has been perpetrated through a new and objectively significant important point, of which the other party had no notice, being brought into the equation – the answer is no. IFI was unable at trial to explain why, in its over 9 years of opposition to the Shot Head salmon farm, and despite participating in the Oral Hearing and making repeated submissions to ALAB, it had not done and submitted to ALAB the results of such an electrofishing exercise as was done after ALAB had made its Impugned Decision. It is difficult to see the IFI report of September 2021 and Dr Gargan’s

affidavit submitting it to the court as other than as an expression of l’esprit de l’escalier as to opportunities

of which it had not availed in time and in the licencing process. L’esprit de l’escalier cannot be a basis for quashing an Impugned Decision.

756. Also, and while I would not rule out the possibility of exception to the rule in a particular case, I agree with ALAB that Dr O’Toole can be analogised to an inspector of An Board Pleanála and that M&F Quirke1234 and Mulhaire1235 are authority that, at least ordinarily, fair procedures do not require that such reports be circulated for comment. For reasons which will be apparent from the foregoing, I do not see this as an exceptional case.

757. I therefore respectfully reject the complaint of unfairness of procedures by not circulating the O’Toole FwPM report 2021 for comment.

1227 Dellway Investments Ltd v National Asset Management Agency [2011] 4 I.R. 1. 1228 State (Haverty) v An Bord Pleanála [1987] I.R. 485. 1229 Evans v An Bord Pleanála, Unrep Kearns J, 7 November 2003, 2004 WJSC-HC 4037, [2003] 11 JIC 0702. 1230 Redrock Developments Ltd v An Bord Pleanála [2019] IEHC 792 (Faherty J, 21 October 2019), §167. 1231 West Wood v An Bord Pleanála [2010] IEHC 16 §44. 1232 Southwood Park Residents Association v An Bord Pleanála & ors [2019] IEHC 504. 1233 Wexele v An Bord Pleanála (No. 1) [2010] IEHC 255. 1234 M & F Quirke and Sons v O'Connor [2009] IEHC 426. 1235 Mulhaire v An Bord Pleanála [2007] I.E.H.C. 478

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

SALMON ESCAPES – EIA & AA, the 2014 ESCAPE, CAGES & SEALS

Escapes – Introduction

758. The questions of the risk of the occurrence of escapes of farmed salmon from the proposed Salmon Farm and of the environmental effects thereof looms large in these proceedings. No doubt escapes can occur for a variety of reasons but amongst them are storm damage and seal predation – both of which risks can be particular to the location of the farm and are at issue in this case. Storms can damage salmon cages in various ways – especially if the farm is in shallow waters or close to land. Predatory seals tear holes in cage netting to get at farmed salmon, through which holes farmed salmon can escape.

759. It is not in any real dispute that if farmed salmon do escape in appreciable numbers from a salmon farm, they may, in greater or lesser, but in any event in concerning, degree,

• transmit sea lice more efficiently to wild salmon than would otherwise occur, at risk of increased wild salmon morbidity and mortality.

• interbreed with wild salmon, so altering the genetic makeup of the offspring – which are less capable of survival in the wild, thereby diminishing the genetic quality of the wild salmon population.

• compete with wild salmon for resources, to the disadvantage of the latter.

Dr Saunders advised1236 that “Escaped farmed salmon can impact on wild salmonid populations via resource competition in the riverine environment and through interbreeding and subsequent genetic dilution of native traits, resulting in reduced fitness. Escapee salmon may also transfer diseases and parasites,1237 although evidence for this is limited.”

760. These risks are in substance reflected in detail in various documents I consider below: including the ICES1238 and NASCO1239 reports of 2016 (including the international objective of 100% containment of farmed salmon), the s.47 Notice to the DAFM in 2017, the November 2017 report of the chairperson of the oral hearing, Condition 4 of the Aquaculture Licence as to the NASCO Containment Guidelines 2001,1240 and, if implicitly and generally, in ss. 7 and 77 of the 1997 Act.

1236 Final Report 8 December 2020 §6.5.5. 1237 Presumably including sea lice.

1238 International Council for the Exploration of the Sea. It was established in 1902 and since 1968 operates pursuant to the Convention for the

International Council for the Exploration of the Sea 1964. Ireland is a member.

1239 North Atlantic Salmon Conservation Organisation. It was established in 1984 by the Convention for the Conservation of Salmon in the North Atlantic

Ocean 1982. The EU approved that convention by Council Decision 82/886/ EEC and it is registered with the UN as an international treaty. It enables 6

Governments & the EU to co-operate to conserve wild Atlantic salmon. Ireland participates via its EU membership. The IFI attends for Ireland. This

information is taken from “Progress and challenges in achieving NASCO’s international goals in Ireland Michael Millane, Paddy Gargan and Cathal

Gallagher, Inland Fisheries Ireland” annexed to the Report of a Theme-based Special Session of the Council of NASCO, 8 June 2016 – addressing impacts

of salmon farming on wild Atlantic salmon. 1240 CNL(01)53 Guidelines on Containment of Farm Salmon.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

761. By way of setting the context both for his efforts thereafter and for the consideration of the issue of escapes in this judgment, I note that Dr Saunders “read in” to the Appeals from April 2016 and as early as by e-mails of 1 & 10 June 2016 advised ALAB as follows:

• “You will doubtless not be surprised that the most lengthy and well-researched appeals (and therefore the issues that require very careful consideration) are those dealing with the potential for impacts on the wild salmon and trout population through sea-lice transfer and farm escapes.”

• He referred to “the major implications in respect of the appeals submissions” overwhelming emphasis on concerns relating to sea lice and escape impacts. This continues to be a subject of considerable debate and I am attempting to establish a credible scientifically-supported response …. to the appeals”

So, from the start, escapes were both a major and a controversial issue in considering the Appeal.

Escapes – ALAB’s Determination

762. While I emphasise that it must be considered in context, it is useful to record here ALAB’s determination on the risk of farmed salmon escape.

“The risk of fish escapes from the proposed salmon farm at the Site was discussed in the 2011 EIS and was considered also at the Oral Hearing and in the Technical Advisor Final Report. The advice of the Board's Technical Advisor in the Technical Advisor Final Report was that the proposed activity at the Site presents a negligible risk for the transfer of fish diseases to wild stocks via escapes into Bantry Bay. The Board noted that certain Appellants had raised concerns over large episodic events associated with holes in nets caused by predators, or as a result of storm events. The Board noted the specifications of the farm cages and their ability to withstand the expected conditions were not supplied but also noted that approval of the cages specification will fall within the jurisdiction of and require approval from DAFM. The Board also noted that a Term and Condition of the licence will include compliance with the most up to date guidelines on fish containment developed by the North Atlantic Salmon Farming Industry and the NASCO Liaison Group.

Notwithstanding, the Board determined that the licence for the Site should include, in Schedule 5 thereof, an additional condition requiring the Applicant to comply with the standards laid out in the Structural Design Protocol, as revised from time to time.”1241

1241 ALAB’s Determination §6.1.3.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Escapes – Pleadings

763. The issue of escapes is pleaded in terms of,

• bias – allegedly disclosed in ALAB’s withdrawal of a s.47 notice to DAFM seeking documents relating to a major farmed salmon escape in February 2014 from the Murphy’s Irish Seafoods salmon farm at Gearhies on the other side of Bantry Bay from Shot Head.

• inadequacy of EIA and AA – primarily as to assessment of the structural reliability of containment of the salmon in the Salmon Farm and as to the risk of escape by reason of damage to the cages by predatory seals.

• an allegation that the acoustic means – seal scarers – likely to be used to deter seals have not been considered in AA.

764. SWI plead:

• that the Aquaculture Licence is invalid as ALAB failed to assess, in both AA1242 and EIA,1243 the structural adequacy of the salmon cages to prevent escapes.1244

• the content of Dr Saunders’ Final Report as to escapes, of ALAB’s determination as to escapes and the content of the Protocol for Structural Design of Marine Finfish Farms. 1245

• that Dr Saunders was unaware of the Protocol and ALAB did not analyse its adequacy for EIA Purposes.

• ALAB’s EIA failed to identify, describe and assess comprehensively the likely significant impacts of storms on cages leading to potential escapes and inter-breeding of farmed salmon with wild salmon and resulting genetic dilution.1246

MOWI’s written submissions took a pleading objection to this case. In my view SWI’s case was adequately clear from the pleadings.

765. IFI plead that:

1242 Contrary to Article 6(3) of the Habitats Directive, as implemented by Regulation 42(8) of the European Communities (Birds and Natural Habitats)

Regulations 2011 as amended).

1243 Articles 2, 3 and 8a of the EIA Directive or their precursors, as implemented by Regulation 3 of the Aquaculture (Licence Application) Regulations 1998

(S.I. No. 236 of 1998) as amended by the Aquaculture (Licence Application) (Amendment) Regulations 2010 (S.I. No. 280 of 2010), Aquaculture (Licence

Application) (Amendment) (No. 2) Regulations 2010 (S.I. No. 369 of 2010) and Aquaculture (Licence Application) (Amendment) Regulations 2012 (S.I. No.

301 of 2012) and European Union (Environmental Impact Assessment) (Aquaculture) Regulations 2012 (S.I. No. 410 of 2012) and / or as required by

Article 3(1) of the Aquaculture Appeals (Environmental Impact Assessment) Regulations 2012 (S.I. No. 468 of 2012). 1244 §E1.7. 1245 §E2.7.

1246 SWI pleads that accordingly, the EIA failed to meet the standard set in Recital 2, Articles 2, 3 and 8a of the EIA Directive ….. as implemented by

Regulation 3 of the Aquaculture (Licence Application) Regulations 1998 (S.I. No. 236 of 1998) as amended by the Aquaculture (Licence Application)

(Amendment) Regulations 2010 (S.I. No. 280 of 2010), Aquaculture (Licence Application) (Amendment) (No. 2) Regulations 2010 (S.I. No. 369 of 2010)

and Aquaculture (Licence Application) (Amendment) Regulations 2012 (S.I. No. 301 of 2012) and European Union (Environmental Impact Assessment)

(Aquaculture) Regulations 2012 (S.I. No. 410 of 2012) and / or as required by Article 3(1) of the Aquaculture Appeals (Environmental Impact Assessment)

Regulations 2012 (S.I. No. 468 of 2012) in particular in relation to impact on biodiversity or fauna SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• it had submitted to ALAB that the potential impact of a large escape of farmed fish (such as occurred in Bantry Bay in 2014) and preventative measures to prevent such escape, were not adequately assessed in the EIS.

• ALAB did not conduct any or a proper EIA.

• ALAB failed to properly consider the threats posed by escapes of farmed salmon.

• ALAB in this regard seemingly gave the benefit of the doubt to MOWI, in a manner which is irrational, illogical and unreasonable having regard to normal best practice in environmental protection and fails to take account of relevant considerations.

766. The Sweetman Applicants plead1247 that:

• The Aquaculture Licence is invalid as in breach of s.47 of the 1997 Act and of the requirements of fair procedures for the following reasons. o ALAB formed the opinion that documents relating to the escape of 230,000 fish from a salmon farm in Bantry Bay in 2014 were necessary to enable it to determine the appeal. o ALAB was therefore obliged to and did serve on DAFM a notice under s.47 of the 1997 Act requiring DAFM to supply those documents. o ALAB refused the DAFM offer to supply the report on a confidential basis as all documents are made available on its website. o ALAB withdrew the s.47 notice before the documents were supplied. o ALAB had no jurisdiction to withdraw the s.47 notice. o The decision to withdraw the s.47 notice was unreasonable and unreasoned.

• ALAB failed to consider the impact of accidental escapes of farmed salmon in breach of the obligation to screen for AA.

• ALAB failed in breach of Arts. 3(1) & (2) of the EIA Directive1248 to acquire and consider the necessary information as to the o impacts of a major accident including a fish escape. o structural integrity of the proposed structures.

767. The Sweetman Applicants also plead1249 that:

• Ireland’s largest recorded escape of farmed salmon – of 230,000 fish – occurred at a fish farm in Bantry Bay during a storm in 2014.

• Media reports of the escape of 50,000 salmon from a MOWI site near Campbeltown, Scotland during Storm Ellen in August 2020.

• The DFAM report into the cause and effects of this incident has not been published.

1247 §E.1.4. §E.2.21, 29, 35, 74, 75.

1248 The 2011 EIA Directive Art 3(1) requires that EIA “shall identify, describe and assess in an appropriate manner, in the light of each individual case, the

direct and indirect significant effects of a project” inter alia on fauna and flora. As amended in 2014, it specifies “biodiversity, with particular attention to

species and habitats protected under the Habitats Directive and the Wild Birds Directive”. 1249 §E.2.47, 79-82.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• The failure to incorporate that report into to ALAB’s environmental assessments was controversial as to the Storm Head site which is subject to the same storm and wave risks.

• At the oral hearing in September 2017, o controversy had been raised as to the withdrawal of the s.47 notice in circumstances in which the Commissioner for Environmental Information in January 2017 had directed release to the public of certain reports within the scope of the s.47 notice. o DAFM had asserted that its report was incomplete and had no standing, that DAFM had told ALAB it would release it and didn’t only because ALAB withdrew its s.47 notice.

• The DAFM/ALAB correspondence was not made available to the public with the other s.47 files on the ALAB website.

768. ALAB pleads in the Sweetman case that:

• Its withdrawal of its s.47 notice to DAFM was reasonable, reasoned and within ALAB’s jurisdiction and it recites the DAFM/ALAB correspondence in that regard.

• The Sweetman Applicants lack standing on this issue. (As will appear, I reject this plea). ALAB does not deny the plea that the DAFM/ALAB correspondence was not made available to the public with the other s.47 files on the ALAB website. Nor does it explain why that correspondence was not made available to the public.

769. ALAB raises a pleading issue in the Sweetman case that it was only in written submissions that the Applicants cited the report of the chairperson of the oral hearing that “the Board should make every effort to consider the potential impacts of large-scale farmed salmon escapes” to the effect that the DAFM documents remained necessary to the determination of the appeal and observed that “It is not clear why the information (either as a draft or final report) was not requested again from the Department during the period of almost 4 years between the withdrawal of the s.47 request and the making of the determination.” ALAB’s point is that this was the first intimation of an unpleaded case that after the oral hearing and the report thereof, the s.47 notice should have been resubmitted to the DAFM.

770. The Sweetman Applicants reply that,

• if the decision to withdraw the request was unreasonable and unreasoned it follows that the request should have been re-submitted.

• if the s.47 request was withdrawn only because the document sought was then in draft form, it is obvious that the document is still necessary and further attempts should be made to ascertain when the report has been finalised. They call in aid their plea1250 of o “the discussion on this in the Oral Hearing” o the exhibited oral hearing transcript of 20 September 2017 pages 47-58.

1250 Grounds §E.2.21.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

771. These two replies seem to me distinct. If the withdrawal was reasonable and reasoned, it follows that Sweetman Applicants’ first reply fails. The second reply in effect asserts that, having withdrawn the s.47 Request having been told by DAFM that the report was in draft, it is obvious that where the documents in question were in fact in the public domain and at least some were not in draft but finalised, the necessity of considering them reasserted itself and the need to renew the s.47 Notice is obvious. In my view this pleading

dispute is a fine issue. I am conscious of the insistence on clarity of pleadings in judicial review – Casey.1251 I

am also conscious of the need to take a fair and reasonable, rather than an excessively narrow view, of the content of pleadings, justice being the key criterion and the question being whether the pleadings are “acceptably clear”– St Audoen’s1252 and Sweetman v An Bord Pleanála & Bord na Mona.1253 In the latter case Humphreys J said “I don’t accept the implicit point that if a general claim is made with particulars provided in later paragraphs, and the particulars don’t flesh out all aspects, then the non-fleshed-out aspects fall away. That is to read pleadings in the narrowest possible sense and of course massively incentivises

prolixity.” He also said “the cry of “particularise that” can echo indefinitely. That process has to stop when the

point being made is acceptably clear.”

772. As stated, the Sweetman Applicants pleaded pages 47-58 of the exhibited oral hearing transcript of

20 September 2017. That content includes:

• ventilation of a complaint that, when by letter dated 28 July 2017 DAFM asserted that “the release into the public domain of this report, which currently remains in draft form, would not be in the public interest”, at least some of the DAFM documents encompassed by the s.47 Notice were, o finalised and not still in draft. o already in the public domain since March 2017 on foot of an order of the Commissioner for Environmental Information made in January 2017 and Friends of The Irish Environment had copies.

• Mr Hodnett of DAFM confirmed to the oral hearing that he had released those reports into the public domain.

• The chairman’s guarantee “that ALAB will consider what we do with this report. It is very helpful …” as “ALAB have made it clear that they felt this might contain useful evidence, that was the reason for the Section 47 request in the first place” and had “already determined that we felt that this report may contain information that would be useful to our determination” and would “have to consider we can rely on the

decision of the commissioner for environmental information, to the effect that elements of this report are not draft or not draft in form, then we will reconsider whether we should take account of these reports.”

• Mr Sweetman’s interventions in these arguments.

773. The point here is that the Sweetman Applicants not merely adduced this content in evidence – they pleaded it by reference and it clearly engages not merely the issue of ALAB’s withdrawal of the s.47 request but the Chairman’s guarantee that ALAB would revisit the issue now that it knew the reports were in the

1251 Casey v Minister for Housing [2021] IESC 42 (Supreme Court, Baker J, 16 July 2021).

1252 The Board of Management of St. Audoen’s National School v An Bord Pleanála [2021] IEHC 453 (Simons J, 15 July 2021).

1253 [2021] IEHC 390. The key criterion is, needless to say, justice, which includes the concept of giving acceptable notice of the point being made.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

public domain. As a result and though he might have phrased it more pointedly, the Chairman in his report on the oral hearing recommended that “the Board should make every effort to consider the potential impacts of large-scale farmed salmon escapes”. Putting the pleaded transcript in the context of the plea that ALAB failed in breach of the EIA Directive to acquire and consider the necessary information as to the impacts of a major accident including a fish escape and the structural integrity of the proposed structures, I consider that on a fair and reasonable reading of their pleadings and by reference to the key criterion of justice, the Sweetman Applicants are entitled to argue that the s.47 notice to DAFM should have been reinstated after the Oral Hearing or that ALAB had to at least consider reinstating it or otherwise having regard to the documents in the public domain. That is especially so given the guarantees given by the Chairman at the oral hearing, which guarantees were pleaded by reference.

Timing & Scope of EIA & AA – Basic Principles

774. It is useful to preface the following description of the factual and regulatory matrix with the fundamental rule that both EIA and AA must precede the grant of development consent. Recital 7* of the EIA Directive states that “Development consent for public and private projects which are likely to have significant effects on the environment should be granted only after an assessment of the likely significant environmental effects of those projects has been carried out.” Article 2(1) of the EIA Directive requires member States to “ensure that, before development consent is given, projects likely to have significant effects on the environment” are subjected to EIA. Many authorities1254 reflect this principle and it is not weakened by the fact that Boland conditions are acceptable in EIA and AA cases.1255

775. The second basic principle is that EIA must be comprehensive. That is to say, it must identify and consider all significant environmental risks and, of those risks, it must perform “as complete an assessment as possible”.1256

Escapes – the Matrix

The Risk of and posed by Escapes – Basic Information and its Statutory Recognition

776. There is in reality, no dispute but that Dr Saunders is correct in stating,1257 as to risks posed to wild salmon by escaped farmed salmon, that: “There is a substantial body of scientific literature and associated

1254 Including Case C-392/96 Commission v Ireland, Opinion of Pergola AG of 17 December 1998; Case C-215/06 Commission v Ireland; Case C-535/18 IL

v Land Nordrhein-Westfalen, Opinion of Hogan AG of 12 November 2019, §41; Boylan v Limerick City and County Council [2020] IEHC 666 §§55 & 56.

Case C-463/20 Namur-Est v Région Wallonne, §§46, 47 & 58. 1255 See further below.

1256 E.g. Case C-50/09 Commission v Ireland, Case C-404/09 Commission v Spain (AG Kokott), Case C-463/20 Namur-Est v Région Wallonne §§58 & 60.

1257 Final report 8 December 2020 §9.2.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

data on this subject, with conclusions ranging between neutral or negative effects for wild populations.” He advises that “Farmed escapee fish can compete with wild salmon for resources, may breed with wild counterparts resulting in reduced genetic fitness, and constitute a disease and parasite transfer risk.”

777. Dr Saunders observes that “The prevention of escapes and approaches to reduce impacts are of fundamental importance to both the interests of the aquaculture industry and the conservation of declining wild salmonid stocks.” Though it does not affect my decision, it bears noting that this observation illustrates the undoubted truth that, on this issue of escapes at least and albeit from different perspectives, [respectively commercial (as to the loss of their stock in trade) and environmental (as to salmon conservation)], the protagonists’ interests are precisely aligned in seeking to prevent escapes. Though, of course, they will weigh the acceptability of their respective risks and the proper means of addressing them, differently.

778. Notably, in 2014 the Department of the Environment1258 had advised DAFM that in deciding the Aquaculture Licence application it should:

“take cognisance of the following:

STRUCTURAL INTEGRITY

It is noted that the site as proposed along the coast between Shot Head and Mehal Head is in an unsheltered and reasonably exposed location in Bantry Bay. The Applicants would need to demonstrate to DAFM's satisfaction the structural integrity of the proposed anchor/mooring arrangements for the cages especially in relation to extreme wave/current and tidal situations linked to extreme weather conditions. They would need to demonstrate that the proposed structures are fit for purpose under all these extreme conditions. Cage and rope/net fatigue and wear and tear over time would also need to be factored into this assessment and the Applicants would need to satisfy DAFM in relation to their contingency plans for dealing with defective or loose structures should they arise.”

779. The significance of escapes is, in broad terms, illustrated in the summary of an exhibited report of an expert group of the ICES1259 of March 20161260 in response to questions posed by NASCO. Dr Gargan of IFI and Dr Jackson of the Marine Institute attended the ICES workshop. To set the context for the ICES findings I should cite Dr Saunders to the effect that:

1258 Observations 19 June 2014. 1259 International Council for the Exploration of the Sea – constituted by international convention of 1964.

1260 ICES Workshop Report 3/16 on possible effects of salmonid aquaculture on wild Atlantic salmon focusing on effects of sea lice and genetic interactions.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“The homing instinct of salmon to breed in geographically and ecologically distinct rivers drives adaptation to a specific aquatic locality. This promotes genetic isolation of wild Atlantic salmon populations with little genetic interaction between populations from different catchments. Farmed salmon production is largely based on a small number of breeding strains selected for traits advantageous to salmon production. This contrasts with the traits acquired in wild salmon, driven by natural selection in a particular river. Farmed strains are therefore genetically distinct from wild populations. Differences between wild and farmed salmon due to domestication and trait selection are likely to be exhibited in respect of growth rate, body size, survival, delayed maturity, stress tolerance, temperature tolerance, disease resistance, flesh quality, and egg production.”1261

780. The ICES findings can be fairly summarised as follows:

• Large numbers of salmon escape from fish farms.

• Escapees are observed in rivers in all regions where farming occurs, although the numbers of escapees vary both spatially and temporally.

• Genetic studies have demonstrated widespread introgression1262 of farmed salmon in a large number of Norwegian populations – the full extent of introgression remained unknown.

• Farmed salmon and their offspring display significant genetic differences to wild salmon in a wide range of fitness-related traits – they display lower fitness than their wild counterparts.

• Where farmed salmon interbreed with wild salmon, it is likely that wild salmon populations will display maladaptive changes in life-history traits.

• Introgression can be expected to lead, long-term, to fewer salmon and more fragile stocks.

• Juvenile escapees and the offspring of farmed salmon compete with wild salmon for territory and food.

Their presence in the natural habitat will reduce the total production of wild salmon.

• Such effects have to be seen in the context of other challenges to wild populations such as disease, sea lice infection, over exploitation, habitat destruction and poor water quality. So, wild populations are more likely to be sensitive to the potential negative effects of introgression and loss of fitness.

• The evidence clearly indicates impacts of escapees / genetic introgression on wild salmon. A substantial reduction of escaped farmed salmon in the wild, or sterilization of farmed salmon, would be required to minimize such effects.

781. Three months later, Dr Gargan of IFI, representing the EU and as a member of NASCO’s Steering Committee, was co-editor of a NASCO report dated June 20161263 as to protecting wild Atlantic salmon from impacts of salmon farming, promoting sustainable salmon farming practices, and goals for sea lice and containment. Notably, it identified best practice and an “agreed international goal” that “100% of farmed

fish are retained in all production facilities.” – i.e. “100% containment”. NASCO noted “with great concern”

1261 Saunders Final Report 8 December 2020 p68. 1262 Roughly, the interbreeding of farmed and wild salmon.

1263 Addressing impacts of salmon farming on wild Atlantic salmon: Challenges to, and developments supporting, achievement of NASCO’s international

goals. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

the ICES report of March 2016 as confirming substantial and growing evidence that aquaculture activities can affect wild Atlantic salmon, through the impacts of sea lice as well as farm escapees and widespread introgression of farmed salmon genes into wild salmon populations in Norway, and the detection of introgression in other countries. The conclusions of the NASCO steering committee recorded that the NASCO Convention requires that, in exercising its functions, NASCO takes into account the best scientific information available to it. Inter alia, it notes that “despite the measures introduced, ICES still advises that very large numbers of domesticated salmon escape from fish farms each year” and, as to introgression of farmed salmon into wild salmon stocks,

“… the consequences of these genetic changes in wild salmon populations are likely to be depression of fitness, decreased overall productivity, erosion of genetic diversity and decreased resilience.

Repeated invasions of farmed salmon in a wild population may cause the fitness of the native population to seriously decline and potentially enter an ‘extinction-vortex’ in extreme cases.

Preliminary analyses of non-introgressed and introgressed adult wild salmon from more than 50 populations in Norway suggest that ecological and life-history changes are widespread in Atlantic salmon populations where there has been introgression. This is a very worrying development.”

It seems to me fair to describe the tone of the NASCO report as one of alarm.

782. I do not adopt these ICES and NASCO reports and findings as necessarily correct in detail or as necessarily directly applicable to Irish conditions – much less to Bantry Bay or the proposed Shot Head salmon farm. They do not seem to constitute peer-reviewed academic literature – but they do seem to reflect considerable expert consensus arrived at by processes having standing in international law. They are not legally binding on ALAB – neither can they reasonably be ignored. Nor do I suggest that the account above is a full, or even a balanced account of those ICES and NASCO reports, which are nuanced and technical and display an appreciation of the requirements of the salmon farming sector. Still less do I suggest that they required refusal of ALAB’s licence application. I also note that the ICES and NASCO reports were made some years ago and further corrective measures may have supervened. However this account does fairly reflect general concerns expressed – and, indeed, the strength of those concerns as emanating from expert bodies having international law standing. Also, the Sweetman Applicants are not contradicted as to the alleged escape of 50,000 salmon from a MOWI site near Campbeltown, Scotland in Storm Ellen as late as August 2020.1264 The significance of this observation seems to me to be not so much that it was a MOWI site but that escapes in large numbers can still occur – have recently occurred – in regulated salmon farms.

783. I recite this content of those reports primarily as recording a respectable and authoritative scientific view of the general importance of, and risks posed by, escapes from salmon farms and the international goal of 100% containment of farmed salmon. They also serve to identify the risk of escape as necessarily to be considered in the “comprehensive assessment” required in EIA. Dr Saunders says as much:

1264 Affidavit of Noel Carr – sworn 27 September 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“A major escape event from a farm of this size would present a considerable risk to wild salmon population. Unlike ‘trickle’ escapes, the risk and impact of any large episodic escape is of considerable importance in the assessment of potential environmental impacts.”1265

784. It is convenient to observe here that ALAB’s pleaded reliance on Dr Saunders’ view that the “proposed activity at the Site represents a negligible risk for the transfer of fish diseases to wild stocks via escapes into Bantry Bay” must be read in context – reading his report as a whole, including the excerpt set out above. As will appear, this observation addresses disease risk only and not genetic risk – the considerable weight of which is not disputed. Dr Saunders’ view also says that the effect of lice on wild salmonids should be considered alongside the potential impacts from the 2014 escape “because of the known relationship between escapee fish and the augmentation or enhancement of natural infections in wild

stocks”. 1266 So it is clear that Dr Saunders’ description of negligible risk is clearly predicated, not on a view

that escapes, if they occur, will be environmentally insignificant, but on a view that escapes will not occur.

Yet, as will be seen, he asserts his own inability to assess the risk – he is “unable to evaluate the risk of fish

escapes from the Shot Head site”.

785. It is clear that these 2016 ICES and NASCO reports reflect long-standing concerns and those concerns are reflected in, inter alia, NASCO “Guidelines on Containment of Farm Salmon” of 2001 and NASCO’s “Williamsburg Resolution” 20031267 and its 2009 “Guidance on Best Management Practices to address impacts of sea lice and escaped farmed salmon on wild salmon stocks”.1268 As they are clearly of long standing, it is not anachronistic to observe that these concerns are broadly reflected in the earlier 1997 Act.

786. S.7 of the 1997 Act empowers ALAB to attach licence conditions specifically requiring measures for preventing escapes of fish from salmon farms, and arrangements for the reporting of such escapes. S.77 of the 1997 Act empowers IFI to take such action as it considers necessary to recapture stock which has escaped from a salmon farm and the Minister may authorise any person to take specified action to recapture such stock. I am satisfied that these statutory provisions are explicable only by the view that, and there is no dispute inter partes but that, escapes of farmed salmon (at least large scale escapes) are potentially highly detrimental to wild salmon and to be prevented. And if they occur they are to be reversed if possible.

787. I need not and do not decide any controversy as to the foregoing general views of the risks of the occurrence of, and risks of the effects of, escapes. I need to and do make findings only that the materials before me are clear that the risks of the occurrence of escape and the risks consequent upon escape could

1265 Saunders Final Report p69.

1266 Technical Advisor’s Final Report 8 December 2020 §9.1 Increased threat to wild salmon and sea trout from sea lice, p64.

1267 “Resolution by the Parties to the Convention for the Conservation of Salmon in the North Atlantic Ocean to Minimise Impacts from Aquaculture,

Introductions and Transfers, and Transgenics on the Wild Salmon Stocks” cited variously in the 2016 NASCO report and identified at p107 thereof.

1268 identified at p107 of the 2016 NASCO report.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

not be regarded as any less than “significant” as significance is understood in EIA. Accordingly they required consideration in EIA and, by necessary implication, the comprehensive assessment necessary in EIA. In short, and as a matter of law, the EIA had to comprehensively address the issue of escapes. Dr Saunders and Professor McIntyre had advised accordingly – the latter in his report on the oral hearing, to the effect that ALAB should “make every effort to consider the potential impacts of large-scale farmed salmon escapes.”

788. It seems to me useful to emphasise that, as with any risk analysis, an assessment and evaluation of the acceptability of the risks associated with farmed salmon escapes require analysis of two distinct aspects of risk: the risk of the occurrence of escape; the risk of the occurrence and severity environmental effects if escapes do occur. After mitigation of each, the net evaluation of the acceptability of the residual risk is based on the product of the two. Accordingly, it is necessary to consider in EIA of the escape issue:

• The assessment and mitigation of the risk of occurrence of escapes.

• The assessment and mitigation of the risks to wild salmon once an escape has occurred.

789. Given their espousal of a 100% containment standard, it is clear that ICES and NASCO at least consider that any such acceptable risk should be very low indeed. It seems inevitable to infer, at least in general terms, that the international adoption of a 100% containment objective implies that even such a demanding objective is less demanding than trying to mitigate the effect of escapes. Dr Saunders1269 clearly expresses that view and no-one has impugned it: “In reality, any attempt to recover escapee fish is likely to result in a very low level of success” and “recoverable numbers may be very low in comparison to total escape numbers”. He cites academic authority that “less than 3% of escaped salmon have been recaptured through organised fishing after large escape episodes.” The appropriate analogy is clearly with the genie and the bottle. While the views of ICES and NASCO do not bind ALAB as to the acceptability of risk of escape, they at least demonstrate and emphasise the necessity of comprehensive assessment of the issue in EIA.

790. Given the analogy with the genie and the bottle, and as to both storm risk and seal risk, it is unsurprising that no-one disputes Dr Saunders opinion that “The most important mitigation measure against fish escapes is the installation of the correctly specified equipment for a given site.”1270

Escapes – The Aquaculture Licence as to Cage Specification

791. ALAB’s Impugned Aquaculture licence permits cultivation of farmed salmon “in accordance with the plans and drawing(s) in Schedule 2 attached as approved of by the Minister”. However while Schedule 2

1269 Saunders Final Report 8 December 2020 §9.2 p71. 1270 Saunders Final Report §9.2 p72.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

includes the words “Schedule 2 contains: the approved plans and drawing(s)”, in fact Schedule 2 contains no such plans and drawings. What Schedule 2 does contain is a “Note” which reads:

“The Licensee will adhere to the standards set out in the Department of Agriculture, Food and Marine's Protocol for Structural Design of Marine Finfish Farms, 2016 and the Floating Facilities shall be approved by the Department of Agriculture Food and the Marine.”

792. Schedule 4 of the Licence includes the following

“The Licensee will adhere to the standards set out in the DAFM Protocol for Structural Design of Marine Finfish Farms, 2016 ("Structural Design Protocol") and the Floating Facilities shall be approved by the Department of Agriculture Food and the Marine.

Before deployment of any farm components at the Site, the Applicant shall obtain the prior approval of the Minister to the Initial layout of the cages on the Site and such plan shall be included in Schedule 2 to the Licence.”

Escapes – History, the 2014 Escape, the Minister’s 2015 EIA as to the 2014 Escape

793. Ireland, per IFI, informed NASCO, as recorded in its 2016 report and by Dr Saunders in his Final Report, that official statistics indicate that approximately 415,000 salmon were reported to have escaped from salmon farms in 1996 - 2004. Since 2009, six escapes had been reported – mostly due to storm damage. That seems at very least consistent with the identification of the risk of escape as a necessary consideration in EIA.

794. Dr Saunders’ Final Report1271 refers to the DAFM EIA in June 2015 as recording that:

“Given the small size of the salmon stocks in Bantry Bay rivers, and other areas along the possible migration or dispersal route of escaped farmed fish, mitigation of potential interactions with escaped farmed fish is essential”. 1272

That EIA also states that,

“An important consideration in relation to impact of escapees is the size of the wild salmon stocks in adjacent rivers. Escaped farmed salmon have been shown to alter the genetic make-up of wild populations.”

1271 8 December 2020 p69. 1272 EIA 2015 §19.1.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“There have been no documented cases in Ireland of negative population impacts leading specifically to loss of wild stock integrity and productivity. However, there is documented evidence from a number of Irish rivers of escapees entering freshwater, spawning with wild salmon and producing viable offspring and the persistence of genetic changes over time in one of the stocks studied. These genetic changes to wild stocks are likely to have a negative effect on a wild stock. Indications from literature suggest that small stocks and stocks at low densities are more vulnerable to these negative genetic impacts.”

Presumably this last sentence calls up the EIA’s reference to “the small size of the salmon stocks in Bantry Bay rivers” as being more vulnerable to adverse effects of large, farmed salmon escapes such as that which occurred in 2014.

795. The Minister’s EIA in 2015 recorded:1273

• Prevention of escapees is of paramount importance to MOWI.

• There have been no recorded escapes in the last 5 years of MOWI’s licensed aquaculture production in Bantry Bay.

• However, escapes from fish farms in Ireland have occurred. Three official reports of escaping farmed fish have been made to DAFM/MI/IFI since 2007: o In November 2009, approximately 25,000 salmon escaped from a cage. o In 2010 a report of an escape of 1,000 fish was made. o In 2010 a report of an escape of 83,000 fish was made.

• The incidence of escapes is generally lower in Ireland than in other countries.

• There was a fish loss at a salmon farm in Bantry Bay in early 2014 due to extreme and prolonged storm conditions but it was not possible to establish whether any fish actually escaped as a result of this event.

• The moorings and pen types deployed at the location were of an older generation of technology less able to withstand extreme weather events than the technology proposed for this application. The technology proposed for this application is to modern specification.

• There follows an appreciation of the risk posed to wild salmon if an escape occurs, such that mitigation of potential interactions with escaped farmed fish is essential and somewhat contradictory views that: o the Standard Operating Procedure – Emergency Plan for Fish Escapes identifies no direct actions aimed at reducing or eliminating any potential impact on wild salmon stocks, other than recapture by net deployment at sea. But the effectiveness of such measures is uncertain and they may cause an undesirable by-catch of wild salmon. o effective mitigation measures (e.g. removal of escapees from freshwater, where possible and practical) will assist.

1273 §19.1.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

796. I pause to observe that, as the DAFM has since accepted that 230,000 farmed salmon escaped in Bantry Bay in February 2014, it is remarkable and difficult to see how, 1 year and 4 months later, the Minister’s EIA is expressed in terms so equivocal as to the occurrence of any escape at all. I note also that Baker J in later proceedings1274 to which the Minister was party records as background to that case (though not itself at issue) that,

“The matters giving rise to the present application arose following a catastrophic storm of hurricane force on 1st February, 2014, as a result of which most of the fish stock amounting in total to 235,000 fish, and almost all of the aquaculture equipment on the site was destroyed.”

As the matter was one of judicial review and as Baker J does not record determination of any facts disputed int this respect, it is safe to infer that the Minister in those proceedings did not doubt the escape.

However, I do not need to further pursue that issue. Also, and as has been seen, Dr Saunders considers that after an escape prospects of recapture are pretty hopeless.

797. The Minister’s EIA in June 2015 concluded:1275

“This application will fall under the aegis of a new protocol governing the design, installation and maintenance of pens and moorings. Based on international experience, the introduction of this important mitigating factor will significantly reduce the risks associated with stock escapes.

The improved cage design and application of up to date technology will ensure the standard of structural design of the proposed development will meet the highest standards available to the industry, thus reducing the risk of escapes and minimising the interaction with wild salmonid species.”

Escapes – MOWI’s s.47 Reply of 2 November 2016 as to Cage Specifications & ALAB’s Response

798. Dr Saunders’ Final report in 2020 recited MOWI’s final position on the issue of cage specification, set out in its s.47 reply of 2 November 2016.1276 ALAB’s s.47 Notice had

“… noted that the Applicant acknowledged that the Shot Head site would be among the most exposed of all salmon farm sites in Ireland, with a location close to a downwind (prevailing wind)

rocky shore and cliff coastline, allowing limited scope for remedial action in the event of cage or mooring system failure or damage. Further information was requested on the intended cage and mooring system, together with evidence that the system had been successfully deployed elsewhere in

1274 Murphy's Irish Seafood Ltd v Minister for Agriculture [2017] IEHC 353. 1275 §19.1 & 2. 1276 Saunders Final report 8 December 2020 §10.1.1 Section 47 requests and subsequent responses, p102.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

similar conditions. Assurances were also sought that the system would be sufficiently robust to withstand a one-in-fifty-year storm event.”

Dr Saunders recorded MOWI’s Response as that MOWI:

“… points out that under current licencing arrangements the Engineering Division of DAFM can only impose a specification and grant certification of the installation design after the licence has been approved, so they are unable to provide final details of the proposed Shot Head farm installation. They do, however, indicate that the system will be comparable to that successfully used at the similarly exposed Clare Island smolt site.”

Dr Saunders’ advice in consequence is that:

“Since the suitability of the system will be subject to scrutiny and approval at a later date, we consider this matter can be managed by deferral to DAFM expert approval.”

It is clear from their effective delegation of the cage specification issue to DAFM that Dr Saunders and ALAB essentially accepted MOWI’s position in this regard.

799. In any event, as a matter of law MOWI’s position is incorrect. There was no legal impediment to its provision of the relevant information to ALAB. The reference to current licencing arrangements and the Engineering Division of DAFM is clearly to the DAFM Structural Design Protocol. As I observe below, that Protocol has no legal status beyond any accorded it by an aquaculture licence condition. Nor could “current licencing arrangements” administratively adopted by DAFM and taking the form of events post-dating the issuing licence affect the proper scope of EIA required by law and which must precede development consent

– which is not to say that Boland conditions as to matters of detail would be necessarily invalid. But neither Dr Saunders nor MOWI considers the terms of the Structural Design Protocol or, specifically, the requirements of the preliminary design stage which it stipulates or the adequacy of the information to hand in that regard from the point of view of adequacy of comprehensive EIA.

Dr Saunders’ Interim Report of December 2016

800. Dr Saunders’ Interim Report of December 2016 considered the issue of escapes. As much of its content was repeated in his Final Report of 2020 which directly informed ALAB’s Determination, I will concentrate on the latter to avoid duplication.1277 However, Dr Saunders’ Interim Report is important in that it informed ALAB’s action thereafter and it advises that “The prevention of escapes and approaches to reduce impacts are of fundamental importance to both the interests of the aquaculture industry and the conservation of declining wild salmonid stocks.”

1277 What follows here is taken from §9.2 of both the Saunders Interim Report of 2016 and Final Report 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

801. It also records that,

• in 2014 “a notable escape of 230,000 salmon farmed fish occurred from the Gearhies salmon farm1278 in Bantry Bay” – “the largest in Ireland’s history”.

• DAFM had in its 2015 EIA asserted that “it was not possible to establish any fish actually escaped as a result of this event”.

Strikingly, remarkably even (though undoubtedly correctly and properly), Dr Saunders’ says of the Minister’s EIA that:

• “It is clear, that contrary to the statement in the EIA, the escape of these fish is not in question. Some 230,000 fish were absent from the cages following the storm.”

• The “escape was confirmed by Minister Coveney, who stated that .. 230,000 fish had escaped the Gearhies farm due to severe and prolonged storms.”

• “Unfortunately, the scale of the escape event was not disclosed in the EIA.”

It should be said that ALAB’s EIA supervened but these observations do seem redolent of the Departments’ later attitude to transparency as to the 2014 escape.

802. Dr Saunders adds: “…… the potential for interaction between farmed escapee fish and wild salmon following a significant escape event …. is particularly relevant to the 2014 Bantry Bay escape” and “As the escaped fish were unaccounted for, there is a significant possibility that the wild Bantry Bay populations may carry a genetic consequence of that escape event, which will now need to be taken into consideration.”1279 Also, as noted earlier, he says that the effect of lice on wild salmonids should also be considered alongside the potential impacts from the 2014 escape “because of the known relationship between escapee fish and the augmentation or enhancement of natural infections in wild stocks”. 1280

803. Dr Saunders makes an observation as to the Minister’s EIA which is entirely of a piece with the concerns underlying the ICES and NASCO reports of 2016, the NASCO Containment Guidelines for 2001, ss.7 and 77 of the 1997 Act, the Chairperson’s report on the oral hearing1281 and ALAB’s determination, reflected in its s.47 notice to DAFM, that the documents sought of DAFM were necessary to its determination of the Appeal.1282 Dr Saunders states of the Minister’s EIA and the 2014 escape:

1278 Unconnected to any of the parties to the Proceedings. 1279 Emphasis added.

1280 Technical Advisor’s Final Report 8 December 2020 §9.1 Increased threat to wild salmon and sea trout from sea lice at p64.

1281 See below. 1282 See below.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“It is our opinion that the EIA undervalues the knowledge to be gained from this escape, in particular with respect to the genetic and sea lice impacts it may have had on the wild salmon population of the Bantry Bay catchment.”

He says of the undisclosed scale and cause of the 2014 escape:

“This is of importance as it demonstrates the possible magnitude of fish escapes associated with open cage farming and the difficulty in recovering and accounting for escapee fish.”

“Despite the present lack of official information on this escape occurrence, this event demonstrates, 1) the clear potential risk of escapes in Bantry Bay due to storm damage, 2) the magnitude of escapes that are possible, 3) the difficulties associated with assessing the scale of the impact due to inclement weather conditions during prolonged periods of stormy weather, and 4) the difficulty in accounting for, or recovering, escapee fish.”

In light of the Saunders’s Interim report it is little surprise that ALAB, reflecting its duty of comprehensive EIA, set about investigating the 2014 escape: indeed it would have been delinquent had it not done so.

Withdrawn s.47 request as to 2014 Escape & Events at the Oral Hearing in February 2017 (and Bias)

804. I now consider the allegation of bias made by reference to ALAB’s making, and later withdrawal, of a s.47 requirement it made of DAFM in 2017 to produce to it documents relating to the major escape of farmed salmon from the Gearhies salmon farm in 2014. As the grounds in question overlap considerably, much of the content of this section of the judgment is also referable to the allegations of inadequacy of EIA as to the risk of such escapes.

805. At expense of some repetition, I recall that DAFM’s own EIA in June 2015,1283 only a year after the 2014 escape, had observed that “Prevention of escapees is of paramount importance to the applicant.” And, “Given the small size of the salmon stocks in Bantry Bay rivers, and other areas along the possible migration or dispersal route of escaped farmed fish, mitigation of potential interactions with escaped farmed fish is essential.” Of particular note, the 2015 EIA states that,

“Efficient monitoring in freshwater for escapees, following large-scale escape events, and effective mitigation measures (e.g. removal of escapees from freshwater, where possible and practical) will assist in the ongoing maintenance of the status of the local wild salmon stocks”.

1283 12 June 2015 §19.1.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

806. This observation by DAFM in its EIA can only have been based on the expectation by DAFM itself that, following such a large-scale escape, such efficient monitoring and effective mitigation would in fact occur. I am unaware, as is ALAB, whether it did occur after the 2014 escape and with what results. Nor am I aware that ALAB was aware of what that monitoring revealed of the “genetic consequence of that escape event, which will now need to be taken into consideration” according to Dr Saunders. Indeed, in 2015 DAFM’s own EIA recorded that “There are very few systematic in-river monitoring programmes for farm escapees in Irish rivers and the scale of the potential risk is uncertain.”

807. Obviously, the records of such expected “efficient monitoring …….. and effective mitigation” following the 2014 escape, and their outcomes in terms of avoidance of – or occurrence of – environmental damage, inter alia to wild salmon stocks in the Bay and the rivers flowing into the Bay would necessarily be of considerable interest to ALAB in assessing, in considering MOWI’s licence application and in performing a comprehensive EIA, the reality of DAFM’s expectation of monitoring and mitigation as constituting reassurance in EIA as to the acceptability of the risk of large scape escape from MOWI’s proposed salmon farm. It is very easy to see why ALAB, in issuing its s.47 notice to DAFM, made a statutory decision that it was necessary to their determination of the Appeal to understand the lessons of experience from the 2014 escape as to the actual efficiency of monitoring and effectiveness of mitigation in the same bay as that for which the licence was sought and as to the largest escape in Ireland to date and as bearing on the “ongoing maintenance of the status of the local wild salmon stocks”. In addition, of course, such documents could be expected to shed useful light on the causes of the 2014 escape.

808. In light not least of its incorrect denial in the 2015 EIA that it was clear that any escape at all occurred a year earlier, DAFM’s resistance to publishing its reports on the 2014 escape, which resistance I describe below, has clearly, and predictably, (putting it as its very lowest) not served to dispel concern and

foster public confidence on these issues. However DAFM’s phrase “it was not possible to establish” is at least

suggestive that DAFM’s investigations were, if not finally complete by mid-2015, at least well-advanced.

809. As to the 2014 escape, IFI stated in their June 2016 paper that “a report on the event is being finalised by DAFM”.1284 At the oral hearing in February 2017, Mr Sweetman complained that a report on that escape by the “Marine Engineering Division” of DAFM was being kept “secret”, expressed his view that ALAB could not decide the application without seeing it and sought to require its production to the oral hearing.1285 Indeed the transcript reveals that Mr Sweetman weighed in on this issue on a number of occasions and I reject ALAB’s assertion that, for want of raising the issue in the process, he lacks standing to do so in these proceedings.

1284 Progress and challenges in achieving NASCO’s international goals in Ireland Millane, Gargan and Gallagher, IFI.

1285 Transcript 14 February 2017 p79, 15 February 2017 p51.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

810. At the oral hearing, FoIE asserted that the Commissioner for Environmental Information (“CEI”) on

16 January 2017 had ruled that release of the DAFM reports (there is more than one – though the position in that regard seems to have been somewhat confused at the oral hearing) was not inhibited by pending proceedings, which related to entirely separate issues, and they should be released.1286 Nonetheless, the DAFM representative at the oral hearing in effect said the report1287 was embargoed pending judgment in proceedings – which judgment was expected in mid-March 2017, after which the DAFM would consider the matter further. It is fair to say that the information disclosed to the oral hearing by DAFM representatives about those proceedings and the effect of the CEI ruling was vague – surprisingly so as it can only have been foreseeable to DAFM that the issue would arise at the oral hearing.

The CEI Ruling of 16 January 2017

811. The CEI ruling of 16 January 2017 was not exhibited. But it was included in the agreed authorities submitted to me.1288 And it was before ALAB when it made its decision.1289 In those circumstances I consider that I may have some regard to it. It records inter alia,1290 as follows:

i. FoIE had in March 2014 requested all records relating to the storm damage – which disclosure the CEI in July 20151291 had directed on appeal and which records had been disclosed.

ii. FoIE in September 2015 had requested any such records created since March 2014 – including in particular, ▪ scientific, technical and veterinary reports including interim draft reports of April and July 2014 and the final report referred to in the CEI Decision of July 2015 as due in August 2014.

▪ documents concerning these reports or concerning the incident, including scientific records or site visits, observations and survey logs of inspection themselves, photographs, and any related records, including emails, records of meeting or phone calls.

iii. The CEI had sight of the relevant documents – of which there was a very large number. They included:

▪ A "Farm Incident Report" dated 16 July 2014 by the Marine Institute. Referred to as “Report 'A'.”

▪ A "Storm Damage Report" dated 17 July 2014 by the Marine Engineering Division of the Department. Referred to as “Report 'B'.”

▪ Report B contained other reports, notably a "Condition Inspection Report" by Marine Specialists Ltd for the Department as to the condition of the fish-cages after the storm.

1286 Transcript 14 February 2014 p169. 1287 I am not precisely clear to which report or reports he was referring but nothing turns on it.

1288 Authorities Tab 139. CEI/16/0004 – Date of decision: 16 January 2017, Appellant: Friends of the Irish Environment Limited Public Authority:

Department of Agriculture, Food and the Marine. 1289 Oral Hearing Transcript 20 September 2017 p51. 1290 It is lengthy and cannot be fully set out here. 1291 CEI/14/0007.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

▪ An "Inspection & Assessment Report" dated 9 June 2014 by an engineering company retained by the fish-farm operator, and an addendum dated 25 June 2014. Referred to as “Report 'C'.”

iv. The findings of the CIE, as to the Department’s arguments against release that the documents consisted of internal communications and material in the course of completion, that ▪ Reports A and B were not in the course of completion. They were completed reports. That status was not altered by their later inclusion in a draft report being prepared for the Minister.

▪ While Report B was an internal communication, the public interest in its being publicly available outweighed the interest served by refusal.

v. The findings of the CEI, as to the Department’s arguments against release, that the documents consisted of commercial information. The Department had proposed redaction of Reports A B and C as to information on numbers and weights of fish.

▪ The CEI held that “The information on numbers of fish (and weights of fish, since that can indicate numbers of fish when their age is known) is crucially important environmental information, without which no assessment of the storm-damage reports (from an environmental point of view)

would be possible. The Department presented no argument to show that the confidentiality of this information is protected by law. I am satisfied that, even if the Department had made such a case, persuasively, the public interest in access to this information would outweigh any private commercial interest served by disclosure. In saying this I am mindful that both reports date from the summer of 2014 and any commercial sensitivity that might have existed at that time is likely to be much reduced now.”

▪ The earlier CEI Decision had identified three drafts, dated April 2014, July 2014 and November 2014, of another report which had been “due in August 2014”. It was apparent that further work had since been done on the November 2014 draft and the up-to-date draft included Conclusions and Recommendations. Some of the content exceeded the scope of the information request – in part as it post-dated the request. Other of the content fell within it. Included is the first report of the incident from the fish farm operator to the Department (information which had already been provided to FoIE following the earlier CEI Decision).

vi. The findings of the CEI, as to the Department’s arguments against release that disclosure would “adversely affect the course of justice, in light of ongoing judicial review proceedings”.

▪ The CEI examined the proceedings and was not satisfied that they concerned the storm damage event: they concerned matters which arose after the storm event which were outside of the scope of his review. So he was not satisfied that disclosure of reports 'A', 'B' or 'C' would adversely affect the course of justice.

vii. The findings of the CEI, as to the Department’s arguments against release that disclosure would be "premature" and "would unduly constrain the Minister in respect of any action which he might deem appropriate" and might breach the licence-holders right to natural justice. I will not recite the CEI findings here save to note ▪ his rejection of the Department’s argument of prematurity, and his view that “On the contrary, it could reasonably be said that the provision of such information to the public is considerably SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

overdue.” and that any prejudice to a third party would be outweighed by the “public interest in disclosure of the information”.

▪ His repetition of his view expressed in his earlier decision that "there is a very strong public interest in openness and accountability in relation to how the Department and Marine Institute carry out their functions under the legislation governing the aquaculture industry.

▪ His view, in similar vein, “….. that openness and accountability serve the interests of those involved in the industry, like the Company in this instance, since the public trust on which they depend, commercially, relies to a considerable degree on public confidence in the effectiveness of regulatory oversight.”

viii. The findings of the CEI, that the request for documents other than the reports was excessively broad and so burdensome but it was open to FoIE to submit a more specific request.

ix. The direction of the CEI to release Reports A, B and C.

ALAB’s s.47 Notice to DAFM & Withdrawal thereof

812. The oral hearing having adjourned, by s.47 request to the Minister dated 13 July 2017, ALAB noted that it had been indicated to the oral hearing that “the report” was due for release after mid-March 2017.1292 The notice contains the following determination by ALAB in accordance with s.47 of the 1995 Act:

“Pursuant to Section 47(1] (a) of the Fisheries (Amendment) Act,1997, where the Board is of the opinion that any document, particulars or other information ls or are necessary for the purposes of enabling the Board determine the Appeal, it may serve a notice on a party. ALAB is of the opinion that further information is necessary for the purposes of enabling the Board determine the Appeal.1293

In accordance with the provisions of section 47(1] (a] of the Act, the Board requires the following information:

A copy of the report/s if any, arising from the salmon escape Incident in Bantry Bay which occurred on 1 February 2014, insofar as they relate to cage specifications, issues concerning fish husbandry and fish escape.

……..

Please also note that a person who refuses or falls to comply with a requirement under section 47 (l)(a) shall be guilty of an offence.”

1292 Presumably a reference to Oral Hearing Transcript 14 February 2014 p168. 1293 Emphasis added.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

813. The Sweetman Applicants correctly observe that this notice constituted a formal statutory act by ALAB based on a formal statutory determination by it that sight of any such report was necessary to its determination of the Appeal. I include in the extract set out above the warning that failure to comply is an offence, not as an observation pointedly made at the Minister but as something known to ALAB such that they must be presumed to have exercised proper statutory care and judgement in determining that they should issue such a notice requiring the documents in question. To put it another way, it cannot be inferred that the notice was lightly or speculatively sent or sent other than on foot of the formal determination of necessity on which such a notice must, by statute, be based.

814. DAFM’s reply, dated 28 July 2017 asserted that,

• The draft report it referred to, while “at an advanced state of completion”, had not yet been finalised or submitted to the Minister for approval.

• DAFM was considering a recent High Court ruling concerning an aquaculture site referred to in the draft report and considering further action, following which the possibility of legal proceedings initiated by other parties cannot be ruled out.

• DAFM was also mindful of the Data Protection Act in respect of the draft report. (There is no elaboration on this point nor has there ever been any as far as the papers before me reveal. It is difficult to see that Data Protection could seriously have impeded compliance with the s.47 Notice)

• DAFM considered that “the release into the public domain of this report, which currently remains in draft form, would not be in the public interest.”

• DAFM offered to forward the draft report to ALAB “for the confidential information of the Board for the purpose of enabling it to determine the appeal as set out in Section 47(1)(a) of the applicable legislation and not for further dissemination.”

• If the Board wishes to receive the draft report intending it may be disseminated further by the Board, please advise accordingly so the Department may seek legal advice.

815. While this letter was portrayed by DAFM at the resumed oral hearing as an offer to provide the Report to ALAB, the Department must, or at very least should, have known that it was an offer ALAB could not possibly accept consistently with its duty of fair procedures in the quasi-judicial process of determining the Appeal. Notably, while the s.47 notice sought “report/s” plural, the DAFM did not mention

• the CEI ruling of the previous January.

• the reports other than that being prepared for the Minister (of which the CEI recorded at least four).1294

1294 "Farm Incident Report", dated 16 July 2014, by the Marine Institute. Referred to as “Report 'A'.”; “Storm Damage Report”, dated 17 July 2014, by

the Marine Engineering Division of the Department. Referred to as “Report 'B'.”; “Condition Inspection Report” by Marine Specialists Ltd for the

Department as to the condition of the fish-cages after the storm set out within Report B; “Inspection & Assessment Report”, dated 9 June 2014, by an

engineering company retained by the fish-farm operator, and an addendum dated 25 June 2014. Referred to as “Report 'C'.”

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Not merely that, but it later became apparent – DAFM admitted it at the oral hearing in September 2017 – that at least some of the DAFM documents encompassed by the s.47 Notice were already in the public domain when that letter was written in July 2017 and had been since March 2017 on foot of the CEI order made in January 2017. Indeed, some may have been in the public domain since the CEI’s decision of 2015.

816. The date of this letter, the information it contains as to the judgment referred to and the reference at the resumed oral hearing1295 to the judgment by counsel for SWI as having been delivered by Baker J on 1 June 2017 in judicial review proceedings challenging the Minister’s revocation of an aquaculture licence in Bantry Bay, readily identify the judgment as that in Murphy's Irish Seafood.1296 It is also apparent from that oral hearing transcript, that ALAB was aware of that judgment – at least by September 2017. It bears observing that the issues in that case were not the causes of, or effects of, the 2014 escape. They related to the later revocation of the licence for the Gearhies salmon farm on foot, as DAFM alleged, of failure by Murphy's Irish Seafood to comply with undertakings given after the escape as to the structural quality of reinstatement of the Gearhies salmon farm. This appears to conform, at least generally, to the view taken by the CEI in January 2017 of the relevance of those proceedings.

817. ALAB, entirely predictably and properly, replied to DAFM by letter dated 11 August 2017 requiring provision of the DAFM report on the basis that it would be disseminated in accordance with its statutory duty “to make all material considered by the Board in the context of an appeal available for inspection by parties to the Appeal.” Notably, ALAB stated that it noted “your wish to take legal advice, but the Board has however formed the opinion in accordance with Section 47(1)(a) of the Act that this Information is required and asks that it be provided.” So, ALAB clearly confirmed, reiterated and insisted on its statutory determination that it required the report as necessary to its determination of the Appeal.

818. ALAB continued in equally insistent terms: “The Board also wishes to draw to your attention the fact that the oral hearing In this Appeal shall resume on Tuesday 19th September 2017. Accordingly the Board requires that the material sought be provided no later than 1 September, to allow time for its dissemination to all parties to the Appeal.” In context, ALAB’s concluding reminder that failure to comply with a s.47 requirement is a criminal offence can only be regard as pointed.

819. The Minister replied in turn by letter dated 15 August 2017. That letter

• noted that compliance was mandatory on pain of criminal sanction and “Accordingly, the Department will comply with this statutory obligation as soon as is practicable.”

1295 Oral Hearing Transcript 19 September 2017. 1296 Murphy's Irish Seafood Ltd v Minister for Agriculture [2017] IEHC 353.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• repeated the (literally) unacceptable offer that “the Department is satisfied to forward the draft Report to you for the confidential Information of the Board for the purpose of enabling the Board to determine the appeal”.

• asserted that s.47(1) related only to the submission of documents and implausibly (given the requirements of fair procedures) asserted that it did not extend to their dissemination.

• as to dissemination, requested ALAB to consider that: o The Report is in draft only, having “undergone a number of iterations since it was commissioned in 2014”. “It will undergo further examination and amendment prior to finalisation”. It has not yet been submitted to the Minister for approval and does not represent his finalised views.

▪ (It bears observing that this was now over 3 years since the report had been commissioned.) o The report, as a draft, had not “directly” informed the Minister’s decision under appeal, such that its dissemination did “not seem appropriate” within the “scope of an appeal”.

▪ (Of course that the report had not “directly” informed the Minister’s decision left open the possibility that it or the escape had in some indirect way done so and in any event the observation in no way circumscribed ALAB as to its relevance. The reference to the “scope of an appeal” is difficult to understand of an appeal to be decided de novo.) o The draft report contained “preliminary observations and conclusions” as to third parties, who had not yet been invited to respond and dissemination would in effect deprive them of their right of reply before Ministerial approval of the report. o The same vague and uninformative information as to consideration of a high court judgment (which, of course, must have been in the public domain) and the possibility of further proceedings was repeated. It stated that “the possibility of legal proceedings initiated, by the Department and/or by other parties, cannot be ruled out”.

▪ Of course, in myriad circumstances, such possibilities can rarely be “ruled out”. o The draft report contained personal data the dissemination of which would be ALAB’s responsibility having regard to Data Protection legislation and the GDPR.

• “Given the legal Issues arising with the dissemination of the draft Report, and given that relevant stakeholders and officials are in many cases currently on leave at this time of year, we propose furnishing you with the draft Report by Friday 8 September 2017.”

This last paragraph, delaying provision of the draft report by about 3 weeks, is difficult to understand as the letter recorded that the decision to provide it had been made. And the letter clearly left the issue of dissemination to ALAB to decide.

820. ALAB, replied in turn by letter dated 31 August 2017 recording that it had taken legal advice and

could not “give an assurance that it will not furnish a copy of the draft report to third parties”. It set out its

reasons, which I need not repeat. It repeated that it was interested only in “the elements (if any) of the draft report dealing with cage specifications, issues concerning fish husbandry and fish escape” and, in that light

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

and as to third parties and data protection issues, requested that the Minister redact the report appropriately and provide the resulting excerpts by 4 September. The letter concluded with the following passage:

“If it is the case that some or all of the elements of the draft report sought are not conclusive or final, or if the draft report contains no such recommendations, then please advise accordingly, and the Board shall proceed to withdraw the request made pursuant to section 47 of the 1997 Act.”

821. Yet again, by letter dated 5 September 2017, the Minister repeated the offer, which by now it can only have known ALAB could not accept, to send ALAB the draft report “for the confidential information of the Board for the purpose of enabling the Board to determine the appeal”. It repeated its request that the matters I have already listed be considered by ALAB as to the issue of dissemination. As to the possibility of withdrawal of the s.47 request, the Minister expressed the view that “as the Report in its entirety is in draft form, no part of it can be considered "conclusive or final" at this point” and it asked how ALAB wished to proceed.

822. By letter dated 8 September 2017, noting the Minister’s view that “as the Report in its entirety is in draft form, no part of it can be considered "conclusive or final" at this point”, ALAB withdrew the s.47 Request. No prior minute of that decision by ALAB is exhibited – as one would have expected of a decision to rescind a statutory notice. However, ALAB ratified the decision on 8 October 2017.

823. It would in my view be wrong to criticise ALAB for its decisions and correspondence to this point, including its withdrawal of the s.47 notice, as unreasonable or unreasoned or for want of jurisdiction. ALAB had displayed considerable, almost confrontational, independence of DAFM to the point of withdrawal of its s.47 request for reasons stated – that the report was in draft – of which reasons one might take a variety of views but the withdrawal seems to me to have been within ALAB’s jurisdiction and discretion. A decisionmaker is entitled in light of new information to review and change its mind as to the necessity of information to its decision. However, the withdrawal proved to have been a decision taken on incomplete information provided by DAFM.

The Resumed Oral Hearing – the 2014 Escape & the Oral Hearing Report

824. The oral hearing resumed on 19 September 2017. Predictably, FoIE, again raised the question of DAFM’s draft report. Prof McIntyre said ALAB had rejected as irrelevant the Minister’s point that the draft report had not directly informed his decision but, on legal advice, had withdrawn its s.47 request because the report was in draft and therefore unreliable. FoIE cited the decision of the CEI dated 27 January 20171297

1297 Oral Hearing transcript 20 September 2017 pages 47 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

which, FoIE said, had rejected each of the same arguments made by the Minister against the s.47 request. FoIE asserted that as a result of the CEI decision, three of the reports which ALAB had sought from the Department had been in the public domain since 22 March 2017. Prof McIntyre said ALAB was aware of the CEI decision.1298

825. FoIE cited the CEI, whose 2017 decision was submitted to the oral hearing, as disagreeing that it would be premature for the public to find out what DAFM's engineers and the Marine Institute's scientist, and the Gearhies operator’s own engineering consultants said about the important storm damage, almost three years after the event. FoIE said two of the reports now in the public domain were final reports and not drafts – these were a farm incident report by the Marine Institute dated 16 July 2014 and a storm damage report by the DAFM Marine Engineering Division dated 17 July 2014. These in turn contained other reports, including by Marine Specialist Limited for DAFM. The latter reported on the condition of the Gearhies cages after the storm. For reasons not apparent, FoIE provided the hearing only with the title pages of the three reports.

826. FoIE sought the direction of the chair of the oral hearing “whether we should present you with these reports or whether you are now able to obtain the reports which you sought in your Section 47 letter as you are no longer constrained to any form of secrecy.” Prof McIntyre unsurprisingly replied: “I can guarantee you that ALAB will consider what we do with this report. It is very helpful.” FoIE interjected that it had presented only the title pages “because I don't think it is appropriate that I should give you the report”. Prof McIntyre replied: “No, it may not be appropriate that you give me the report but this was something that ALAB have made it clear that they felt this might contain useful evidence, that was the reason for the Section 47 request in the first place”. He said that ALAB would consider whether it could rely on the CEI decision that the reports were not in draft and would “reconsider whether we should take account of these reports.” He said: “…… as I said we will take advice, we have as I said already determined that we felt that this report may contain information that would be useful to our determination”.

827. Prof McIntyre observed that ALAB had been unaware that the reports were in the public domain. He asked the DAFM representative whether he knew they were in the public domain. I should say that the

1298 The transcript includes the following:

“MR. LOWES: I don't know if you have seen this document. It is a decision of the information commission on the 27th January?

MR. CHAIRMAN: Yes, we are aware of it.

MR. LOWES: 2017. You are aware the document was released into the public domain, I have those documents here and providing you with the

title pages of documents, please note, not document. These three reports are the reports that you were seeking from Mr. Quinlan and the Department

of the Marine. They have been in the public domain since the 22nd of March 2017. Each of the grounds that I have just read out were rejected by the

Information Commission and Mr. Kevin Hodnett provided me with a report. MR. CHAIRMAN: The question?

MR. LOWES: The question is this report was in the public domain since 22nd of March 2017.You asked the Department of the Marine for this

report, for those reports and they did not give them to you because it was not in the interest to have them released to the public domain. In fact, they

would only give it to you if you gave them assurances that you would give it to no one else when the documents were in the public domain.

MR. CHAIRMAN: We did not realise these were in the public domain as yet.” (Mr Hodnett, of DAFM, was a co-author of the DAFM EIA of 2015 in the subject Aquaculture Licence application.) SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

DAFM representative at the oral hearing had himself signed the letters to ALAB. He did not answer the question – he merely repeated the DAFM’s position as to, it seems, specifically the report to the Minister (which, presumably, would include be based on or cite the other reports): “The report is not complete, it has no standing. There are elements of it which are subject to regular change and no doubt will be subject to further change. The Department did not indicate to you that it wouldn't release it. In fact, it stated specifically that we would release it to you and that the reason that you don't have it is because you withdrew your request. So the matter again is a matter for ALAB's procedures and the Department's position is crystal clear, I think, from the correspondence”. 1299 I confess to finding this account of the correspondence to lack the necessary nuance but, as the correspondence had been disseminated, no doubt those involved formed their own views.

828. However, the oral hearing transcript records that Mr Hodnett of DAFM (not the DAFM representative to whom I have referred above) later confirmed to the oral hearing that he had released the

reports (other, it seems than the draft report to the Minister) “into the public domain” by letter to FoIE – it

seems in March 2017.1300

829. One can only observe that the allegation that relevant reports were already in the public domain damagingly impugned DAFM’s correspondence with ALAB. It is disquieting that the DAFM representative at the oral hearing did not immediately respond to, much less dispute, the allegations that while it was, putting it at its lowest, strongly discouraging provision of any reports to ALAB and offering provision on terms as to their circulation which it should have known ALAB could not accept,

• at least some reports had been in the public domain since March 2017 and

• the CEI had rejected the very objections which DAFM repeated in its later correspondence with ALAB.

830. Some time later, the question was put again to the DAFM representative (not Mr Hodnett) – whether when writing to ALAB he was aware that relevant reports were in the public domain because of the CEI decision. The DAFM representative replied unimpressively: “I think it is not something I would have dwelt upon” and then, equally unimpressively, retreated to the position that “ALAB, having considered it, choose to withdraw its request. That is a decision the Department respects and I can certainly not interfere with the

internal workings of ALAB.” The question was pressed – “I am just looking for a 'yes' or 'no', thanks. Just were

you aware it was put in the public domain?” The DAFM representative replied “I am not in anyway, I think a 'yes' or 'no' answer to this is going to mislead one person or the other. The fact of the matter is that I am

aware of lot of things on files and I dare say I forget a lot of things as well. The reality is we got a very specific

request from ALAB. We responded in good faith and very comprehensively to that and we offered ALAB a report, everything else is a matter for ALAB.” And on being pressed further: “If I could just, I would be generally aware of correspondence from the Ombudsman in relation to this and other matters. Did I focus on

1299 Oral Hearing Transcript 20 September 2017 p53.

1300 Oral Hearing Transcript 20 September 2017 p58/59. Mr Hodnett is erroneously identified in the transcript as “Mr Hognett”.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

that matter? No, I did not. I am still unclear and I would have to check our records but I don't think it is relevant. The relevant issue here is we were asked for a specific report from ALAB and we offered it to them.” He also said “I want to make it crystal clear in the responses that were drafted to ALAB, the matter of the Ombudsman was not in my mind or anybody else’s mind. We are and were dealing with the issues raised with us by ALAB.” As far as I have been made aware, the allegations that at least some of the reports covered by the s.47 request made of the DAFM are not in draft and had been in the public domain since March 2017 remain uncontradicted.

831. Prof McIntyre intervened to observe that “there is information here that is actually very useful to ALAB and ALAB will act upon. But what we are not aware of was that the decision of the Commissioner questioned or otherwise cast doubt on the draft status of evidence in this report. Now, I am not aware that our legal advisors were aware of that either and that may make a difference to our advice as to whether or not we can use that report. And that will be revisited. … I believe the Board will find that very helpful and something we can act upon.”

832. The DAFM representative then concluded of the reports “If you want it you can have it and you can

disseminate it … And it is up to you. This is entirely a matter for ALAB.”1301 ALAB has never explained why this

offer, in unconditional terms, was not taken up – or at least pursued in correspondence with DAFM.

833. One must be careful in reading transcripts – they lack the immediacy of the oral hearing. But it is impossible to be impressed by the DAFM’s account of matters as just recited. The suggestion that, in resisting delivery of reports on foot of a statutory process requiring their delivery it was irrelevant to consider and, for that matter, to tell ALAB, that the reports were already in the public domain (or at least that their disclosure to the public domain had been directed by the CEI) and that some were finalized and not in draft, is untenable. It led ALAB to consider the matter and withdraw its s.47 notice on significantly incomplete information. It is not for me to weigh the significance of that information. It suffices that ALAB’s issuing the s.47 notice to DAFM in the first place reflected its statutory determination, pursuant to s.47, that the information was “necessary for the purpose of enabling it to determine an appeal” and that its withdrawal of the notice was based on a significantly incomplete understanding of the position regarding the documents in question.

834. While Prof McIntyre’s report on the oral hearing did not directly address the matters described above, he did cite an objection as to “DAFM’s failure to release the incident report(s) relating to the major escape of farmed salmon in Bantry Bay” and he cited “the calls of a number of objecting Appellants (FoIE and Save Bantry Bay) that an incident report(s) into a large-scale escape of farmed salmon in Bantry Bay be immediately made available by DAFM.” He also, and significantly, recommended that, before determining

1301 Oral Hearing Transcript 20 September 2017 p62.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

the appeal ,“the Board should make every effort to consider the potential impacts of large-scale farmed salmon escapes.” It is perhaps a little unfair to regret that Prof McIntyre was not, in his report, more explicit as to the information available to him and on the transcript of the oral hearing which underlay this recommendation.

835. But, taking his report with the transcript, it is impossible not to conclude that Prof McIntyre:

• Accepted the importance of the issue of escapes – an importance reflected in the words “every effort”.

• By the words “every effort” recommended reconsideration of the withdrawn s.47 notice issued to the DAFM as to the 2014 escape, now that it had become apparent at the oral hearing in 2017 that at least some of the documents encompassed by that notice were by then o in final, not draft, form (unsurprisingly given the passage of time since the 2014 escape) and o in the public domain, having been released pursuant to a CEI decision which rejected at least some of the arguments on which the DAFM had relied in resisting the s.47 notice. o being offered by DAFM to ALAB on an unconditional basis.

Escapes – DAFM Reports on 2014 Escape – Adequacy of EIA – Decision

836. The reports mentioned by Mr Lowe as already in the public domain in 2017 are not before me – or even their cover sheets as placed by him before the oral hearing. The transcripts of the oral hearing are in evidence before me. They are, it seems to me, evidence of what was said at the oral hearing, though not, strictly, of the truth or accuracy of what was said save by way of admission against interest. However, and as informing ALAB’s discharge of its duties in EIA, what matters for present purposes is the fact that the oral hearing transcript and Prof McIntyre’s report of November 2017 constituted part of the information before ALAB from that point on.

837. There clearly was sufficient sufficiently reliable information available to ALAB as a result of the oral hearing – not least the ruling of the CEI of 16 January 2017 and DAFM’s confirmation per Mr Hodnett that the reports had been in the public domain since March 2017 and its offer to make the documents available unconditionally as to their dissemination – such as to require ALAB, in light of its earlier statutory determination that the documents were necessary to its determination of the appeal, to revisit the issue of bespeaking those documents from DAFM or otherwise obtaining those which were in the public domain. In the very particular, indeed peculiar, circumstances of this case this obligation seems to me to flow from

• the indisputable statutory and scientific recognition of the importance of escapes generally,

• ALAB’s statutory determination of the necessity of the documents,

• the fact that those documents were by now in the public domain and

• the reason to discern a significant possibility that ALAB’s withdrawal of its s.47 notice may have been based on incomplete information as provided by the DAFM.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

838. It flows also from the legal requirement, recited in Commission v Ireland C-50/091302 that EIA requires:

• As complete an assessment as possible of the direct and indirect effects of the project concerned.

• Accordingly, that the competent environmental authority must undertake:

o not merely analysis, but investigation.

o not merely examination of the substance of the information gathered but also consideration of the expediency of supplementing it, if appropriate, with additional data.

Clearly, in making its s.47 determination that the documents which it required of DAFM were necessary to its determination of the appeal, ALAB was carrying these duties into effect. As AG Kokott said in Namur- Est:1303

“The starting point is the objective of a comprehensive assessment of all environmental effects in the environmental impact assessment ..”

That requires not merely that each significant effect be identified but that each significant effect be comprehensively assessed.

839. Ordinarily, a competent authority such as ALAB is afforded a wide margin of appreciation and judgment as to what the requirements of investigation and getting supplementary information amount to in a particular case. However, and ex hypothesi, even a wide margin of appreciation has boundaries. The circumstances here are very unusual, striking and, in my view, compelling.

• First, it is undeniable that escapes cannot but have been high on the list of risks requiring comprehensive assessment in EIA.

• Second, ALAB had formed the opinion pursuant to statute that the DAFM documents as to the 2014 escape were necessary inform its determination of the licence application.

• Third, and, as ALAB points out in written submissions as to the alleged issue of excessive s.47 notices, where ALAB forms such an opinion it “shall” serve a notice accordingly – it has an obligation to seek the “necessary” information.

• Fourth, ALAB had so determined against the backdrop of the striking content of Dr Saunders’ Interim Report of 2016 – his description of what amounted to the DAFM’s failure to admit the 2014 escape in its 2015 EIA.

1302 See also Case C-404/09 Commission v Spain (Brown Bear) Opinion of Kokott AG of 24 November 2011, §177 – “Article 3 of the EIA Directive may

require the competent authorities to gather additional information, where necessary, in order to reach as complete an assessment as possible of the

direct and indirect effects of the project concerned …” 1303 Case C-463/20 Namur-Est Environnement ASBL v Région Wallonne, Opinion of Kokott AG of 21 October 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Fifth, and despite that determination, ALAB had withdrawn its s.47 notice in the face of DAFM resistance.

• Sixth, one notes the events of the oral hearing – which put at their lowest – cast appreciable doubt on the DAFM assertion that the documents relating to the 2014 escape were in draft form only, at least as to some of those documents. They also suggested that the DAFM’s arguments which lead ALAB to withdraw its s.47 request had, unbeknownst to it when withdrawing the request, already been rejected by the CEI with the effect that the documents, or some of them, were already in the public domain. It is not possible for me, or necessary, to find the various allegations made at the oral hearing to have been correct. It is also apparent that ALAB was aware of the CEI decision of January 2021. The information disclosed, whether accurate or not, amply sufficed to put ALAB on inquiry in those regards. Indeed the chairperson of the oral hearing expressly so acknowledged at the oral hearing.

• However, and seventh, the transcript is particularly notable for Mr Hodnett’s admission for DAFM that the reports in question had been in the public domain since March 2017 – three months before the start of the DAFM correspondence objecting to their circulation in ALAB’s quasi-judicial process.

• Eighth, the transcript is also particularly notable for the DAFM’s unconditional offer to provide the documents to ALAB without restriction as to dissemination.

• Ninth, there is the report of the chairperson of the oral hearing, considered in light of the foregoing factors, to the effect that ALAB should make “every effort to consider the potential impacts of large-scale farmed salmon escapes.”

• Tenth, there is the passage of about 7 years between the 2014 escape and ALAB’s determination, prior to which ALAB must have reasonably anticipated that in the ordinary course of proper and timely public administration any reports which had been in draft would have been finalised. Indeed in 2016, 5 years before ALAB’s determination, Ireland had told NASCO that a report on the 2014 event is “being finalised by DAFM” – suggesting it was in an advanced state of preparation. Further, in the context of Dr Saunders’ account of the true position as to the 2014 escape as confirmed by the Minister, it is notable that Ireland had told NASCO on that occasion that DAFM had advised that ‘it is not possible, at this time, to exclude the possibility that fish escaped nor is it possible to quantify the potential number of mortalities versus escapees’. And by letter of 28 July 2017, 4 years before ALAB’s determination, DAFM had stated that its report was “at an advanced state of completion”.

840. Notwithstanding the wide margin of appreciation and judgement to which I have referred, in this

strikingly unusual – compelling – set of circumstances, I cannot see that ALAB’s duties in EIA of investigation

and consideration of the pursuit of supplementary information, in the discharge of its obligation of “as complete an assessment as possible”, could have been satisfied by anything less than a reconsideration of renewing the s.47 notice to DAFM or otherwise obtaining those which were in the public domain. There is no suggestion, much less evidence, that these possibilities were considered. Notably, surprisingly, though Dr

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Saunders’ Final Report devotes its Addendum A1 to a consideration of the response to the recommendations of the chairperson of the oral hearing, it does not record any manner of addressing of his recommendation that ALAB make every effort to consider the potential impacts of large-scale farmed salmon escapes. Indeed, that recommendation is not even recorded in Dr Saunders’ Final Report.

841. As far as I can discern, ALAB did nothing to pursue its knowledge that material was, by DAFM’s admission, now in the public domain and available to ALAB, of which material it had previously formed the statutory opinion, pursuant to s.47, that it was necessary to its determination of the aquaculture licence application. My view seems to me consistent, and ALAB’s inaction is inconsistent, with the advice of its own technical advisor Dr Saunders that the “ever-present risk of fish escapes” which “present a considerable risk to wild salmon population …. the risk and impact of any large episodic escape is of considerable importance in the assessment of potential environmental impacts.” My attention has been drawn to no advice to ALAB or material before it upon which it could properly have concluded (not that they recorded any such

conclusion) that Dr Saunders and Prof McIntyre were incorrect in their view of the significance of the risk of salmon escape generally and the 2014 escape in particular or upon which ALAB could have relied to resile from their statutory opinion that the information in the public domain was necessary to their determination.

842. And even if, which I do not consider, the draft status of a report bears on the issue, given that as of July 2017 DAFM stated that it was “at an advanced state of completion”, it remains notably unexplained why, in the 4 years thereafter to ALAB’s determination, during which, as one would imagine, good administration by DAFM would have ensured finalisation of any report, ALAB did not revert to DAFM for that report.

843. While it is not a basis of the view I take, as I do not see that any duty to identifiable parties arose nor was one argued, I note in passing that good administration is a general obligation of EU states (including of state organs such as the Minister and ALAB) and a fundamental EU Law right of those affected by it – Eco Advocacy.1304

844. By reason of these matters, there was here was a failure by ALAB to perform an adequate – that is to say a comprehensive – EIA as to the risk of fish escape. The Aquaculture Licence will be quashed on this account.

845. My provisional view is that the matter should be remitted to ALAB’s reconsideration of the question whether it should again require from DAFM documents relevant to the 2014 escape. Any such requirement, if made, would presumably cover a lengthier time period than that which governed the CEI. Though it is a

1304 Case C-721/21, Eco Advocacy CLG v An Bord Pleanála, Judgment of 15 June 2023.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

matter upon which any necessary decision would have to await events, it would be surprising for ALAB to learn, in response to any such requirement, that the draft report to the Minister was, over 10 years from the 2014 escape, still in draft though it was in 2017 at “an advanced stage of completion”. In that event, the question could arise whether there remained any intention to complete it or reasonable prospect of its completion, the answer to which questions might arguably affect the outcome of any request by ALAB for its provision.

Escapes – EIA – CAGES, STORMS & SEALS

846. I must also consider whether ALAB’s EIA of the issue of escape was defective as to the issue of the structural and more general adequacy of the proposed salmon farm facilities having regard to the risks of fish escape due to storm and due to seal predation.

Mitigation of Escape once it has Occurred

847. The DAFM EIA 2015 is, at very best, equivocal as to the prospects of mitigation of the effects of an

escape once it has occurred. As to its identification as “essential” the “mitigation of potential interactions

with escaped farmed fish” (the predicate of which is that escape has occurred). I contrast DAFM’s expectation that, after a large-scale escape, efficient monitoring and effective mitigation would in fact occur with both its recognition in its EIA that the only measure contemplated “in the SOP aimed at either reducing or eliminating any potential impact on wild salmon stocks” was “recapture at sea”,1305 its doubt as to the efficacy and side effects on wild salmon of such efforts1306 and Dr Saunders’ dismal expectations (less than 3% recovery escapees) of such efforts.1307

Prevention of escape – DAFM EIA 2015

848. In the foregoing light it is no surprise, if not inevitable, that the DAFM EIA in 2015 clearly grounded its view of the acceptability of risk of escape in measures to prevent escape as opposed to mitigating the effects of escapes.1308 I have referred earlier to the DECLG advice to the Minister in August 2014 that the Site is in an unsheltered and reasonably exposed location such that MOWI would need to demonstrate the structural integrity of the proposed anchor/mooring arrangements and cages especially in relation to storms.

1305 EIA 2015 §19.1. p45. 1306 EIA 2015 §19.1. p45.

1307 See above: Saunders Final Report 8 December 2020 §9.2 p71: “In reality, any attempt to recover escapee fish is likely to result in a very low level of

success” – “recoverable numbers may be very low in comparison to total escape numbers” – “less than 3% of escaped salmon have been recaptured

through organised fishing after large escape episodes.” 1308 See generally §19.1&2.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

849. The DAFM EIA 2015 records, as to the 2014 (non-) escape,1309 that “… the moorings and pen types … were of an older generation of technology less able to withstand extreme weather events than the technology proposed for this application.” In contrast,

“The technology proposed for this application is to modern specification. The use of improved standards for cage design, together with better training, has been shown in Norway to reduce escapees significantly.”

“This application will fall under the aegis of a new protocol governing the design, installation and maintenance of pens and moorings. Based on international experience, the introduction of this important mitigating factor will significantly reduce the risks associated with stock escapes.

It seems likely that this observation, made in 2015, as to Norwegian escapes, falls to be considered in light of the later reports of ICES and NASCO in 2016, as recounted above, regarding the scale of the problem of escape, not least in Norway, and the work yet required to achieve the international objective of 100% containment.

850. The DAFM EIA records its “Conclusion” as to the risk of escape that,

“The improved cage design and application of up to date technology will ensure the standard of structural design of the proposed development will meet the highest standards available to the industry, thus reducing the risk of escapes and minimising the interaction with wild salmonid species.”1310

In other words, what would set the MOWI salmon farm apart from the Gearhies farm from which 2014 escape occurred, would be the better quality of its design and specification – though DAFM does not attempt to analyse, even roughly, the resultant reduction of risk.

Cage Specifications, Site Conditions & Seals – Saunders’ Final Report – unable to evaluate risk of escapes

851. Dr Saunders’ Final Report1311 bears extensive recall – albeit I have edited and paraphrased it somewhat below. I have also omitted certain positive statements in the report about the proposal as they are recited earlier:

1309 As it was all but portrayed at that time. 1310 §19.2. 1311 8 December 2020 p69 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• “Given the small size of the salmon stocks in Bantry Bay rivers, and other areas along the possible migration or dispersal route of escaped farmed fish, mitigation of potential interactions with escaped farmed fish is essential”1312

• A major escape would present a considerable risk to wild salmon – a risk of “considerable importance” in EIA.

• This is particularly relevant to the 2014 Bantry Bay escape due to storm – the largest in Ireland’s history.

Contrary to a statement in the EIA the occurrence of this escape is not in question. As the 230,000 escapees were unaccounted for, there is a significant possibility that the wild Bantry Bay populations carry a genetic consequence of that escape which will now need to be taken into consideration in EIA.

• Cages have improved but the Shot Head site is more exposed to weather and waves than the escape site.

• The official report to DAFM on this escape remains unavailable.

• The EIA undervalues the knowledge to be gained from this escape, in particular as to possible genetic and sea lice impacts on the wild salmon population of the Bantry Bay catchment.

• Unfortunately, the scale and cause of the escape was not disclosed in the EIA. This is important as it demonstrates the possible magnitude of fish escapes from open cage farms and the difficulty in recovering and accounting for escapees.

• Despite the lack of official information on this escape it demonstrates, o 1) the clear potential risk of escapes in Bantry Bay due to storm damage, o 2) the magnitude of escapes that are possible, o 3) the difficulties in assessing impact in prolonged stormy weather inclement o 4) the difficulty in accounting for, or recovering, escapee fish. In reality, recovery is likely to be very low.

• The most important mitigation measure against fish escapes is the installation of the correctly specified equipment for a given site.

• The EIS provides no details of cage robustness, resilience and reliability to withstand predicted storms, key elements in preventing fish escapes and relevant to the granting of any licence. The EIS states1313 that o before installation the final specifications will have to be submitted to DoAFM1314 for certification under its Structural Design Protocol. o Thus precise specification is not a matter for this document.

• Without the installation’s specifications we can’t comment on its structural suitability for the location or

ability to withstand a weather event of a particular magnitude.

• It is recommended and anticipated that these details will be submitted for expert consideration prior to the granting of a licence. 1315

• Seals pose a particular risk of escapes, as predation on farmed salmon and associated net biting, resulting in a net breach, make up 47% of reported escapes.

1312 EIA §19.1. 1313 Vol 1 §3.3.2.

1314 Engineering Section of the Aquaculture and Foreshore Management Division of the newly named Department of Agriculture Marine and Food.

1315 Emphases added.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Frequent visits from seals may be expected at the Shot Head site.

• From the EIS it is difficult to determine the anticipated risk of seal nuisance from seals and the effectiveness and impact of seal control.

• Due to the ever-present risk of fish escapes it would be prudent to establish whether this is an issue of concern to existing operations in the area, and, if so, whether measures are effective. A statement on the current impact of seals at the Roncarraig Site would enable a more robust risk assessment.

852. Notably, I am unaware that any further information on these issues accrued or expert consideration occurred prior to the granting of the licence. Had what Dr Saunders envisaged happened, the expert consideration in question could have been incorporated in ALAB’s EIA. But it did not happen.

853. Further, it appears clear from his Final Report that, though it was introduced in 2016 and sent to ALAB by MOWI in 2016,1316 Dr Saunders in 2020 had not seen, much less considered the content of, the DAFM Structural Design Protocol. His only reference to it was as described in the 2015 EIA as awaited.1317 Had he adverted to it, it is inconceivable that he would have given no account of it.

854. As to the cited content of MOWI’s 2011 EIS, the sentence: “Thus precise specification is not a matter for this document”, is a non-sequitur – at least to the extent of the degree of specification required for comprehensive EIA. That another public authority is expected to consider, after development consent has issued, a foreseen environmental risk does not absolve the authority competent in EIA from assessment of that risk prior to the issuing of development consent – Namur Est.1318

855. As the foregoing records and as bearing specifically on the issue of the withdrawn s.47 Notice, immediately prior to his observations that the DAFM EIA “undervalues the knowledge to be gained from” the 2014 escape and “Unfortunately, the scale and cause of the escape event was not disclosed in the EIA”, Dr Saunders had said1319 that, while the cages will be better, the Shot Head Site is more exposed to weather and waves than the 2014 escape site.

856. Dr Saunders generally had no objection to, and expected no additional adverse environmental impacts from, MOWI’s proposed “increased number of cages from 12 plus two temporary cages for grading to 16 with an additional two cages for management of disease treatments”.1320 But he entered a caveat: that “consideration should be given to the structural integrity of the preferred configuration, in particular its

1316 See ALAB Determination §2.8. 1317 Saunders Final Report 8 December 2020 p72. 1318 Case C-463/20 Namur-Est Environnement ASBL v Région Wallonne, Opinion of Kokott AG of 21 October 2021. 1319 Saunders Final Report 8 December 2020 p70. 1320 §9.11.3 Cage number and configuration.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

ability to withstand the previously evaluated exposure regime and the adopted 1:50 year storm event reference. Should a significant difference be predicted between each arrangement, a preference should be given to the least vulnerable pen arrangement.”1321 That did not occur in the process before ALAB.

857. On the risk of escape, Dr Saunders’ conclusion1322 was as follows:

“We do, however, have some concerns over the exposed nature of the site and its close proximity to the equally exposed steep and vertical rocky shores. The greatest risk of escapes is either from large episodic events associated with holes in nets caused by predators, or as a result of storm events.

While we have no doubt that MHI will adhere to strict inspection and maintenance regimes, the specifications of the farm cages and their ability to withstand the expected conditions are not supplied in the EIS, preventing any objective assessment of installation’s suitability for the potentially challenging location and leaving us unable to evaluate the risk of fish escapes from the Shot Head site. We accept that this element will therefore fall within the jurisdiction of, and require approval from, the DAFM Chief Engineer.”

858. Crucially, given the undeniable environmental significance of large-scale fish escapes if they occur, it is striking and vital that Dr Saunders found himself, ultimately, “unable to evaluate the risk of fish escapes from the Shot Head site”. Inasmuch as, as will be seen, ALAB relied on Dr Saunders’ report and took the matter no further in its determination than Dr Saunders had, Dr Saunders’ statement amounts to an admission of inadequacy of ALAB’s EIA as to the obligation of comprehensiveness. While that conclusion suffices to quash the Impugned Aquaculture licence in this regard, some further comment is required – not least as to ALAB’s imposition of a Boland condition abdicating assessment of the structural design to DAFM.

Escapes – ALAB’s Determination & the Aquaculture Licence

859. I have recited above ALAB’s Determination1323 as to the risk of fish escapes. I return to it now. Inter alia it states:

“The Board noted that certain Appellants had raised concerns over large episodic events associated with holes in nets caused by predators, or as a result of storm events. The Board noted the specifications of the farm cages and their ability to withstand the expected conditions were not supplied but also noted that approval of the cages specification will fall within the jurisdiction of and require approval from DAFM.”

1321 Saunders Final Report 8 December 2020 §9.11.3 p93. 1322 Saunders Final Report 8 December 2020 p73. 1323 §6.1.3.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

This passage is clearly derived from Dr Saunders’ final report.1324 But it fails to adequately record and address his view in three crucial respects. It omits Dr Saunders’,

• expression of “some concerns over the exposed nature of the site and its close proximity to the equally exposed steep and vertical rocky shores” (which must be read also in the context of his contrasting the Site as more exposed than the more sheltered Gearhies site from which a storm caused the 2014 escape.1325)

• observation that the non-supply of the cage specifications and their ability to withstand the expected conditions prevented “any objective assessment of installation’s suitability for the potentially challenging location”.

• observation that he was “unable to evaluate the risk of fish escapes from the Shot Head site”.

860. Nowhere does ALAB contradict that asserted inability or assert its own ability to objectively assess the installations suitability for the location or evaluate a risk which, on any conceivable view, required comprehensive assessment in EIA. Instead, ALAB merely invokes the so-called “jurisdiction” of the DAFM.

861. Rather, the recognition of the risk but the delegation and deferral until after the development consent of the task of its assessment is repeated in the following text:

“The location of the Site is exposed to prevailing winds with a possible considerable fetch. Since the suitability of the cage structures and system will be subject to scrutiny and approval by DAFM, as outlined in paragraph 6.1.3 above, the Board consider this can be managed by DAFM approval and adherence by the Applicant to the standards set out in the Structural Design Protocol, as revised from time to time.”1326

This passage fails to recognise the requirements within the licencing process imposed on ALAB by both the Protocol itself (which Dr Saunders had not described but ALAB had seen) and by the requirements of EIA.

862. Condition 4 of the Aquaculture License, headed “Containment of Stock”, requires MOWI to “take all steps necessary to prevent the escape of fish from its cages/pens”, to record any such escapes and to notify DAFM, the Marine Institute and IFI of such escapes within 24 hours of their occurrence. It also requires compliance with “the most up to date guidelines on fish containment developed by the North Atlantic Salmon Farming Industry and the North Atlantic Salmon Conservation Organisation (NASCO) Liaison Group” (i.e. the NASCO Containment Guidelines 2001 - 2006 version).1327

1324 §9.2 at p73. 1325 Saunders Final Report p70. 1326 ALAB Determination §6.1.10. 1327 CNL(01)53 Guidelines on Containment of Farm Salmon.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

863. The operative grant clause of the licence is for the cultivation of Atlantic Salmon inter alia, “in accordance with the plans and drawing(s) in accordance with Schedule 2 attached as approved of by the Minister”. Schedule 2 reads as follows:

“Schedule 2 contains:

• the approved plans and drawing(s)

Note: The Licensee will adhere to the standards set out in the Department of Agriculture, Food and Marine's Protocol for Structural Design of Marine Finfish Farms, 2016 and the Floating Facilities shall be approved by the Department of Agriculture Food and the Marine.”

864. The first two lines above appear to be pre-entered descriptions in the blank precedent licence form, awaiting completion by specific identification of the approved plans and drawing(s). The Minister’s draft licence of September 2015 included a drawing which depicted the proposed Salmon Farm in its original 12- cage format in both plan and sections. I make no observation as to the adequacy of that drawing (though the copy exhibited was very poor): the point is that it was appended to the licence as Schedule 2. In contrast, Schedule 2 of ALAB’s impugned licence contains no drawing or assertion that any have been approved. Instead the “Note” set out above appears.

865. No approved plans and drawing are attached and the DAFM has approved none. Instead, Schedule 4 of the Aquaculture Licence envisages retrofitting the drawings to Schedule 2 after the licence has been granted: “Before deployment of any farm components at the Site, the Applicant shall obtain the prior approval of the Minister to the Initial layout of the cages on the Site and such plan shall be included in Schedule 2 to the Licence.” Schedule 41328 repeats the “Note” to Schedule 2 as set out above.

866. §2.2. of the licence reads “Method: Cages/Pens subject to the stocking and/or deployment limits as specified in Schedule 4 attached.” Schedule 4 contains, inter alia, the following:

The Licensee will adhere to the standards set out in the DAFM Protocol for Structural Design of Marine Finfish Farms, 2016 ("Structural Design Protocol") and the Floating Facilities shall be approved by the Department of Agriculture Food and the Marine.

Before deployment of any farm components at the Site, the Applicant shall obtain the prior approval of the Minister to the Initial layout of the cages on the Site and such plan shall be included in Schedule 2 to the Licence.

1328 Inserted by ALAB Determination §6.5.3.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

867. Schedule 5 in the licence is not introduced in the body of the licence itself. It is headed “Additional conditions applicable to the licence.” It commences “The Licensee shall:” and a series of bullet-pointed entries follow. One reads:

“comply with the standards laid out in the Structural Design Protocol, as revised from time to time.”

868. §4.2 of the licence states that MOWI “shall comply with the most up to date” NASCO fish Containment Guidelines. The Sweetman Applicants1329 say that the NASCO Containment Guidelines 20011330 thus invoked are “very generic”. I agree. This seems to be another example of the practice of incorporating documents in permissions and licences without adequately considering what those documents actually say and whether their terms are adequately prescriptive to merit incorporation in conditions which must be legally binding. The NASCO Containment Guidelines clearly consist of statements of broad principle applicable worldwide. They are clearly not written or intended for direct and enforceable application as licence conditions. Accordingly, and by §7, the NASCO Guidelines envisage that they will be carried into effect by way of national action plans – presumably to flesh out in a manner suitable to local conditions and make concrete and enforceable the principles set out in the Guidelines. ALAB’s counsel asserted that “in particular, the NASCO standards, the international standards are kept very up to date, as we understand it.” I am unaware of any evidential basis for that submission. I have not been referred to any such action plan. I have seen protocols addressing some of the issues one would expect to see addressed in such an action plan. Notably however, the DAFM Structural Design Protocol neither invokes the NASCO Containment Guidelines nor purports to be part of an action plan within §7. Nor does it recognise NASCO’s 100% containment objective. Indeed, while in one sense it is obvious, it is nonetheless striking that the statement in the Protocol that it “primarily addresses structural design requirements to prevent structural failure and/or fish escape” appears only as an aside in a footnote.

EIA – Cages & Seal Scarers- Reliance on Later Regulatory or other Decisions & Boland Conditions

869. As I have said, the fact that an element of a project requires a separate and later regulatory consent by another public authority, does not absolve an authority legally competent in EIA, obliged to perform comprehensive EIA, from assessing significant environmental effects likely in the context of that later regulatory consent. That is, of course, unless further EIA will ensue in that later regulatory consent (e.g. Martin1331). But no-one has suggested DAFM will do an EIA when approving detailed design of the cages pursuant to its Structural Design Protocol.

1329 Affidavit of Noel Carr – sworn 27 September 2021. 1330 2006 version. 1331 Martin v An Bord Pleanála & Indaver, [2007] IESC 23, [2008] 1 I.R. 336 – amongst many other cases.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

870. Such later consent may be relevant in one or both of two ways: it may authorise mitigation of significant environmental effects likely to derive from the project and thereby affect assessment of those effects in EIA and/or it may itself generate significant environmental effects of the project which will require assessment in in EIA and/or AA.

871. The first possibility can be illustrated in the present case in that the cage design and specification to be approved by the DAFM is intended to mitigate the risk of occurrence of an escape and thereby the risk of significant environmental effects occurring by reason of such an escape. The EIA inevitably recognises that risk as significant and so as requiring comprehensive assessment. Obviously, evaluating in EIA the acceptability or otherwise of the risk of escape must – cannot but – be informed by a view of the likely efficacy of the cage design and specification in mitigating the risk of escape and the risk residual to such efficacy.

872. Both possibilities are illustrated in the present case by the prospect, whatever it may be, of the use of seal scarers. Their use, if they are used, will be intended to mitigate the risk of escapes by discouraging seals from preying on farmed salmon and damaging cages in so doing. But their use may also involve possible significant detrimental environmental effects by way of acoustic disturbance of the seals thereby discouraged. Such risk of disturbance requires EIA and, likely, AA as the seals in question are a conservation interest of the Glengarriff Harbour and Woodland SAC.

873. As to EIA as to the cage design, the necessity of such consideration, as part of a comprehensive assessment, is a reason why, at least in general, it is at least desirable that such other separate regulatory, or other, consent,

• precede the regulatory consent process in which EIA (and if needs be AA) is done (thus tending to render more certain the factual premises on which EIA is done and lending substance to rights of public participation in EIA) or

• itself be subjected to EIA or

• itself be limited in its scope by a Boland condition, imposed in the development consent for which EIA was done, leaving matters of detail – detail only – to later regulatory agreement

874. These principles are found in substance in the opinion of AG Kokott in Namur-Est1332 in the specific context of derogation licences under Art. 16 of the Habitats Directive. AG Kokott emphasised that EIA must be comprehensive. So, “all environmental effects of the project” must be assessed and “Member States may not exclude certain environmental effects” from EIA. Accordingly, she made clear, the environmental effects of effecting a derogation licence for purposes of the project for which development consent was sought had to be considered in EIA. Though the CJEU did not express itself in the same terms, in substance it agreed.

1332 Case C-463/20 Namur-Est Environnement ASBL v Région Wallonne & Cimenteries CBR SA, Opinion of Kokott AG of 21 October 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Namur-Est was considered in Jennings1333 prior to the trial of the present case in terms which, in substance prefigured the decision, after the trial of the present case of the CJEU in Hellfire-Massy.1334 In Hellfire-Massy the CJEU confirmed what had been prefigured in Namur Est: if a project subjected to EIA also requires an Art.

16 derogation, the derogation must precede the EIA1335 as otherwise the EIA would be incomplete for want of consideration of the environmental effects of the derogation. The CJEU said:

“It follows from that case-law1336 that, in the specific case where,

• first, the execution of a project that is subject to the dual requirement for assessment and development consent laid down in Article 2(1) of Directive 2011/92 involves the developer applying for and obtaining a derogation from the plant and animal species protection measures prescribed in the provisions of national law transposing Articles 12 and 13 of Directive 92/43 and where,

• second, a Member State confers power to grant such a derogation on an authority other than the one on which it confers power to give development consent for the project, that potential derogation must necessarily be adopted before development consent is given.

If it were otherwise, that development consent would be given on an incomplete basis and would not, therefore, meet the applicable requirements (see, … Namur-Est …§§52 and 59 and the case-law cited).”1337

I should make clear that I do not rely on Hellfire-Massy for present purposes as having stated new law since the trial of these proceedings: it merely better expresses principles already apparent in Namur-Est (cited by the CJEU in Hellfire-Massy) and recognised in Jennings.1338

875. While I need merely consider the comprehensiveness of the EIA and it is not for me to suggest in detail to ALAB how it should proceed, it may assist to observe that the Supreme Court in Krikke1339 has observed the practical necessity, in the planning context, of “Boland conditions”1340 delegating to other authorities the agreement, after the issue of development consent, of technical details.1341 That Boland conditions are acceptable in development consents subjected to EIA and AA is well-established.1342 I cannot see why the same practical necessity should not apply in the case of Aquaculture and Foreshore licenses. In

1333 Jennings & O’Connor v An Bord Pleanála & Colbeam [2023] IEHC 14, §668 et seq. 1334 Case C-166/22, Hellfire Massy Residents Association v An Bord Pleanála, Judgment of 6 July 2023.

1335 The CJEU spoke in terms of development consent rather than EIA but as it did so in the context of the “the dual requirement for assessment and

development consent laid down in Article 2(1) of Directive 2011/92” nothing turns on that distinction. 1336 Emphasis added. 1337 Layout changed for purposes of exposition.

1338 The Court in Hellfire Massy proceeded to judgment without an advocate general’s opinion, which tends to occur where no new law is at issue.

1339 Krikke v Barranafaddock Sustainable Electricity [2023] 1 I.L.R.M. 81, Hogan J §13 et seq. and Woulfe J §65. 1340 Boland v An Bord Pleanála [1996] 3 I.R. 435. 1341 So-called for their origin in Boland v An Bord Pleanála [1996] 3 IR 435.

1342 Arklow Holidays v An Bord Pleanála [2006] I.E.H.C. 15, [2006] I.E.H.C. 102, [2007] 1 I.L.R.M. 125, [2007] 4 IR 112 and People Over Wind v An Bord

Pleanála [2015] IECA 272, Case C-461/17 Holohan v An Bord Pleanála, [2019] PTSR 1054; Donnelly v An Bord Pleanála [2021] IEHC 834.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Shadowmill1343 it was noted that the decision-maker in considering whether and in what terms to impose a Boland condition may have regard to:

• the desirability of leaving to a developer who is hoping to engage in a complex enterprise a certain limited degree of flexibility having regard to the nature of the enterprise;

• the desirability of leaving technical matters or matters of detail to be agreed between the developer and the planning authority, particularly when such matters or such details are within the responsibility of the planning authority and may require re-design in the light of the practical experience;

• the impracticability of imposing detailed conditions having regard to the nature of the development.

876. However the Supreme Court in Krikke1344 also noted long-standing concerns as to how Boland conditions operate in practice – not least given that agreements on foot thereof, made after the relevant permission/licence has issued, are closeted from public participation. This suggests cautious scepticism as to their use and scrutiny accordingly of their use in a particular case. That seems to be particularly so in cases involving EIA and AA – in which the degree of discretion that can be left over to later determination is “considerably lessened” as compared to development consents not involving EIA or AA.1345 Generally, Boland conditions cannot be such as to render the EIA and/or AA incomplete or inadequate by allowing significant environmental effects to escape assessment – remembering that EIA must be “comprehensive” (Namur- Est1346) – that is to say, be “as complete an assessment as possible” (Case C-50/091347) and remembering that AA must guarantee, no matter what is later agreed, that there will be no adverse effects on the integrity of the European site(s) in question. See generally, Donnelly and cases cited therein.1348

877. In the present case, the argument that requiring MOWI to specify cage design and specification in the licence application process and address it in EIA would hobble its flexibility to avail of the environmental protection advantages of improvements in cage design between the grant of the licence and the construction of the farm is distinctly unconvincing. It purports to hold out the unsure prospect of the better yet to come as a pretext for not describing the good already to hand and for not demonstrating that the good is acceptable, when the good may be what is implemented. It implies a general proposition which could be applied to justify a failure to detail many mitigation measures and tend to hollow out EIA in considerable degree. EIA is not intended to stymie progress in environmental protection. Nor should it trap a developer into use of less effective technologies. But that is not a basis for abdicating responsibility in EIA. EIA must

• establish whether, as matters (including the terms of the development proposal and the state of technology) stand at the time of the EIA, the environmental risks of a project are significant,

1343 Shadowmill v An Bord Pleanála & Lilacstone [2023] IEHC 157. 1344 Krikke v Barranafaddock Sustainable Electricity [2023] 1 I.L.R.M. 81. 1345 Donnelly v An Bord Pleanála [2021] IEHC 834.

1346 Case C-463/20 Namur-Est Environnement ASBL v Région Wallonne, Opinion of Kokott AG of 21 October 2021 – “The starting point is the objective of

a comprehensive assessment of all environmental effects in the environmental impact assessment” and “the development consent must also

comprehensively regulate those effects.” 1347 Case C-50/09 Commission v Ireland.

1348 Donnelly v An Bord Pleanála [2021] IEHC 834, Holohan v An Bord Pleanála (Case C-461/17); Sliabh Luachra v An Bord Pleanála [2019] IEHC 888; Kemper

v An Bord Pleanála [2020] IEHC 601; People Over Wind v An Bord Pleanála [2015] IECA 272. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

and

• investigate examine, analyse, evaluate and describe1349 the environmental risks qualitatively and, to the extent reasonably possible, quantitatively, so that the decision-maker as to development consent can decide if those risks are acceptable.

878. In that regard, and given in particular the duration of the licensing process in this case, during which cage design likely advanced, it was open to ALAB to:

• consider the application on the basis of such cage1350 design as was proffered in the licence application. or

• serve a s.47 notice at an appropriate date, perhaps close to the anticipated decision of the licence application, inviting updated and/or more detailed cage design and specification reflecting improved technology as compared to that described in the licence application.

• consider the licence application on the basis of that updated cage design and specification and the possibility of its circulation for comment and perform EIA accordingly.

• impose, in light of that consideration, a condition as to minimum standards of cage design. And/or

• insert a Boland condition allowing flexibility of cage design and specification in matters of technical detail to reflect any technological improvements after the licence has issued and before development of the salmon farm.

879. Whether a particular matter amounts merely to technical detail will turn on the particular circumstances of the case – what is mere technical detail or design in one case may be a vital matter in another. As Barr J said in Donnelly1351 a Boland condition, “must be read in light of the planning documentation and the inspector’s report”. The agreement on foot of a Boland condition may only “implement that which has already been decided in essence.” McKechnie J in Kenny1352 said that whether a Boland condition “is intra vires is a matter of degree and depends on the nature of the matter left for

resolution ….”. That the validity of such conditions was a matter of degree derives explicitly from the decision

in Boland itself. For an example of the effect of circumstances, see Shadowmill,1353 in which it was accepted that “what might be merely matters of detail as to other buildings and so fit for later agreement with the planning authority, are not matters of detail as to protected structures”. It was also observed in that case

that “the word “limited” in the phrase in Boland, “limited degree of flexibility”, sheds appreciable light on the

1349 Case C-50/09 Commission v Ireland, EU:C:2011:109, §§35, 37-41; Commission v Spain, Case C-404/09, EU:C:2011:768, §§78-80.

1350 Here I used the word cage as encompassing the entire installation. 1351 Donnelly v An Bord Pleanála & Cavan County Council [2021] IEHC 834. 1352

Kenny v An Bord Pleanála & Ors [2001] 1 I.R. 565, §9. Mr Kenny impugned a Board decision, of which Condition 8 required the submission of revised

drawings of the development for agreement with the Planning Authority. Inter alia, he impugned that Condition as ultra vires in leaving the developer

and the planning authority is at large, as to the agreement in private and without public participation of the appearance, nature and scale of the ultimate

development. He said that thereby what was granted was permission for an unknown development. McKechnie J rejected that challenge – essentially as

failing to read Condition 8 as part of the permission as a whole, including Condition 1 which required the development to be carried out in accordance

with revised plans which had already been submitted to the planning authority. Read in context of the inspector’s report Condition 8 was directed at

correction of certain identified discrepancies and ambiguities in the drawings already submitted. 1353 Shadowmill v ABP & Lilacstone [2023] IEHC 157, §407.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

concept of matters of detail” – which concept in that case was to be considered in light of the degree of statutory protection of protected structures. It was also observed in Shadowmill that,

“Two things at least, are notable in this observation by McKechnie J: first, that it describes an issue of jurisdiction, necessarily for the Court to decide. Second, that it represents one of those situations in which the Court cannot escape exercising at least some degree of what might otherwise be considered planning judgement proper to the Board rather than the court. I need not interrogate its underlying rationale to accept the view of McKechnie J, but it may proceed from the principle that the Board cannot be allowed to determine the scope of its own jurisdiction to impose a Boland Condition, which is a jurisdiction limited to technical matters and matters of detail.”

In Dooner1354 O’Neill J said that the scope for change left by a Boland condition “would have to be of a very limited and technical nature and not such as to excite significant public interest and/or objection.” That can certainly not be said of salmon cage integrity.

880. One implication of the foregoing, combined with the principles that EIA must both precede development consent and be comprehensive, must be that a Boland condition is invalid if it operates to exclude from the scope of EIA (and/or, a fortiori, AA) assessment of significant environmental risks by deferring their consideration to after the issuing of development consent such that they escape comprehensive EIA or, for that matter, AA. Such an exclusion would deprive the EIA and/or AA of its mandatory comprehensiveness.

881. Whether a particular matter amounts merely to technical detail will also turn on whether the condition adequately stipulates the principles on which, and bounds within which, the details are to be agreed. The Court of Appeal in Krikke1355 pithily summarised Boland as requiring that in such conditions the decisionmaker is to state “…. the overall objective to be achieved by the matters which have been left for such agreement; to state clearly the reasons therefor and lay down criteria by which the developer and the planning authority can reach agreement.” As I have said, the resultant agreement may only “implement that which has already been decided in essence.”

882. Finally, I should address the Respondent’s reliance on People over Wind1356 – a wind farm case in which Hogan J, in the context of risk to a FwPM population in the river Nore, upheld a Boland condition requiring agreement with the planning authority prior to the commencement of the development of a construction management plan, to include means to ensure that surface water run-off is controlled such that no silt or other pollutants enter watercourses. Hogan j held that the Board’s statement of principle was crystal clear and the realisation of that objective by the expert design and implementation of the requisite

1354 Dooner v Longford County Council [2009] 4 IR 619, §13 et seq. 1355 Krikke v Barranafaddock Sustainable Electricity Ltd [2021] IECA 217 (Donnelly J, 30 July 2021), §59. 1356 People Over Wind Environmental Action Alliance Ireland v An Bord Pleanála [2015] IECA 272.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

mitigation measures was essentially a matter of detail in the sense envisaged by the Supreme Court in Boland. It must be said that the relatively brief consideration of the issue in the court of appeal is likely to have been influenced by the finding in the high court that specification of such matters – such as the size and position of proposed drains and settling ponds – in the planning permission was “wholly impractical”. In my view no such impracticality arises here. Further, what is at issue here is not an ancillary aspect of the proposed development: it is the design and structure of the development. To borrow a planning concept – it consists of the essential works themselves as opposed to their use or woks ancillary to them. Further, it relates to a risk clearly identified on the papers as one of high significance in which the international objective (I do not suggest it is legally binding but it is highly relevant) is of 100% containment despite whatever risk there is of extreme weather events on an exposed site near and downwind of the coastline. I am far from suggesting that ALAB cannot be reassured on these issues and consider any residual risk acceptable. But, remembering that what constitutes detail depends on the circumstances of the case and is a matter of degree, I do not see the issues of cage integrity as being of detail.

883. One method of adequately identifying in a Boland condition the applicable principles, criteria or objectives is by reference to another standard or protocol setting them out. That is the method ALAB adopted here in invoking the DAFM Structural Design Protocol. The question is whether it did so successfully. In this context it seems to me that any such Boland Condition must reflect consistency as between the aquaculture licencing process, including EIA, and the terms of the Protocol itself. By this I mean, as to the present case, that,

• the premise of the DAFM Structural Design Protocol is that certain steps as to preliminary design will already have been taken in the licensing process before the licence issues and, impliedly, will have informed EIA in the licensing process,

• that premise must have been realised if the Boland condition invoking the DAFM Structural Design Protocol is to be valid.

884. In general terms there is nothing legally untoward in Dr Saunders’ observation1357 that MOWI “should be given the flexibility to explore improvement options and where appropriate select and upgrade the cage specification and design as required, subject to consultation with, and subsequent approval from, the licencing authority.” (While that approval could be “subsequent” to the grant of the licence it must be “prior” to effecting the development.) But, importantly, that observation and that flexibility can apply only

• if comprehensive EIA has been achieved and

• within the proper limits of Boland conditions permitting such approval. Such flexibility cannot operate to diminish the proper scope of the information required to be provided by MOWI as to structural design and specification or to diminish the assessment required of ALAB – as required by the law as to Boland conditions (that the decision be made “in essence”, leaving only matters of detail outstanding) and by the law as it requires comprehensive EIA and/or AA. That Boland flexibility may allow of a better solution in time cannot imply that it is unnecessary to decide the acceptability of the solution

1357 Saunders final report 8 December 2020, §9.11.2.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

proffered now. As I have said, there is no reason that such a Boland condition, approving of a cage design and specification, could not allow for the later agreement of a design and specification amended as to detail to improve environmental protection. Nor need flexibility for other good reasons be excluded as long as the resultant environmental protection is no worse than that provided by the approved cage design and specification the risks (if any) whereof have been deemed acceptable. Such a Boland condition would provide certainty as the baseline of environmental protection considered acceptable while allowing the degree of flexibility desirable and permissible in law as to improved specification and design.

885. It must be remembered also that if improvements in specification and design outstrip the flexibility afforded by a Boland condition and the concept of detail, s.68 of the 1997 Act provides for amendment of a licence.

Escapes – DAFM Structural Design Protocol, April 2016

DAFM Structural Design Protocol – Legal Status & Boland Conditions

886. The DAFM Structural Design Protocol was published in April 2016 – after the EIS and the Minister’s 2015 EIA (which envisaged it1358) but before the Technical Advisor’s Final Report and the impugned ALAB Aquaculture licence. Both Dr Saunders and ALAB contemplate its application in approval of a final design of the Salmon Farm after the Aquaculture Licence had issued and as a matter of the “jurisdiction” of DAFM. But neither adverts to the specific requirements of the Protocol as to the information to be provided and assessed in the licensing process (as opposed to after the licence had issued) or assesses the information provided by MOWI by reference to those requirements.

887. The ALAB Determination delegates the “approval of the cages specification” to the “jurisdiction” of DAFM. So, it is necessary to consider the legal status of the Structural Design Protocol. That is also necessary as the Minister relies on that Protocol in defending the escape grounds in these proceedings. Mr Waldron for DAFM states1359 that implementation of the Protocol is overseen by the DAFM Marine Engineering Division which provides technical advice to DAFM in respect of all aquaculture licence applications. Design drawings submitted to DAFM must be approved by an engineer. Such drawings are reviewed by the Marine Engineering Division as part of the licensing process and it provides technical reports in respect of the structural / engineering aspects of all applications. However, and as will be seen, these averments must be understood as relating to two different stages – one preceding and one succeeding the grant of the aquaculture licence.

1358 §19.1 & 2.

1359 Affidavit of Ultan Waldron, Principal Officer at the Department of Agriculture, Food and the Marine, sworn 13 July 2022.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

888. It should be emphasised that, ceteris paribus, there is nothing wrong with the issuing of administrative protocols such as the Structural Design Protocol as, for example, advancing good practice or making known DAFM expectations of good practice. However that is – or at least can be depending on circumstances – a different matter from the legal effect and enforceability of such protocols. In its terms, the Structural Design Protocol does not assert or invoke any legal authority or basis for its promulgation. It is not promulgated pursuant to any statute or statutory instrument. As far as I am aware, the DAFM Structural Design Protocol has, per se, no legal force and no DAFM “jurisdiction” derives from it per se. It is not apparent on what legal basis – other than a condition of an aquaculture licence – DAFM could insist on compliance with the Structural Design Protocol before installation of a salmon farm or on what legal basis and with what legal effect the Department may certify compliance with that Structural Design Protocol as a basis on which installation of a salmon farm may proceed. The State agrees these views of the legal status of such protocols.1360

889. Nor is it in law permissible, by the application of the scheme of such a protocol, to defer consideration of or limit or exclude from the scope of the statutory aquaculture licensing process or any EIA or AA process any issue properly requiring assessment in those processes. Also, a non-statutory approval pursuant to such a protocol could not constitute a development consent for EIA purposes or a project approval for AA purposes such as to provide a legal context for the carrying out of any EIA or AA process. So there is no prospect of EIA or AA in any approval process under the Protocol.

890. In referring to the “jurisdiction” of the DAFM, ALAB appears to have been under a misapprehension. The DAFM “jurisdiction” derives solely from requirements of the Aquaculture Licence by way of a condition

requiring compliance with the “standards laid out in the Structural Design Protocol”.1361 Accordingly, it is not

a jurisdiction to which ALAB must defer: it is a jurisdiction which ALAB itself confers by its condition requiring compliance with the Structural Design Protocol.

891. In general terms, it is of course, permissible to impose a condition in an aquaculture licence requiring compliance with such a protocol – thereby indirectly, but nonetheless effectively, conferring legal effect on the protocol in the individual case of a particular aquaculture licence. But to be valid such conditions, insofar as they leave matters to agreement after the licence has issued, must comply with the legal requirements of Boland conditions to which I have referred.

1360 Transcript Day 12 p131. 1361 ALAB Determination §6.1.3.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

DAFM Structural Design Protocol Content & the Aquaculture Licence Boland Condition

892. As stated above, the Aquaculture Licence in substance includes a Boland condition letting cage design and specification to later agreement with and approval by DAFM. Leaving aside whether that is properly a matter of technical detail, it is necessary to consider the terms of the Structural Design Protocol to discern if the principles it states are compatible with the Boland condition viewed in the context of the content of the materials before ALAB.

893. Though undesirably tucked away in a footnote to the Structural Design Protocol, its purpose and

objective is incidentally stated as follows: “… this protocol primarily addresses structural design requirements

to prevent structural failure and/or fish escape”. The stated fundamental principle of the Structural Design

Protocol is that “All marine finfish farms shall be subject to full structural design. They shall be designed to at

least meet the standard set out in this protocol.”

894. The Protocol envisages two stages: preliminary and detailed design. Importantly, The Protocol envisages that the preliminary design stage must precede and be considered in the aquaculture licensing process. The two stages are described in the Introduction to the Protocol as follows:

“The preliminary design is carried out to the extent that preliminary design loadings and resultant applied forces are established and these are evaluated in the design analysis to produce a provisional outline design for the farm structure.

The preliminary design will indicate position and orientation of the main structural components of the finfish farm, identify principal dimensions of nets, pens and moorings and include sufficient detail to allow production of a set of site specific drawings that illustrate the structure of the proposed farm. These preliminary design drawings will be submitted as part of the licence application and will also be referenced in the licence (if issued). It is important that the drawings are drafted to a professional engineering standard and include all relevant information. Drawing requirements are set out later in this protocol.

The detailed design stage follows later, following issue of a licence and before farm deployment.

Following selection of the equipment supplier, the licence holder will be in a position to specify in detail all elements of the structure proposed for deployment at the site. The capacity of each element to withstand the applied design forces shall be evaluated in detail in this second detailed stage of the design.”

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

895. The sequence envisaged by the Protocol is important. Preliminary design precedes and informs ALAB’s determination of the aquaculture licence application. Detailed design is done after the licence issues and DAFM approval of that design is required before installation of the salmon farm.1362

896. §2 of the Protocol is headed “Design Information to be provided at licence application stage” and includes the following:

• 2.1 Preliminary Design o A preliminary design shall be carried out prior to an application being submitted. o The preliminary design shall consider in sufficient detail the provisions outlined in Sections 4, 5, 6, 7 and 8 of this protocol,1363 to allow the drafting of a set of site-specific drawings. o It shall also include an initial assessment and sizing of the proposed set of nets, pens and mooring system to withstand the maximum design loads acting on the farm. o This preliminary assessment shall be such as to identify the principal structural components of the proposed farm and their principal dimensions. o The principal output from the preliminary design shall include a set of drawings showing the proposed farm as designed following completion of the preliminary structural design which output shall “be to such a level of detail so that the later detailed design will not significantly alter the principal dimensions and orientation of the proposed net, pen and mooring system.”

I pause to observe that this latter requirement of the Protocol is consistent with the necessity to the validity of Boland conditions that they leave only matters of technical detail to decision after the licence has issued. And the word “significantly” in this context must, it seems to me, encompass significance as to environmental effect.

• 2.2 Drawing Requirements (Preliminary Design) o The minimum drawing requirements considered necessary for illustrating the preliminary design are set out in the Appendix to this protocol.

▪ The Appendix specifies standards to which drawings are to be prepared1364 and describes their required content.

1362 §3.2. 1363 Emphasis added – as to their content see below.

1364 All drawings showing the structural design of the marine finfish farm shall be drafted to a professional engineering standard. They shall include the

following details:

• Farm Company name / logo

• Engineer’s Name / Logo

• Project Title.

• Drawing Title.

• Drawing Number (plus revision version as appropriate)

• Drawing status.

• Scales in use (for each drawing detail)

• Scale at sheet size or scale bar as appropriate

• Drawing date.

• Draughtsman’s and Approver’s name, initial and dates

• Drawing notes as appropriate. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o The licence applicant shall include the preliminary design drawings with the application.

o If the preliminary design and/or drawings submitted with the application are deemed inadequate or incomplete the application may, in the case of serious deficiencies, be rejected; or, if requested, the applicant shall revisit the preliminary design and provide revised design drawings prior to issue of the application for public consultation.

o If a licence is granted the preliminary design drawings submitted will form an integral part of the licence; and they will be specified as approved drawings in a schedule to the licence.1365

897. Remembering that the Protocol cites them specifically as to the content of preliminary design, for assessment before a licence is granted, it is relevant to note that Sections 4 to 8 of the Protocol identify, inter alia,1366 the following matters to be addressed in preliminary design:

• 4. Design conditions; Worst case assumptions (50-year return) as to wind speed, current velocity and storm wave height and period from at least 8 equally-spaced directions

• 5. Estimated worst case loadings wind, wave and current loadings acting on the farm components, including in combination, mathematically calculated and analysed to derive loads and forces acting on the net pen and moorings

• 6. Net Design “designed to structurally withstand the maximum design loads acting on the net”.

“Appropriate material factors shall be applied to yield strength values in providing for sufficient reserve structural capacity in materials used. Allowance shall be made for reduction in net twine breaking strength over time and the threshold strength values set to determine net serviceability.”

• The design shall provide sufficient clearance between net and sea bed under all tidal and wave conditions.

The net weighting system, for maintaining net shape under hydrodynamic action, shall be designed to be effective at maximum design loads.

• The design shall consider the adequacy of net containment for anticipated maximum fish stock levels and shall ensure against excessive net bagging/ fish stock crowding at maximum design loads on the net.

• The net shall be designed to maintain full containment of stock for the predicted design load conditions.

• 7. Pen Design; The pen system1367 shall be designed to structurally withstand maximum design loads acting on the pen from direct wave, wind and current actions and loadings transferred from the net, net weighting system and the mooring system.

• Again as to pen design, “Appropriate material factors shall be applied to yield strength values in providing for sufficient reserve structural capacity in materials used.”

1365 Emphasis added. 1366 What follows is not a complete list. 1367 The pen system is the cage superstructure and flotation rings or units.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• 8. Mooring Design. The requirements are similar to those imposed as to net and pen design.

Also, “each mooring element and fitting shall have a stated minimum load capacity requirement arising from the design.” And “All main components in the mooring system such as mooring ropes and chains, grid ropes, bridles, mooring buoys, anchors and all load bearing connections, such as shackles, swivels, links and chain plates, shall be designed to withstand the maximum calculated design loadings.”

898. More generally and as a matter of EIA law – Nordrhein-Westfalen1368 – it seems to me inevitable that the preliminary design information which must be supplied in the licence application must be sufficient (within the limits of reasonable requirement of a private developer) to enable

• the public to obtain an accurate and adequate understanding of the likely environmental impact of the project.

• ALAB to comply with its obligation of comprehensive EIA. On the particular facts of this case, that includes the data as to the function of the cages in mitigating the risk of escape and the assessment of residual risk of escape after mitigation by a cage installation characterised by structural integrity. Sufficiency of available information for purposes of EIA is generally a matter of judgment for the licensing authority – in this case ALAB. However, the court has a duty to verify the adequacy of the information to inform the public – Nordrhein-Westfalen1369 – by necessary extension that implies at least a residual duty to verify its adequacy to enable ALAB to comply with its obligation of comprehensive

Escapes – Cage Information and Drawings before ALAB & its Adequacy

899. The “general layout of components” is shown in Figures 63 (“Generalised cage and farm Layout and Specification Design”) and 64 of the 2011 EIS and their proposed orientation in the area is shown in Figure 4. These are a general plan layouts of a 12-cage farm but do not contain the detail specified in the Structural Design Protocol. The grid buoys, mooring system layout and expected anchor positions are schematic only – not to scale, so there is no drawing depicting the overall 19.20 hectares, being 45% of the proposed licensed foreshore area. I have not been directed to any cross-sectional elevation in the EIS. Though Figure 641370 may provide some indication it seems to be an “off the shelf” illustration. The Minister’s 2015 Draft Aquaculture licence at Schedule 2, did include a 12-cage drawing which included some sections/elevations (generally illegible as to hand) but neither it nor any 16+2 cage equivalent were included in ALAB’s Aquaculture Licence. More generally, all those drawings – all the drawings before ALAB – are specific to a cage grid area of 5.88ha and 12 cages. My attention has been drawn to no drawings of the proposed 16+2 cages on a footprint of 8.82 hectares plus anchors and moorings occupying an overall area of 24.9 hectares, being 60% of the licensed area for which the Aquaculture Licence was granted.

1368 Case C-535/18 IL et al v Land Nordrhein-Westfalen, Judgment of 28 May 2020, §81 et seq.

1369 §89 – “it is for the referring court to verify whether the file to which the public had access before the project at issue was approved satisfies all of

the requirements stemming from Article 6(3) of Directive 2011/92, read in conjunction with Article 5(1) and (3) of that directive, ..”

1370 Vol 1 p154.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

900. There is no suggestion that the limits of reasonable requirement of a private developer in this case were such that it could not have been expected to provide adequate preliminary design information. It is difficult to see that such a suggestion could have been made given the terms of the DAFM Structural Design Protocol and the reality that whether earlier or later, the developer must finalise its own design in order to construct it. A fortiori, it can be expected to provide a preliminary design to the standard contemplated by the Structural Design Protocol and, as a distinct matter, to the standard required to enable comprehensive EIA.

901. On the particular facts of this case it does not seem to me necessary to resolve questions of the standard of review on which the court must verify the adequacy of the information supplied. At risk of excessive recapping, here the explicit and uncontradicted advice of ALAB’s Technical Expert, Dr Saunders, as to the risk of escape was that that the information available to him as to cage specification and design:

• was such as to prevent “any objective assessment of installation’s suitability for the potentially challenging

location”.

• left him “unable to comment further on either the installation’s structural suitability for the location, or its

ability to withstand a weather event of a particular magnitude”.

• left him, “unable to evaluate the risk of fish escapes from the Shot Head site”.

Importantly, in the light of that advice of its own Technical Expert, ALAB made no judgment to the contrary. The net position therefore in EIA is a finding that the information supplied was insufficient to enable comprehensive EIA of the risk of escape.

902. The solution to this inadequacy of information adopted by Dr Saunders and ALAB was the same: to let this assessment and evaluation to be done by DAFM after development consent had issued. In my view that was a legally flawed solution in that it abdicated, and so failed to fulfil, ALAB’s duty to perform a comprehensive EIA.

903. Though it is not essential to that conclusion, I may add that this abdication is amplified in that it was by reference to a Protocol Dr Saunders had not seen. Though ALAB did have it, ALAB did not record any analysis of its content.1371

904. As has been seen, the Aquaculture Licence, instead of constituting the required preliminary design drawings “an integral part of the licence … specified as approved drawings in a schedule to the licence” as contemplated by the Protocol, identified no such drawings at all. Indeed no 16+2-cage drawings exist – at least in the Aquaculture Licence process. Despite the 2014 advice by the Department of the Environment as

1371 ALAB Determination §2.8.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

to the issue of structural integrity,1372 I have been directed to no “initial assessment and sizing of the proposed set of nets, pens and mooring system to withstand the maximum design loads acting on the farm”. Schedule 2 to the licence instead required compliance with the Structural Design Protocol and DAFM approval. The difficulty with that is that ALAB’s approval of the preliminary design and appendage of the preliminary design drawing to the licence is a basic premise of that Protocol. ALAB in effect, skipped the first stage of the “two-stage process” which is the method of the Protocol. It is not difficult to infer that the Protocol envisages a first stage which enables ALAB, in the relevant Boland condition, to decide “in essence” what is required such that the detailed design stage is of “a very limited and technical nature” such as to “implement that which has already been decided in essence.” The Aquaculture Licence itself breaches the Protocol with which it requires compliance.

905. That this is so is confirmed by the requirement in the Protocol that the preliminary design shall

consider “in sufficient detail the provisions outlined in Sections 4, 5, 6, 7 and 8 of this protocol, to allow the

drafting of a set of site specific drawings.” I have summarised that content above. While this description of sufficiency is unobjectionable in itself (though prompting the question – what is the criterion for sufficiency?), the law inevitably superimposes a criterion of sufficiency to enable ALAB to perform its duties both generally (including by reference to the factors identified in s.7 of the 1997 Act, which include “measures for preventing escapes of fish”) and, as a distinct matter, as to comprehensive EIA, inter alia as to risks of fish escape and its environmental consequences.

906. It seems necessary to repeat that §2.1 of the Structural Design Protocol explicitly requires that the preliminary design address all matters listen in §§4, 5, 6, 7 and 8 of the protocol and that they inform the drawings to be submitted in the licence application and appended to it. Obviously, how far “down the road” the preliminary design must go as to these matters in informing ALAB’s determination of the licence application and how much can be left to the DAFM to finalise afterwards will be a matter of some judgment for ALAB bearing in mind that ALAB must:

• decide the design and specification issues “in essence”, leaving only “very limited and technical” details to

decision by DAFM.

• have sufficient information as to design and specification to enable it to assess the risk of escape in a comprehensive EIA.

907. What ALAB cannot properly do, it seems to me, is omit to decide the design and specification of the cages “in essence” and to leave even that decision to DAFM, in effect skipping the preliminary design stage which the protocol itself requires it to address. To do so is to fail to comprehensively assess in EIA the risk of escape. That it did so in the present case is amply demonstrated by the content of Schedule 2 of the Licence and §6.1.3 of the Licence in its failure to engage with Dr Saunders’ observation that for want of cage specifications and design, “unable to evaluate the risk of fish escapes from the Shot Head site”.

1372 Set out above.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

908. Looked at otherwise, and to paraphrase the Court of Appeal in Krikke,1373 ALAB failed in its Boland condition to provide the “criteria by which [MOWI and DAFM] can reach agreement." That may seem a strange observation given ALAB cites the Protocol and given the content of the Protocol generally. But it seems to me correct given ALAB’s failure to carry out the tasks specifically envisaged by that Protocol as necessary to provide the criteria, by way of preliminary design, required to enable the DAFM to perform its task of approval of detailed design. Not for the first time, I respectfully observe that the incorporation of documents by reference in permissions and licences has its pitfalls and such documents must be carefully considered before incorporation.

909. While I am of the view that the Boland condition in this case is deficient as failing to provide the necessary criteria for the decision delegated to DAFM, one can look at the same issue from a slightly different perspective. Bearing in mind the view of McKechnie J in Kenny1374 that whether a Boland condition

“is intra vires is a matter of degree and depends on the nature of the matter left for resolution ….” I cannot

see that the matters left to DAFM were matters of “technical detail” where they were necessary to enable ALAB to evaluate, and their absence was such as to render it “unable to evaluate”, “the risk of fish escapes from the Shot Head site”. There is simply no doubt that the risk of escape had to be a central consideration in EIA and required comprehensive assessment – which ALAB did not perform.

910. At this point it is convenient to address MOWI’s reliance on Alen-Buckley1375 to the effect that, for Boland conditions to be impermissible, the matters they leave to later agreement “would need to be of such central importance or magnitude so as to be an abdication of the Board’s statutory duties.” It might be thought, taking it out of context and though I do not consider that such was the intent of the judgment, that the formula used in the High Court in Alen-Buckley represented an expansion of the scope of permissible Boland conditions beyond the “technical matters of detail”, to anything short of something of “such central importance or magnitude”.

911. But whatever view one takes of this point, the High Court’s decision in Alen-Buckley is perfectly understandable in its own terms. The Boland condition impugned in that wind farm case was criticised as allowing for the construction of a grid connection and haul routes. As a matter of interpretation, the High Court held that it did not permit such works so the point fell away. The only other complaint as to the impugned condition appears to have been the exclusion of the public from the process for reaching agreement pursuant thereto. Haughton J summarised the view he took as follows:

1373 Krikke v Barranafaddock Sustainable Electricity Ltd [2021] IECA 217 (Donnelly J, 30 July 2021) §59. 1374 Kenny v An Bord Pleanála & Ors [2001] 1 I.R. 565, §9.

1375 Alen-Buckley v An Bord Pleanála [2017] IEHC 541. While the precedential value of Supreme Court determinations is very limited, the determination

in this respect represents an entirely orthodox view of the law. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“In summary it is clear that the Board is entitled to impose conditions and to leave matters over for future agreement between the developer and the planning authority and any challenge to such derogation would need to be based on an alleged abdication of statutory functions of the Board – an allegation which is not made in the present case. Furthermore, for the most part this argument appeared to be based on the false premise that planning permission was granted by the Board for the grid connection/haul route works, which I have already found was not the case. In any event, the point seems to be moot in that the sole complaint related to impermissible delegation relates to public participation. In circumstances where the public have actively participated in the current development process, and have appropriate opportunities to participate in any existing or future applications for planning permission for the grid connection and haul route works, it seems to me that the submission of public exclusion from the planning process is not borne out.”

Alen-Buckley is not authority for any more expansive than orthodox view of the proper scope of Boland conditions.

912. MOWI also cites the Supreme Court’s refusal of leave to appeal in Alen-Buckley.1376 It cited the caselaw from Boland on “which establishes that as long as conditions relate to “technical matters of detail”, they are consistent with the EIA Directive.” The Supreme Court made clear – “as long as” – that Boland conditions are confined to “technical matters of detail”.

913. However, lest there be doubt, I am of the view that the papers before me – perhaps exemplified in the international objective of 100% containment and the concerns prompting it – repeatedly and indisputably disclose that the risk of occurrence of salmon escape and its consequences for wild salmon and the resultant necessity to ensure that their cages are sound in design and construction, not least in an exposed site close to shore, are indeed matters of “such central importance or magnitude” as to require ALAB to itself determine, and not delegate the determination of, at least minima by way of proposals for actual cage design and construction.

Escapes – ALAB argument - The Minister’s EIA – Cage Design & Specification

914. ALAB courageously, if somewhat diffidently, argued that the position as to ALAB’s EIA was satisfactory as the Minister doing his EIA in June 2015 had, it is said, had regard to the Structural Design Protocol published in April 2016 and so was “obviously satisfied that the details that were provided met the

standard for preliminary design phase”.1377 I put the argument only slightly inaccurately, but I think fairly, for

the purpose of highlighting the anachronistic illogicality of the argument. What was argued in fact was that

1376 [2018] IEDSCT 45 §23. 1377 Transcript Day 10 p67.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“when the Minister completed his EIA, he was aware of what he was intending the protocol to contain.” The Minister’s EIA does record satisfaction with the preliminary design from a structural point of view and refers to the Department being the final stages of introduction of a Structural Design Protocol. But it also but clearly records that the protocol had not yet been “approved”. It does not record that this draft had been published or that participants in the process had been informed of its content or advised that EIA would be done in its light. Neither is the draft protocol in its form and content in June 2015 described in the 2015 EIA and we are not told what is the content of any relevant differences between it and the protocol actually approved the following year. Leaving that aside the 2015 EIA contains no analysis of the structural design proffered by reference to the criteria for preliminary design even as set out in that draft in the form it took in 2015. And, of course, the design had changed – notably as to numbers of cages – by the time ALAB considered the matter as compared to that which the Minister had considered.

915. However, and more fundamentally, ALAB was obliged to conduct its own EIA de novo.

Escapes – EIA – Information on the 2014 Escape and as to Cage Design & Specification – Decision.

916. While I here give only a summary, in my view, as explained above, ALAB’s EIA was deficient and so its determination must be quashed as ALAB failed, i. to reconsider, as an element of comprehensive EIA whether to renew its s.47 Notice to DAFM as to documents relating to the 2014 Escape or otherwise acquire such of those documents as were in the public domain. This not least in light of ▪ events at the oral hearing and of the Oral Hearing Report – including information to the effect that at least some significant documents in the possession of DAFM relevant to the 2014 Bantry Bay fish escape had come into the public domain, ▪ Dr Saunders’ Final Report of December 2020 as to the importance of escape risk to EIA, the importance of the 2014 Escape to inform EIA in that regard and that the non-availability of relevant documents in the possession of DAFM was both important and unfortunate, ▪ its having earlier decided that information as to the 2014 Escape was material, and by way of formal statutory determination that the DAFM reports on that escape were necessary to its

decision.

ii. to conduct a comprehensive EIA as to the risk of escape – in that its Technical Advisor, Dr Saunders in his final report recorded himself as objectively unable to assess that risk and ALAB made no judgment to the contrary. Its delegation of that task to the DAFM via its Structural Design Protocol reflected the same failure to conduct a comprehensive EIA as to the risk of escape.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

SEALS – ESCAPE OF SALMON AND ACOUSTIC DISTURBANCE

Seals – AA & EIA Aspects & Introduction

917. The Harbour, or Common, Seal1378 is listed in Annex II of the Habitats Directive as a species whose conservation requires the designation of their habitats as Special Areas of Conservation. It is listed in Annex V as a species whose taking in the wild and exploitation may be subject to management measures. It is not listed in Annex IV as a species requiring strict protection under Article 12 of the Habitats Directive.

918. Bantry Bay contains, at Glengarriff Harbour at the head of the bay about 9 km from the proposed Shot Head Salmon Farm Site, the largest haul-out location for common seals in south-west Ireland1379 and Glengarriff Harbour is an “important site for harbour seals”.1380 Accordingly, Glengarriff Harbour and Woodland SAC1381 is designated as such for, inter alia, its common seal population and its conservation objectives include avoidance of adverse anthropogenic effect on seals.

919. Common seals have been known to prey on salmon in salmon farms. In doing so, they may tear cages to get at the salmon, such that salmon may escape into the wild. So, any deterrence of seal predation is directed, at least primarily, at the risk of salmon escape and thereby at both the commercial and environmental risks of such escape – the latter as described above.1382 Deterrence of seal predation can be attempted by various means – no doubt not mutually exclusively – including killing seals, use of antipredation nets and use of seal scarers. Seal scarers work by making noise – they are more technically termed

“Acoustic Deterrent Devices” or “ADDs”. I will refer to “seal scarers” and “ADDs” interchangeably. The

controversy in this case concerns seal scarers.

920. Accordingly, seals are relevant to environmental assessment of the proposed Salmon Farm in two distinct but connected respects:

• First, as to EIA, they pose a risk of causing escapes of farmed salmon by damaging the cages.

• Second, as to AA, there is at least some prospect that, in mitigation of that risk of escape, seal scarers will be used which, the Sweetman Applicants suggest, would cause acoustic disturbance (within the meaning of the Habitats Directive1383) of the seals, which are a conservation interest of the Glengarriff Harbour and

1378 Phoca vitulina. The names will be used interchangeably here and all references in this judgment to “seals” are to this species.

1379 Technical Advisor Saunders Final report p71 citing EIS Vol. 1, §2.1.2, p32. 1380 Coram, infra. 1381 Special Area of Conservation designated for purposes of the Habitats Directive.

1382 Genetic damage to wild salmon by interbreeding with farmed salmon, competition with wild salmon for resources and increased risk of sea lice

infestation of wild salmon.

1383 Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Woodland SAC, such as, it is suggested, to require consideration of that prospect in AA of the proposed salmon farm.

One can contrast these two issues as relating, respectively, to the prospect of damage by seals (an EIA issue as to escape of farmed salmon) and, in the cause of mitigating that risk of escape, the prospect of damage to seals (a potential AA issue as to acoustic disturbance by seal scarers).

921. As will be seen, the pleaded grounds relate only to the AA issue as to the prospect of adverse impact to seals by seal scarers. However, to consider the AA issue it is also necessary also to consider the first, EIA, issue. That is because the prospect of the use of seal scarers arises as a reaction to and a function of the risk of salmon escapes due to seal predation. The greater the risk of salmon escape by seal predation then, at least generally, the greater the need – commercial and environmental – for anti-predation measures such as seal scarers. Absent an assessment in EIA of the risk of occurrence of such predation, it is at very least difficult to assess the likelihood that mitigation measures such as seal scarers to prevent or minimise seal predation and consequent salmon escape will be required. Indeed, Dr Saunders links these issues as follows:

“…… it would be advantageous to establish whether seal damage has been an issue with the farms already operating in Bantry Bay, as this may provide some insight in any possible future requirement for deterrent devices.”1384

922. Though – or more accurately because – they are intertwined in particular on the facts of this case, it is important at all points to bear in mind that distinction between the EIA issue of mitigation by seal scarers of the risk of escape on the one hand and, on the other, the possible AA issue of impact of seal scarers on the seals thus scared away.

Seals – Pleadings

923. While the SWI Grounds, in describing the licensing process, refer to seals it is not apparent that any of SWI’s grounds of challenge relate to either the risk of salmon escape by reason of seal predation or to the risk to seals by reason of steps taken to mitigate that risk of escape. The IFI Grounds do not mention seals.

1384 Saunders Final Report 8 December 2020 §9.8.1 p85.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

The Sweetman Applicants’ Grounds

924. The Sweetman Applicants plead core grounds as to seals that ALAB’s Aquaculture Licence decision breaches Article 6(3) of the Habitats Directive as transposed by the Habitats Regulations 2011 as follows:

• Art 42(6)1385 – in that ALAB failed to determine that AA was required as it could not be excluded, on the basis of objective scientific information following screening that the project, … will have a significant effect on the Glengarriff Harbour and Woodland SAC.1386

• Art 42.(1)1387 – in that ALAB failed to do a proper AA screening, in view of best scientific knowledge and in view of the conservation objectives of the site, of whether the project, was likely to have a significant effect on that SAC.1388

• Art. 6(3) of the Habitats Directive – in failing to contain complete, precise and definitive findings and conclusions capable of removing all reasonable scientific doubt as to the effects of the proposed works – in incorrectly adopting an AA screening test of “no significant impact” which does not establish certainty beyond reasonable doubt.1389

• Art 42.(1) – in that ALAB’s AA screening took account of mitigation – measures intended to avoid or reduce harmful noise effects of the project on seals foraging outside that SAC.1390

• Art 42.(1) & Art 2(1)1391 o by granting development consent without being certain that it guarantees that Acoustic Deterrent Devices permitted outside the current consent process will not adversely affect the integrity of the Glengarriff Harbour and Woodland SAC.1392 o in failing in its AA to make an explicit and detailed statement of reasons, capable of dispelling all reasonable scientific doubt as to the effects of the Proposed Development, for rejecting, ▪ Mr Coram’s findings as to the possibility of acoustic deterrents causing hearing damage to seals from the Glengarriff Harbour and Woodland SAC.1393

925. It will be noted that the pleaded grounds above relate only to AA and risk to seals – not to EIA and risk by seals. The Sweetman Applicants did plead failure to consider in AA screening the impact of accidental

1385 Of the Habitats Regulations 2011 (S.I. No. 477 of 2011). 1386 PS Core Ground 8.

1387 of the Habitats Regulations 2011 (S.I. No. 477 of2011). Art 42.(1) requires AA Screening “to assess, in view of best scientific knowledge and in view of

the conservation objectives of the site, if that plan or project, individually or in combination with other plans or projects is likely to have a significant

effect on the European site”. Article 42(6) requires AA “if it cannot be excluded, on the basis of objective scientific information following screening under

this Regulation, that the plan or project, individually or in combination with other plans or projects, will have a significant effect on a European site.”

1388 PS Core Ground 9. 1389 PS Core Ground 11. – citing Case C-236/01 Monsanto Agricoltura Italia and Others [2003] ECR I-8105. 1390 PS Core Ground 12. 1391 of the Habitats Regulations 2011 (S.I. No. 477 of 2011). 1392 PS CG12. 1393 PS CG13.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

escapes of farmed salmon in terms of their effect on wild salmon. But for reasons explained early in this judgment, I do not see that that issue arises in AA – at least on the evidence adduced.

926. I may as well say now that, in my view, the plea is misconceived that ALAB breached Art 42.(1) by conducting AA screening which took account of measures intended to avoid or reduce harmful noise effects of the project on seals. This plea confuses the EIA and AA processes. Seal scarers are mitigation only of the risk, assessable in EIA but not in AA Screening, of escape of salmon due to seal predation. The only risk relevant to in AA Screening is that posed to the seals by the seal scarers themselves. There is no proposal to mitigate that risk and so question arises of taking mitigation into account in AA Screening.

927. By way of particulars, the Sweetman Applicants plead that:

• MOWI’s NIS did not consider seals – stated in Mr Coram’s report to be potentially impacted by the development.1394

• ALAB breached Art 46.(1) of the Habitats Regulations 2011 in failing to determine that AA was required as to seals, given Mr Coram’s report.1395 (Note this is an additional plea – by reference to Art 46.(1) – to the pleas in the Core Grounds. I mention this only for clarity. It does not mean the plea will not be considered.)

• that Mr Coram found that hearing damage to seals from the SAC straying outside it by seal scarers and other devices could not be excluded, should have resulted in seals being screened in for AA1396 (in submissions they cite Holohan1397).

• ALAB failed to properly screen for AA as to possible effect on seals and the Glengarriff Harbour and Woodland SAC.1398

• ALAB failed in AA to make complete, precise and definitive findings and conclusions capable of removing all reasonable scientific doubt as to the effects of the proposed works.1399

• ALAB applied an AA screening test of ‘no significant impact’ – which does not establish certainty beyond reasonable scientific doubt. o I observe that this plea confuses the substance of the conclusion with the standard of its proof.

928. By way of further particulars of these pleas, the Sweetman Applicants also plead1400 contravention of Article 6(3) of the Habitats Directive in that:

1394 §E2.26. 1395 §E2.27. 1396 §E2.86 & 87. 1397 Case C-461/17 Holohan v An Bord Pleanála, Judgment of 7 November 2018. 1398 §E2.27 & 29. 1399 §E2.30. 1400 §E2.31.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• ALAB unlawfully took account in AA Screening of measures intended to avoid or reduce harmful noise effects on seals foraging outside the Glengarriff Harbour and Woodland SAC, of which seals are a conservation interest.

• Contrary to Holohan, and despite Mr Coram’s1401 report that ADDs pose a risk to seals foraging outside the SAC, the Technical Advisor’s Final Report relies on the DAFM EIA to state that this risk will not be assessed in the process before ALAB but will be assessed under a separate consent process in the future.

• Contrary to Case C-323/17, People over Wind, ALAB’s minutes of 20 March 2020 record ALAB’s assumption in AA screening that the impact would be mitigated by NPWS guidelines on marine noise.

For the avoidance of doubt, I read all these pleas as referring and referring only, to the risk allegedly posed to seals by ADDs.

929. The Sweetman Applicants also plead1402 contravention of

• Article 6(3) of the Habitats Directive in failing to make an explicit and detailed statement of reasons for rejecting Mr Coram’s findings as to the possibility of ADDs causing hearing damage to seals from the Glengarriff Harbour and Woodland SAC – reasons capable of dispelling all reasonable scientific doubt as to those effects as required by Holohan.

• S.40(8)(a) of 1997 Act, which required that ALAB’s determination state the main reasons and considerations on which it was based.

The Pleaded Opposition

930. Beyond traverses ALAB pleads1403 that

• ALAB’s Determination screened out seals from AA, having adopted what it calls the Coram “Seal Screening

Report” (sic)1404 to the effect that “on the basis of scientific evidence” it is concluded that “the operation of

a fish farm at Shot Head has no potential for significant effects and it is not likely to negatively impact the conservation status of the population of harbour seals within the Glengarriff Harbour and Woodland SAC.”

• The Coram Seal Report, o did not take into account any NPWS guidelines on marine noise.

1401 The plea refers to Dr Gittings but clearly should have referred to Dr Coram. 1402 §E2.32. 1403 §§50, 59, 64 et seq & §73 et seq. & 86. 1404 See ALAB’s Determination 29 June 2021 §§2.16, 5.3 & 7.2.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o said that seal scarers are “in widespread usage” and that any hearing damage to seals by seal scarers would be temporary. o did not screen in the seals for AA. o In its observation that the “possibility of acoustic deterrents [seal scarers] causing hearing damage to individuals from the … SAC cannot, however, be excluded”, relates to a possible future need for seal scarers as to which Mr Coram observes that “the need for the use of … seal scarers will be assessed, if the licence is granted”.

• The use of seal scarers was not certain and not proposed in the aquaculture licence application nor permitted by the Aquaculture Licence.

• Any future use of seal scarers would require permission under s.42 of the Wildlife Act 1976 as amended and their impact on seals would be assessed in that process. Accordingly ALAB’s Technical Advisor’s Final Report1405 concluded: “We have therefore not considered the use of ADDs in this report, beyond acknowledging that they may be necessary in the future.”

• Mr Coram’s report, which ALAB adopted, sets out the reasons for ALAB’s screening out AA of seals.

931. It is not unfair to observe that, at trial, the opposition to this ground essentially boiled down to the propositions that:

• the “project” for which the aquaculture licence was sought does not include the use of seal scarers and so no obligation arose to do AA of the risk of acoustic adverse impact on seals by seal scarers. In this regard reliance was placed on FitzPatrick.1406

• Any future use of seal scarers would require permission under s.42 of the Wildlife Act 1976 as amended and any impact on seals would be assessed that process.

932. MOWI pleaded that

• It had “indicated in the 2011 EIS that Acoustic Deterrent Devices (“ADDs”) may have to be considered in the future and that they would be the subject of a separate statutory consent process;

• It had stated in the 2011 EIS that it did not intend to use acoustic devices.

• It had clarified to ALAB in November 2015, that the only noise from the cages would be from feed spreaders and vessels.

• ALAB expressly relied upon this material at §7.2 of its Determination

• “at this point in time” 6 of ALAB’s fish farms in the State are Aquaculture Stewardship Council certified and

“a condition of such certification is not to use acoustic devices given the threat they may pose to the hearing of mammals. Two of these six sites are in Bantry Bay.”

1405 §9.13 (page 96). 1406 Fitzpatrick v An Bord Pleanála, Galway County Council & Apple Distribution [2019] IESC 23, [2019] 3 IR 617.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

ALAB’s Determination – as to AA

933. ALAB’s Determination of the Appeal, dated 29 June 2021, included Part 5 headed “Appropriate Assessment”. Part 5 largely duplicates its AA Conclusion Statement of May 2021. I confess that I am unclear what ALAB intends to be the legal relationship between and difference between the two documents or why there are two. In Part 5 of its Determination, dated 29 June 2021 ALAB,1407

• Recorded that “in accordance with the Recommendations arising from the Oral Hearing Report” it had undertaken “detailed evaluation of the potential threats to seal populations” being qualifying interests of Glengarriff Harbour and Woodland SAC. Specifically, it commissioned the Coram report.

• Deemed the Coram Report a “Seal Screening Report” and accepted its conclusion of the Coram report1408 that the proposed fish farm is not likely to have any deleterious effect on the qualifying features of the Glengarriff Harbour and Woodland SAC and there is no potential for significant effects and it is not likely to have a significant effect on that SAC.

• Recorded that the Marine Institute “carried out” AA Screening of Aquaculture Activities in Outer Bantry Bay and that ALAB had regard to the Marine Institute’s June 2018 and September 2020 “Screening Reports”1409 and accepted their Finding of no Significant Impacts for Glengarriff Harbour and Woodland SAC and determined that the proposed fish farm has no potential for significant effects and it is not likely to have a significant effect on that SAC.

934. I observe of that Determination dated 29 June 2021 that,

• it did not record or recognise that in February 2018, ALAB had determined the scope of the AA, screened out risk to seals and required an NIS in terms which did not include consideration of risk to seals.

• its reference to “Screening Reports” is a reference to the Marine Institute Matrix – which was not in existence when, in February 2018 it screened out AA of risk to seals.

• It did not refer to or address Mr Coram’s advice that acoustic effects of seal scarers on seals of the SAC could not be excluded.

1407 I have edited what follows to confine it to materia relating to seals of the Glengarriff Harbour and Woodland SAC.

1408 In fact the reference is to “all of the scientific literature in the material referred to in Part 2 of this Determination” but that includes Mr Coram’s report

– which report is the relevant literature as to seals. 1409 The phrase isa that used by ALAB.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Seals – the Evidence & Analysis Thereof

EIS 2011, EIA 2015 & Oral Hearing Report 2017

935. The 2011 EIS says1410 of common seals that “The closest haul-outs are approximately 5km from the proposed Shot Head site area and there is a likelihood that seals will visit the site.” Dr Saunders states: “From the EIS it is difficult to determine the anticipated or predicted risk of nuisance from seals”. 1411

936. MOWI, in these proceedings, not so subtly reconfigures its account in the EIS. It says1412 accurately,

that it “never sought permission to use seal scarers” but also says that “the 2011 EIS .. stated that it did not

intend1413 to use ADDs. The 2011 EIS did indicate that ADDs may have to be considered in the future and that they would be the subject of a separate statutory consent process.” In fact, the 2011 EIS says: “It will therefore be necessary to assess whether or not anti-predator nets or even seal scarers will be needed to protect the stock from seal attack early in the development of the site,1414 if the licence is granted.”1415 This contradicts MOWI’s plea1416 that it had “stated in the 2011 EIS that it did not intend to use acoustic devices …” and puts in illuminating and unflattering perspective its plea1417 that it had in its 2011 EIS merely

“indicated” that ADDs “may have to be considered in the future”. In fact, the EIS did not say that MOWI “did

not intend to use ADDs”. Nor did it depict their consideration as merely a “future” possibility (“may have

to”). Rather it said that consideration of their use would be “necessary” – and that “early in the development

of the site”.

937. The EIS also says: “Although there are many haul-out sites in the bay, they are at a considerable distance from the proposed site. Whilst there is no doubt that seals will visit the site on occasions, the

distances to the haul-out sites renders the risk of impact low.”1418 For clarity, I should note that the impact in

question here is that of disturbance of seals in situ in their haul-out sites – as to which there is, at least in these proceedings, no issue. This observation does not relate to disturbance of seals to prevent predation at the salmon farm.

938. The DAFM EIA of June 2015 states1419 that noise effects “potentially” include effects by ADDs “which could potentially impact on marine mammals. …….. ADDs if proposed … would require separate statutory consent …. at which time a full evaluation of the effects would be carried out and any mitigation measures

1410 p172 §3.4.11. 1411 Technical Advisor Saunders Final Report December 2020 p71. 1412 Affidavit if Catherine McManus 6 July 2022, §37. 1413 Emphasis added. 1414 Emphasis added. 1415 2011 EIAS p172 §3.4.11. 1416 Statement of Opposition §124. 1417 Statement of Opposition §123. 1418 EIS Vol 1 §5.3.5. p252. 1419 p14.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

would be specified, as required”. Dr Saunders, in considering “Noise Impacts” notes that content of the DAFM EIA and states: “We have therefore not considered the use of ADDs in this report, beyond acknowledging that they may be necessary in the future.” 1420

939. The precise nature of that “separate statutory consent” process is not identified in the DAFM EIA but has been since identified as an application for a licence pursuant to s.42 of the Wildlife Act 1976.1421 While assured that in that application “a full evaluation of the effects would be carried out”, I have not been referred to the regulatory basis of that evaluation. Looking at it from an EU Law perspective, it doesn’t seem to be a matter merely (thought the word “merely” may give the wrong impression) of species protection under Art 12 of the Habitats Directive and a derogation from that protection under Art 16 of the Habitats Directive. That is because the seals in this case are not listed in Annex IV of the Habitats Directive as requiring strict protection. Rather, what the DAFM EIA envisages here is a statutory consent procedure as to potential anthropogenic disturbance of a species for which the Glengarriff Harbour and Woodland SAC is designated and the conservation objectives of which require the avoidance of anthropogenic adverse effect on that species. That suggests that it will require AA or at least AA screening. Of course, the result of neither can be assumed at this point. AA may, or may not, require refusal of a licence to use seal scarers. As the proposal to deploy seal scarers will have arisen by reason of a risk of escape of farmed salmon to the wild, a prohibition of their deployment as a result of AA would bear on the capacity of MOWI to prevent such escapes. Accordingly, on this analysis and as matters stand, EIA in any remitted process, of the risk of salmon escapes due to seal predation, must proceed on the precautionary and explicit footing that seal scarers will not be used.

940. As stated, MOWI Pleaded in answer to this issue of acoustic seal scarers that it “had clarified to ALAB in November 2015, that the only noise from the cages would be from feed spreaders and vessels.” Indeed it had – in its Reply dated 20 November 2015 to the Appeals.1422 But, regrettably, this plea did not draw any attention to the fact that, in the same document, the following passage appears:

“It will therefore be necessary to assess whether or not anti-predator nets or even acoustic deterrent devices will be needed to protect the stock from seal attack early in the development of the site, if the licence is granted.”1423

1420 Saunders Final Report 8 December 2020, §9.13 1421 S.42 is lengthy. It suffices here to record the relevant elements:

“(1) Where serious damage is being caused by … protected wild animals to ….. (h) aquaculture installations, .. the Minister (authorise) such steps,

including …. the capture or killing of any such … wild animal, as he thinks appropriate to stop the damage.

(3) Where damage described in subsection (1) of this section is being caused, the owner or occupier of the property to which the damage is being

caused, … may apply to the Minister for a permission under this section.

(6)(c) .. the Minister may grant the permission subject to conditions which may include one or more of the following:

(i) that any scaring, capture or killing pursuant to the permission is to be effected by a specified means,

(v) that the total number of any particular species of … protected wild animal captured or killed pursuant to the permission shall be limited to such

number as is specified in the permission.” 1422 §9. 1423 §3.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

941. The chairperson of the oral hearing1424 in 2017 was exercised to point out and recommend:

• that “The Board should make every effort to consider the potential impacts of large-scale farmed salmon escapes.” That implies the necessity of consideration inter alia of the risk of occurrence of such escapes due to seal predation.

• as to seals that “it is firmly established in the jurisprudence of the ECJ/CFEU that a national competent authority must establish beyond reasonable scientific doubt that the project will not impact these species to the extent that it affects the integrity of the site having regard to its conservation objectives.”

• MOWI’s statement that it may be necessary to assess the requirement for anti-predator nets or acoustic deterrents.

• The Board should conduct a desk-top study of the potential impacts of the Salmon Farm on the seals of the Glengarriff Harbour and Woodland SAC.

The chairperson’s reference to the EU caselaw is, of course correct – and has since been confirmed as such by Holohan. So too has the strictness of that test, though absolute certainty is not required: see Kokott AG in Eco Advocacy to the effect that “the criteria governing the screening must be just as strict as the criteria for the appropriate assessment itself”.

Coram Report February 2018 & Observations Thereon

942. On 13 December 2017, ALAB resolved to accept the recommendations made by the chairperson of the oral hearing and asked Dr Saunders to prepare terms of reference for ALAB approval to bespeak the recommended desk-top study as to impact on seals. On the redacted minute exhibited, there is no evidence of action by ALAB on the chairperson’s recommendation of every effort to consider the impact of salmon escapes. There is no evidence before me of those terms of reference or their approval by ALAB. By ALAB’s next meeting however, the desktop study was to hand. It was by Mr Alex Coram of St Andrews Marine Research,1425 is on the headed paper of Dr Saunders and is dated 1 February 2018. I will concentrate on Mr Coram’s consideration of ADDs as noise sources but background is required.

What was Mr Coram’s Report About?

943. The Coram report is entitled as a “Technical Advisor's Report: Supplementary Briefing Note”. It is not entitled as a “Common Seal Impact Assessment” or as a “Seal Screening Report” – both of which are

1424 Report 8 November 2017.

1425 Mr Coram is in some papers in the case, referred to as “Dr Coram”. Mr Coram does not, as he should have done, state his expertise in his report. I

infer, perhaps incorrectly, that “St Andrews Marine Research” is attached to St Andrews University in Scotland. Nothing turns on the issue as his expertise

was not challenged. No doubt that was sensible. Mr Coram is co-author of a number of the cited academic papers. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

descriptions later attributed to it by ALAB in its Determination.1426 Mr Coram does not refer to the concepts of AA or AA Screening. He does not cite the standard of scientific certainty of absence of adverse effect required in both.

944. Mr Coram does record the protection of seals by the Habitats Directive as requiring designation of their habitats as SACs and that:

• the oral hearing chairperson had recommended a desk-top study of the potential impacts of the Salmon Farm on the seals of the Glengarriff Harbour and Woodland SAC.

• the method and purposes of his report are to o draw on scientific literature to further evaluate impacts on seals.

o determine the level of threat to the conservation status of the SAC.

o establish whether further material is required to determine whether the proposed fish farm will have an unacceptable impact on seal conservation status.

945. It is important to say that Mr Coram considers whether seal predation at the Site is likely and whether it can be excluded that, if seal scarers are used, they will causing hearing damage to seals from the SAC. He omits, perhaps very properly as beyond his expertise – though he does not say so – an assessment of the issue intermediate to those two considerations: the likelihood that seal scarers will or will not be used.

Coram Report – Background, The SAC and the likelihood of Seal Predation at the Site

946. Mr Coram records that:

• the Eastern Atlantic Harbour Seal is listed by the IUCN1427 as ‘least concern’, meaning that its population size and geographical range are not endangered, vulnerable or near threatened. In Ireland its population status is favourable.1428

• the South-West appears to be a significant stronghold and important site for harbour seals in Ireland, and Bantry Bay stands out as the most significant local population. The population estimates since 2005 suggest that a stable population size may have been reached in the Bay. Current pressures on the population are low.

• the NPWS assesses the threat to seals by aquaculture and by noise as each of low importance1429

• the DAFM EIA states that the need for ADDs will be assessed if the licence is granted.

947. Mr Coram records that,

1426 ALAB Determination 29 June 2021 §2.16. 1427 International Union for Conservation of Nature. It is associated with UNESCO.

1428 Citing NPWS 2008. "The Status of EU Protected Habitats and Species in Ireland" 3:375-88. Mr Coram’s bibliography cites quite a number of papers

the titles of which suggest supportive content as to the status of the seals in Ireland. 1429 Citing NPWS 2008. "The Status of EU Protected Habitats and Species in Ireland" 3:375-88.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• the SAC is designated pursuant to the Habitats Directive.

• the nearest known seal haul-out site,1430 Orthans Island, is outside the SAC and about 4km from the Shot Head site.

• the conservation objectives for the SAC1431 identify conservation target 5 as that “Human activities should occur at levels that do not adversely affect the harbour seal population at the SAC”.

• the conservation objectives set out in the NPWS Supporting Document – Marine Species1432 for the SAC state, as to target 5, that “Proposed activities or operations should not introduce man-made energy (e.g.

aerial or underwater noise, … ..) at levels that could result in a significant negative impact on individuals and/or the population of harbour seal within the site”.

948. Mr Coram records that evidence1433 suggests that seals are generalist feeders. Their diet is highly varied and opportunistic. They probably consume prey in proportion to its availability. Salmonids have been infrequently detected in seal diet in SW Ireland – in low proportion compared to more important prey species. In 2006/2007, as part of a larger project, 2 tagged seals1434 in Bantry Bay were tracked. One was based in Adrigole Harbour, the other in the SAC. The latter moved between the SAC and Castletownbere and regularly passed through the area of the proposed Shot Head site. Both “seals made extensive use of the existing Roancarrig farm site … This marked association with the existing fish farm is similar to …. data … observed In Scotland, and is probably due to foraging opportunities at the site.1435 These opportunities may be aggregations of wild fish found around farms … or may be direct predation on the farmed stock.1436 Mr Coram notes that the EIS states that it is likely that seals will visit the site. He concludes that the home/foraging range of seals from the SAC, and nearer haul-outs, is likely to include the area of proposed salmon farm.

949. Considering “Interaction with Aquaculture” Mr Coram states that “Scientific literature on the finescale interactions between seals and aquaculture is relatively scarce.” For all that, he says:

“In Scotland between 72 and 86% of salmon farm sites experience some level of seal depredation.1437 Cronin et al. (2014b) suggest that all Irish salmon farms experience some degree of seal depredation, especially in the south-west, although they note that the problem is unquantified. Farms on the west coast were estimated to have lost 5 tonnes of salmon each to seal depredation (presumably per growth cycle).”

1430 Though marine mammals, seals “haul-out” onto land for a variety of reasons and tend to do so in specific identifiable sites.

1431 NPWS. 2013. "Conservation Objectives Series – Glengarriff Harbour and Woodland SAC 000090;' no. November 1-24.

1432 Citing NPWS. 2015. "Glengarriff Harbour and Woodland SAC (Site Code: 90) Conservation Objectives Supporting Document – Marine Species," no.

April: 1-9. 1433 Which Mr Coram sets out. 1434 Which he identifies as a small sample.

1435 Citing Northridge, Simon, Jonathan Gordon, Cormac Booth, Susannah Calderan, Alexander Cargill, Alexander Coram, Douglas Gillespie, Mike Lonergan,

and A Webb. 2010. "Assessment of the Impacts and Utility of Acoustic Deterrent Devices. Scottish Aquaculture Research Forum SARF044."

1436 Citing Northridge, Simon, Alex Coram, and Jonathan Gordon. 2012. "Recent investigations on Seal Depredation at Scottish Fish Farms. Report to

Marine Scotland.” 1437 Citing Northridge et al., 2010; Northridge, Coram, and Gordon, 2012.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Coram on ADDs

950. Mr Coram states that harbour seals communicate by underwater sound but habituate to noise: for example, the Orthans Island haul-out is immediately adjacent a sailing and powerboating centre. He cites papers, including one of which he was lead author and specifically related to ADDs,1438 to the effect that:

“The cumulative impacts of increased ambient noise, e.g. from vessels or from acoustic deterrent devices, has been theoretically linked to hearing damage. The lack of empirical data on seals'

behavioural response to elevated noise levels makes it impossible to quantify the risk of hearing damage, but stable or increasing populations of harbour seal do exist within areas of relatively high marine traffic.”

951. Mr Coram states that,

• the use of seal scarers may constitute introduction of man-made energy at significant levels.1439

• that temporary hearing damage may be caused especially by ADDs “cannot be dismissed, but it is noted that such devices are in widespread usage.”

• the possibility of seal scarers causing hearing damage to individual seals from the SAC “cannot be excluded”.

952. Mr Coram’s conclusion is that “On the basis of scientific evidence it is concluded that the operation of a fish farm at Shot Head is unlikely to negatively impact the conservation status of the population of harbour seals within the Glengarriff Harbour and Woodland SAC.”

Coram – Comment & the Standard applicable in AA Screening

953. I confess to having struggled with the legal significance of the Coram report. This is not a case of evidential burdens on the applicants or requiring them to show a basis on which scientific doubt arises. They simply say of the Coram report that it will not do. On the one hand, I am conscious that it is by an expert conscious of the task set him by the Chairperson’s recommendation, the context of the Habitats Directive,

1438 Coram, A, J Gordon, D Thompson, and S Northridge. 2014. "Evaluating and Assessing the Relative Effectiveness of Acoustic Deterrent Devices and

Other Non-Lethal Measures on Marine Mammals." Scottish Government, 1-145;

Gotz, Thomas, and Vincent M. Janik. 2013. "Acoustic Deterrent Devices to Prevent Pinniped Depredation: Efficiency, Conservation Concerns and Possible

Solutions." Marine Ecology Progress Series 492 (October): 285-302. doi:10.3354/meps10482;

Jones, Esther L., Gordon D. Hastie, Sophie Smout, Joseph Onoufriou, Nathan D. Merchant, Kate L. Brookes, and David Thompson. 2017. "Seals and

Shipping: Quantifying Population Risk and Individual Exposure to Vessel Noise." Journal of Applied Ecology, no. April. doi:10.1111/1365-2664.12911;

Grigg, Emma K., Sara G. Allen, Deborah E. Craven-Green, A. Peter Klimley, Hal Markowitz, and Deborah L. Elliott-Fisk. 2012. "Foraging Distribution of

Pacific Harbor Seals (Photo vitulina Richardii) in a Highly Impacted Estuary." Journal of Mammalogy 93 (1): 282-93. d o i:10.1644/ 11-M AM M-A-128.1.

1439 Mr Coram adds “but these levels would be well outside the boundary of the SAC”. This is pretty irrelevant as ex situ effects on a species for which a

site is designated may, in principle, adversely affect the integrity of the site having regard to its conservation objectives. E.g. Highlands Residents

Association v An Bord Pleanála [2020] IEHC 622. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

reliance on scientific evidence and the criterion of impact on the conservation status of the SAC. Yet, on the other, he does not even mention the concepts of AA and AA Screening – concepts basic to an AA Screening Report. It was only in retrospect – indeed only in its final determination – that ALAB deemed it a “Seal Screening Report”.1440 These issues might be overlooked as merely formal if the substantive content of the report was more reassuring. Mr Coram concludes that the Salmon Farm “is unlikely to negatively impact the conservation status” of the seals of the SAC. But he precedes that conclusion, in the context of a conservation objective contraindicating noise, “at levels that could result in a significant negative impact on individual” and an earlier observation that “Temporary hearing damage caused by … more especially acoustic deterrent devices cannot be dismissed”, with findings that

• Seal scarers may introduce noise at significant levels.

• “The possibility of acoustic deterrents causing hearing damage to individuals from the … SAC cannot … be excluded.”

954. Mr Coram gives no advice that AA is not required – as I said, he does not even mention AA. That is the obvious thing to say if screening out AA or opining that it should be screened out. The sentence last-cited above is not the formula of conclusion one expects to find in an AA screening report. Nominally, the test in AA screening as suggested in the 10th recital to, and Article 6(3) of, the Habitats Directive is indeed whether

the project is “likely to have a significant effect on the conservation objectives” of a designated site. But any

expert in AA knows, or should, that the criterion is not in reality whether effect is likely or “unlikely” in the

ordinary sense in which those words are used in the English language and that use of these phrases as expressing the test is apt to mislead.

955. Barniville J in Eoin Kelly1441 traced the caselaw as far as the test posed by Sharpston AG in Sweetman1442 in 2012 and adopted by Finlay-Geoghegan J in Ted Kelly1443 – which adoption had been noted in Harten.1444 That was “whether the plan or project concerned is capable of having an effect” and her view

was that “It follows that the possibility of there being a significant effect on the site will generate the need for

an appropriate assessment ..”. She said that AA Screening operates “merely as a trigger” and the threshold in AA Screening for requiring AA is “a very low one” -as the question to be answered is “should we bother to

check?”. She described the requirement of “significant” effect in AA screening as a “de minimis threshold”

designed to prevent the “legislative overkill” of requiring AA of projects with “no appreciable effect on the site”. Thus, it seems to me, Sharpston AG had in mind that, in AA Screening, the possibility of an effect combined with the possibility that it would be significant, sufficed to trigger AA.

1440 ALAB Determination 29/6/21 §2.16.

1441 Eoin Kelly v ABP & Aldi [2019] IEHC 84 §38 et seq. citing Ted Kelly v An Bord Pleanála [2014] IEHC 400; Connelly v An Bord Pleanála [2018] IESC 31;

[2018] ILRM 453, Case C- 127/02 Waddenzee 200 ECR I-07405, Case C-258/11 Sweetman & Others v An Bord Pleanála ECLI:EU:C:2012:743; Case C-

323/17 People Over Wind and Sweetman v Coillte Teoranta ECLI:EU:C:2018:244. 1442 Case C-258/11 Sweetman & Others v An Bord Pleanála, Judgment of 11 April 2013, ECLI:EU:C:2012:743. 1443 Ted Kelly v An Bord Pleanála [2014] IEHC 400. 1444 Harten v An Bord Pleanála [2018] IEHC 40 (McDermott J, 29 January 2018).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

956. However, Barniville J cites and agrees with Haughton J in his view, expressed in Alen-Buckley,1445 that, properly understood, Sharpston AG did not opine that, and Haughton J rejected an argument that, a mere possibility of a significant effect on a European Site would trigger a need for AA. Rather, they

considered that her view was that the “the “likely significant effect” did not have to be proved, it just had to

be a possibility.”1446 Simons J In Heather Hill1447 noted Haughton J’s rejection of what he called a “different

test” to that suggested by Sharpston AG – one “with a lower threshold, based on the mere possibility of significant effect and without reference to any likelihood.”

957. I confess to respectfully wondering if a threshold set at the possibility of a likelihood, or for that matter, at the likelihood of a possibility, will be of practical assistance to those attempting AA Screening. I am not clear where one ends up with a test of “possibility” if one must gloss it in some way by reference to a “likelihood”. The two words seem to me to denote, at least ordinarily, different concepts on different sides

of a watershed. But in Eoin Kelly the likelihood in question is described to the effect that “the word “likely”

should be read as being less than a balance of probabilities” – this seems close to pitching the test as the possibility of a possibility. I am fearful lest I have misunderstood and diffident in my view accordingly. But if indeed I have got “greim daingean” on the wrong end of the stick, perhaps I am not alone and that may speak to the practicality of such a test. For my part, I confess to the view that,

• Sharpston AG’s emphases that the threshold was “de minimis” and “very low” and operated “merely as a

trigger” to see “should we bother to check?”,

• her invocation of the precautionary principle and

• her reference to the other language versions of the Directive as “suggesting that the test is set at a lower level and that the question is simply whether the plan or project concerned is capable of having an effect”, combine strongly to suggest that she did indeed intend to opine that the possibility of significant effect suffices to trigger the need for AA.

958. For my part – and perhaps it is the lack of linguistic imagination of an anglophone common lawyer cleaving to the concept of the balance of probabilities – I think Waddenzee1448 started the confusion by

positing that the “mere probability or a risk that a plan or project might have a significant effect is sufficient

to make an “appropriate assessment” mandatory”. That permeated later cases. After the trial in the present case, the CJEU in Eco Advocacy may have tried to diminish the confusion by substituting the phrase

“likelihood or a risk that the plan or project will have a significant effect on the site concerned.”1449 Linguistic

meanings in this context are not set in stone, nor used uniformly. And, as Sharpston AG points out in Sweetman,1450 in the case of EU legislation one may need to consider authoritative versions in many

languages. That said, ordinarily in English that something is “probable” or “likely” means it is more likely than

not. “Probable” is contrasted with “possible”. A “probability” is contrasted with a “possibility”. The latter

1445 Alen-Buckley v An Bord Pleanála [2017] IEHC 541. 1446 Alen-Buckley §81. 1447 Heather Hill Management Company clg v An Bord Pleanála [2019] IEHC 450 (Simons J, 21 June 2019), §127. 1448 Case C-127/02. 1449 Case C‑721/21, Eco Advocacy CLG v An Bord Pleanála, Judgment of 15 June 2023, §37. 1450 Case C-258/11 Sweetman & Others v An Bord Pleanála, Judgment of 11 April 2013, ECLI:EU:C:2012:743.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

more closely approximates, it seems to me, with the word “risk” as used in Waddenzee. However, and confusingly, in ordinary English while a probability or a likelihood of something means it is more likely than not, the probability or the likelihood of something may be less likely than not – may amount to merely a possibility.

959. However, as I say, “probable” usually means more likely than not. Yet clearly, no-one believes that Waddenzee means that in AA screening AA is required only if significant effect is more likely to occur than not. Waddenzee, it seems to me, is more helpful in its stricture that AA is required, if “it cannot be excluded, on the basis of objective information and, in particular, in the light of the characteristics and environmental conditions of that site, that it1451 will have a significant effect on that site.” The CJEU has repeated that formula in many cases since.1452 Article 42(6) of the Habitats Regulations 2011 is in similar, though not identical, terms – AA is required “if it cannot be excluded, on the basis of objective scientific information following screening under this Regulation, that the plan or project, individually or in combination with other plans or projects, will have a significant effect on a European site.”

960. It seems to me that that which is possible “cannot be excluded”.

961. It is in that sense of possibility simpliciter that I read the reference to the opinion of Sharpston AG in

Sweetman1453 by McDonald J in Dublin Cycling.1454 He said, “It is sufficient that there is the possibility of there

being a significant effect.” In similar vein in Protect East Meath,1455 McDonald J said that “in carrying out a screening test, the precautionary principle applies such that, in any case of doubt as to the absence of significant effects, a full appropriate assessment must be carried out. This is so notwithstanding that the

words “likely to have a significant effect” are found in article 6(3) of the Habitats Directive.”1456 It is in this

sense also that I read Allen J in Kemper1457 when he, typically pithily, says of the low threshold that “… it can

come as no surprise that a site screened in may later be shown not to be affected. The sites are screened in because it is not clear that there is no possibility of adverse effects and later ruled out if the outcome of the appropriate assessment is that there is no such possibility.” In Carroll1458 in 2016 Fullam J cited Sharpston AG

to the effect that if, in AA Screening, “ … there is a possibility that there would be a significant effect, there

will then be a need for an appropriate assessment ..” – AA is required if AA Screening concludes that there

“may be” a significant effect. I read Pearse1459 similarly. To the extent of any difference between Kelly, Alen-

Buckley and Heather Hill on the one hand and, on the other, Carroll, Dublin Cycling, Protect East Meath and

1451 i.e. the project.

1452 Case C-6/04 Commission v United Kingdom [2005] ECR I-9017, §54; Case C-418/04 Commission v Ireland [2007] ECR I-10947, §226; Case C-538/09,

Commission v Belgium §39; Case C-323/17, People Over Wind, v Coillte Teoranta §34. 1453 Case C-258/11 Sweetman & Others v An Bord Pleanála, Judgment of 11 April 2013, ECLI:EU:C:2012:743. 1454 Dublin Cycling Campaign CLG v. An Bord Pleanála [2020] IEHC 587, §103. 1455 Protect East Meath v An Bord Pleanála, & Ravala [2020] IEHC 294 [2021] 2 IR 796. 1456 Emphases added. 1457 Joyce Kemper v An Bord Pleanála [2020] IEHC 601, §310. 1458 Carroll v An Bord Pleanála [2016] IEHC 90, §§24 & 25. 1459 Pearse v An Bord Pleanála [2019] IEHC 865 §42.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Kemper (and perhaps I am seeing shadows as to any such difference), I respectfully follow the latter as, in my view, better reflecting the views of Sharpston AG and the principles, concerns and precautionary approach which underlay them and as providing a test which is more easily understood and applicable. Though I do not rely on it given the view I have just expressed, I note that this view seems to me also to conform to the view of AA Screening taken, since the trial of this case, by the CJEU in Eco Advocacy1460 that, to decide against doing AA, what was required is “certainty … that there was no reasonable scientific doubt as to the possibility that that project would significantly affect that site”.

962. I am heartened as to both my confusion and my conclusion by a footnote to the opinion of Sharpston AG in Sweetman. She says of the Habitats Directive:

“An example of the type of confusion that this poorly-drafted piece of legislation can give rise to can, I suggest, be seen in the judgment in [Waddenzee], the Court talks of an appropriate assessment being required if there is a ‘mere probability’ that there may be significant effects. In paragraph 43, it refers to there being a ‘probability or a risk’ of such effects. In paragraph 44, it uses the term ‘in case of doubt’. It is the last of these that seems to me best to express the position.”

Remembering that an NIS is usually prompted by, and its scope is usually determined by, the output of an AA Screening, I note also that by Art 2(1) of the Habitats Regulations 2011, an NIS is required to “to identify and

characterise any possible implications of the …project …. in view of the conservation objectives of the site …”.

963. Of course, though a low threshold in AA Screening, that of possibility of significant effect is nonetheless a threshold. First, the effect of which there is a possibility must be a significant effect. Second,

the only gloss I would put on the concept of “possibility” is that it must be real not fanciful – while cautioning

that even that gloss must not exaggerate the height of the threshold. Perhaps that the possibility must not be fanciful expresses it best.

964. I am conscious that it is unusual to criticise a report for adopting the explicit statutory wording of the

test – whether the project is “likely to have a significant effect”. But it is surely to be expected of experts in

AA that they will be aware of the gloss placed by caselaw on that test. The views of Sharpston AG in Sweetman go back to 2012. I am conscious that an AA Screening opinion passed muster in Alen-Buckley which expressed its conclusion in terms that the proposed development was “not likely” to have significant effects on any European sites. So too with the Coram report. But, as Barniville J said in Eoin Kelly,1461 the issue is one of substance not of form or pedantry. That observation tends to cut in favour of the validity of a

1460 Case C‑721/21, Eco Advocacy CLG v An Bord Pleanála, Judgment of 15 June 2023, §42. 1461 §98 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

screening decision – Barniville J cited Ó Gríanna1462 as to pedantry. But it can cut either way. A report in form

inadequate can be in substance adequate – and vice versa.

Coram as an AA Screening - Conclusion

965. I must read the Coram Seal report as a whole. As I say, I have struggled with it. On the one hand it (properly) cites the Habitats Directive, invokes scientific evidence, sets out to assess threat to conservation status and concludes, echoing the Directive, that significant effect is “unlikely”. On the other hand, echoing Waddenzee and the cases following it, it says that the possibility of acoustic damage to seals by seal scarers cannot be “excluded”. I tend to think that the use of the word “excluded”, echoing Waddenzee, is not a coincidence given the salience of that case to anyone involved in AA. That hearing damage may be “temporary” does not seem to me of itself to compel an inference that it is not significant (though temporary effects may generally be less likely to be significant). That is perhaps particularly so of animals which “use underwater sound for communication” and as to which the conservation objectives impugn negative effect by noise. And Mr Coram says the risk of such damage arises “more especially” from seal scarers. Finally, the observation that “such devices are in widespread usage”, while perhaps inferentially or anecdotally reassuring, does not seem to me, at least without more, to comprise “objective information” as contemplated by Waddenzee.

966. Not without hesitation, I have concluded that the Coram report does not provide a legally adequate basis for screening out AA as to the effect of seal scarers on seals. But it may be that that matters only if there is in reality a prospect of the use of seal scarers such that they have to be considered in the first place.

After the Coram Seal Report, ALAB’s AA Screening out of Risk to Seals, February 2018, ALAB Decision to do AA April/June 2019, MOWI’s NIS July 2020 & Consideration of Seals thereafter.

967. The Coram Seal report was circulated for comment. NPWS had no comment.1463 DAFM replied1464 that “Given that this man-made noise would not extend to the SAC, we concur with the report's conclusion that the Shot Head farm and its operations would not negatively impact the conservation status of harbour seals at the Glengarriff and Woodland SAC.” DAFM’s premise is misconceived, as effects ex situ on fauna for which a European Site has been designated are assessable in AA if those effects are liable to affect the

1462 Ó Gríanna v An Bord Pleanála (No. 2) [2017] IEHC 7 (“Ó Gríanna (No. 2)”). “The principle of effectiveness is not a mandate for construing the Directive

in the most onerous manner possible. This involves the courts being astute to ensure the objectives of the Directive are met but not in an overly pedantic

way.” (per McGovern J. at §38, p16). 1463 NPWS to ALAB 16 November 2018. 1464 DAFM to ALAB 27 April 2018.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

conservation objectives of the site – Holohan.1465 For example, in Holohan, Kokott AG cites Moorburg1466 in which a power plant cooling system drew large amounts of water from the Elbe, thereby killing fish traveling to European Sites, a considerable distance upstream, designated for the conservation of those fish. By creating the risk that fewer fish would reach those sites, the power plant adversely affected their integrity. That effects are ex situ may in particular circumstances contribute to screening out AA but not on an ipso facto basis.

968. ALAB considered the Coram report at its meeting of 19 February 2018. Dr Saunders was in attendance as “Technical Advisor”. The minutes summarise the content of the Coram report but do not

analyse it. They merely record ALAB’s acceptance of the “conclusion” of the “Technical Advisor” “that on the

basis of scientific evidence the operation of a fish farm at Shot Head is unlikely to negatively impact the conservation status of the population of harbour seals within the Glengarriff Harbour and Woodland SAC”. That is a verbatim repetition of the conclusion of the Coram report and no other opinion or contribution of the Technical Advisor is recorded.

969. In particular, ALAB record but do not analyse the Coram finding that “… the possibility of acoustic deterrents causing hearing damage to individuals from the Glengarriff Harbour and Woodland SAC cannot be excluded.” Neither does the minute record a view that, as is now argued before me, seal scarers are not proposed, so their use need not have been screened for AA. It is not apparent to me that this minute advances beyond the content of the Coram Report the matter of AA Screening of the risk of acoustic damage by seal scarers to seals. I have already concluded above that the Coram report does not provide a legally adequate basis for screening out AA as to the effect of seal scarers on seals.

970. Clearly, any AA Screening decision screening out particular posited risks and screening in others for AA must precede the performance of the AA. Not least, that is because the AA Screening decision determines the scope of the ensuing AA. ALAB’s meeting of 19 February 2018 was the last occasion on

which, prior to its starting AA, it considered the issue of seals. So the only “candidate” for identification as its

AA screening decision screening out risk to seals is its decision at that meeting.

971. At this point it is useful to recall the description by Kokott AG in Eco Advocacy, of the role of AA Screening:

“Screening serves to identify plans and projects which do not require a full assessment because it is clear even without a full assessment that they may qualify for approval.1467 Screening is not intended,

1465 Case C-461/17 Holohan v An Bord Pleanála, Opinion of Kokott AG of 19 January 2023 & Judgment of 15 June 2023, [2019] PTSR 1054.

1466 Case C-142/16 Commission v Germany (Moorburg) (, EU:C:2017:301, §§29-31).

1467 Citing her Opinion in Case C-127/02 Waddenzee (EU:C:2004:60), §72 – which, she says that the Board (i.e., An Bord Pleanála), probably on account

once again of the English translation, misunderstands. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

however, as a mechanism for circumventing a full assessment,1468 let alone for implementing plans or projects which would not qualify for approval if fully assessed [by way of AA]. For that reason, the criteria governing the screening must be just as strict as the criteria for the appropriate assessment itself.”

972. The EIA Directive specifically provides for reasons for screening decisions, whereas the Habitats Directive does not. But the CJEU in Eco Advocacy implied the same principles by analogy to AA as apply expressly to EIA.1469 Kokott AG noted1470 that a screening decision screening out EIA must contain a statement of reasons. They must form part of the decision. It does not suffice that they be given later. Given the CJEU’s reasoning by analogy, it does not take a great leap to conclude that the same rules apply to an AA screening – such that the reasons for it must be given in the decision screening out AA and not later. However, even such a leap is unnecessary as Art 42(18(a) of the Habitats Regulations 2011 provides that “any determination” (which includes an AA screening determination), shall be made available for inspection at its office and on its website “as soon as may be after the making of the determination”.

973. Though in Eco Advocacy the CJEU ruled out as inadmissible the 6th question referred to it by way of preliminary reference,1471 it is difficult to doubt but that Kokott AG, in her opinion on that question, was correct in advising the CJEU as to AA Screening that the competent authority must ensure that its statement of reasons for a decision not to carry out an AA “are recognisable as such and comprehensible”. That is consistent with the view taken in Christian,1472 EMI Records1473 and Connelly1474 that “Legal certainty requires … that it must be possible to accurately determine what the reasons were. There should not be doubt as to where the reasons can be found.”

974. The advice of Kokott AG in Eco Advocacy was that a competent authority in screening out AA “must at least provide an express and detailed statement of reasons which is capable of removing all reasonable scientific doubt concerning the harmful effects of the works envisaged on the integrity of the protected site concerned.” The CJEU essentially accepted that advice: it held that EU law does not require, in the statement of reasons for a decision screening out AA, a response “one by one, to all the points of law and of fact raised by the interested parties during the administrative procedure”. But it must “state to the requisite standard

1468 Citing Case C-323/17 People Over Wind and Sweetman, Judgment of 12 April 2018 (EU:C:2018:244) §37.

1469 Judgment of the CJEU §§32 & 33: Neither Article 6(3) of Directive 92/43 nor any other provision of that directive lays down requirements as to the

statement of reasons for decisions taken pursuant to Article 6(3). That said, it should be recalled, in the first place, that the right to good administration,

in so far as it reflects a general principle of EU law, has requirements that must be met by the Member States when they implement EU law. Among those

requirements, the obligation to state reasons for decisions adopted by the national authorities is particularly important, since it puts their addressees in

a position to defend their rights under the best possible conditions and decide in full knowledge of the circumstances whether it is worthwhile to bring

an action against those decisions. It is also necessary in order to enable the courts to review the legality of those decisions (Case C-46/16 LS Customs

Services, Judgment of 9 November 2017, EU:C:2017:839, §§39 & 40 and the case-law cited). 1470 Case C-721/21 Eco Advocacy CLG v An Bord Pleanála, Opinion of Kokott AG of 19 January 2023, §70.

1471 Whether where AA is screened out “there should be an express, discrete and/or specific statement as to what documents exactly set out the reason

of the competent authority.” 1472 Christian v Dublin City Council [2012] IEHC 163, [2012] 2 IR 506. 1473 EMI Records v Data Protection Comm [2013] IESC 34, [2014] 1 ILRM 225 §250. 1474 Connelly v An Bord Pleanála [2018] IESC 31; [2021] 2 IR 752.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

the reasons why it was able .. to achieve certainty, notwithstanding any opinions to the contrary and any reasonable doubts expressed therein, that there was no reasonable scientific doubt as to the possibility that that project would significantly affect that site.” This requirement of reasons must in logic equally apply to a Screening Decision ruling out AA of some risks where requiring AA of others.

975. As to screening out from AA any risk of seal scarers to seals, ALAB’s minute of 19 February 2018 does not invoke the concept of AA or AA screening or the standards applicable in AA Screening nor does it purport in terms to be an AA Screening decision. Nor does it in terms purport to be or in substance consist of an AA screening decision setting out the reasons considered by Kokott AG in Eco Advocacy1475 to be required. Nor, which amounts to the same thing, does it state, as required by Art 42(7) read with Art 42(18)(a) of the Habitats Regulations 2011 how, as to any risk of seal scarers to seals, it “excluded, on the basis of objective

scientific information following screening … that the … project, … will have a significant effect on a European

site.”

976. However, for all that it may not have recorded it properly, it is clear that screening out risks to seals for AA purposes is what ALAB in fact did on 19 February 2018. Indeed that is the clear gravamen of ALAB’s

written submission that risk to seals did not “did not need to be expressly identified in the NIS”, as, in its view

(with which I respectfully disagree) seals were ‘screened out’ in Dr Coram’s report. The only possible basis of

such a submission is that ALAB’s AA screening decision (not Dr Coram’s – the decision was not his to make and in my view his report does not support such a decision) preceded ALAB’s notice of determination dated

20 June 2019 that AA was required of listed matters which list did not include risks to seals. MOWI’s written submissions are to the same effect: “The NIS is prepared after the proposed development has been subject to screening for Appropriate Assessment which will have screened in / screened out SACs / SPA’s as well as SCIs. The Appropriate Assessment is only concerned with those SACs / SPAs and the SCI’s that have been screened in. ” and “The NIS did not address … the common seal as these qualifying species had been screened out in the reports of …, Dr. Coram …”.

977. I note that the first version of the Marine Institute “AA Screening Matrix for Aquaculture in Outer Bantry Bay” had not yet issued when ALAB made its screening decision of February 2018 to exclude risk to seals from AA.

978. At its meetings on 30 April and 15 May 2019 ALAB accepted the recommendation of the supplemental AA screening report for birds1476 that AA be done and on 20 June 2019 ALAB issued to MOWI a “Notice Of Determination That Appropriate Assessment is Required”.1477 That notice, as it necessarily did,

1475 Case C-721/21 Eco Advocacy CLG v An Bord Pleanála, Opinion of Kokott AG of 19 January 2023, §90. The Court has since agreed by judgment dated

15 June 2023, §31 et seq. 1476 Crowe, April 2019. 1477 Citing Art. 42(8) of the European Communities (Birds and Natural Habitats) Regulations 2011.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

identified the matters of which AA was required.1478 Those matters did not include any risk to seals. In consequence, MOWI’s NIS did not address any risk to seals.

979. ALAB does not seem to have considered seals again until its meeting of 20 March 2020, when it asked Dr Saunders to confirm, with reference to the issue of seal scarers as quoted in the Seal report, that the SAC distance from the site is 4.5km. That was confirmed to ALAB on 15 May 2020. While this was a relevant inquiry, as observed above, the answer was not necessarily dispositive of the issue of acoustic damage to seals give they would be attracted to the salmon farm – Holohan and Moorburg. Seals are next mentioned in the ALAB minutes of 30 November 2020 when it asked its Technical Advisor clarify whether there are NPWS protocols on marine noise.

Position in February 2018 as to AA Screening of Risk to Seals – Decision

980. From the foregoing I find that:

• ALAB made an AA Screening decision on 19 February 2018 excluding risk to seals from AA. This was reflected, if only negatively by omission,1479 in the scope of its notice of determination dated 20 June 2019 that AA was required. That notice listed the risks requiring AA and it did not include risk to seals.

• Despite his general conclusion as to absence of risk to seals, and for reasons set out more fully above, the Coram Report did not suffice as basis for screening out AA of the acoustic risk to seals posed by seal scarers. Not least, it recorded that such a risk “cannot be excluded”. Such exclusion is a condition of screening out a risk from AA.

• ALAB was required to publish its AA Screening decision, including its reasons for concluding that it could be “excluded, on the basis of objective scientific information following screening ..that the project, will

have a significant effect on a European site.” as to risk to seals. It was required to publish that decision “as

soon as may be” after it was made.

• ALAB’s minute of 19 February 2018 of its AA Screening decision is the only candidate document for publication “as soon as may be” after it was made and, leaving aside whether and when it was published (I am unsure of the position in that regard) it is deficient as to reasons.

1478

Saying, “… it is not possible to rule out potential significant adverse impacts resulting from the Installation of the proposed salmon farm in respect of

selected nearby SPAs and their associated bird species of conservation interest (SCI). The specific SCIs and Natura sites of concern are Fulmar (Beara

Peninsula SPA, lveragh Peninsula SPA, Deenish Island and Scariff Island SPA), Gannet (The Bull and The Cow Rocks SPA and Skelligs SPA) and Guillemot

(lveragh Peninsula SPA). The Appropriate Assessment screening report determined that: 1.The proposed development of aquaculture sites within Bantry

Bay will result in the loss of 42.5 ha of inshore habitat that could potentially be used by the abovementioned SCIs for feeding, and 2.cumulative impacts

have been identified that may, in combination with the above development, exacerbate further the Impacts on the SCls. They include additional fish

farms in the area (elsewhere in Bantry Bay) and associated levels of disturbance.” 1479 I do not suggest it was an erroneous omission given its screening decision to exclude risk to seals from AA.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• No AA was done as to risk to seals.

981. As to my last conclusion, the position became fogged by later documents reflecting continued consideration of risk to seals. However, such documents did not include the formal documents generated in the AA process save as indicated below. Not least, no public participation in AA as to risk to seals ensued.

Seals – Dr Saunders’ Final Report December 2020 & ALAB’s Reaction

982. Dr Saunders’ Final Report states as to “Threat of escape resulting from seal predation” that “Frequent visits from seals may be expected to occur at the Shot Head site.”1480 While the papers suggest that salmon are not a major element of the diet of seals, nonetheless, Dr Saunders states that “From the EIS it is difficult to determine the anticipated or predicted risk of nuisance from seals”. But he cites academic authority to the effect that “Seals pose a particular risk with regard to escapes, as predator interference and associated net biting, resulting in a net breach, make up 47% of reported escapes.”1481 And he later refers, in the context of seals, to the “the ever-present risk of fish escapes”. Given the uncontradicted concerns as to the environmental consequences if large scale farmed salmon escapes (though I do not rule out that the degree of those consequences could be disputed) it seems difficult to avoid the view for EIA purposes that the risk of farmed salmon escapes due to seal predation is sufficiently significant as to require assessment in EIA.

983. Dr Saunders’ Final Report also states that:

“In the EIS1482 it is stated that seal nets or seal scarers may be employed if required. From the EIS1483 it is difficult to determine … the effectiveness and impact of seal control by anti-predator nets, seal scarers, or licensed shooting of seals. Antipredator nets are not commonly used in European farms because of several recognised disadvantages, including reduced water flow and impacts on marine life, particularly bird entanglement.”1484

984. As to both these concerns – the risk of escapes due to seal predation and the efficacy of antipredation measures – it seems probable that MOWI had useful information which it had not provided to ALAB. Dr Saunders states:

1480 Technical Advisor Saunders Final Report December 2020 §9.2 p71. Emphasis added. 1481 Emphasis added. 1482 The text says “EIA” but that is clearly a typo. The text cites Vol. 1, Section 3.3.2, page 152. 1483 The text says “EIA” but that is clearly a typo. The text cites Vol. 1, Section 3.3.2, page 152. 1484 Citing Northridge et al 2013. Layout changed for exposition.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“…… it would be advantageous to establish whether seal damage has been an issue with the farms already operating in Bantry Bay, as this may provide some insight in any possible future requirement for deterrent devices.”1485

“Due to the ever-present risk of fish escapes, it would be prudent to establish whether this is an issue of concern facing existing operations in the area and, if so, whether current measures are considered effective. Accordingly a statement on the current impact of seals at the Roncarraig1486 Site and effectiveness of control measures would enable a more robust assessment of the risk of seal damage to nets.” 1487

985. One may add, though I am sure Dr Saunders intended it by implication, that such a statement would not merely describe the effectiveness of current measures in preventing escapes by seal predation – it would inevitably, in doing so, identify those measures. If seal scarers have never been used at Roancarrig that would be one thing. If they are already in use there or have been in use there, that might be quite another – though that would be for ALAB to consider, not me.

986. Dr Saunders’ phrase “more robust” is puzzling in the absence of any assessment for EIA purposes, by Dr Saunders of this risk of seal damage to nets. He notes that, generally, escapes due to seals “make up 47% of reported escapes” but there is certainly no site-specific assessment. Dr Coram had not assessed this risk. At least in part, any such assessment would seem to turn on a risk analysis as to the

• likelihood of occurrence of a net tearing incident due to seal predation on farmed salmon.

• likelihood of large scale escape by reason of any such incident.

• the quantum of environmental risk resulting from any such large scale escape.

What can be said is that Dr Saunders’ report provides no basis for a conclusion in EIA that, absent mitigation, the risk of farmed salmon escape due to seal predation is insignificant. In truth, Dr Saunders’ report tends appreciably to the contrary.

987. Dr Saunders’ final evaluation of Noise Impacts notes:1488

• that Acoustic Deterrent Devices “may have to be considered in the future”.

• however that, as stated in the Minister’s 2015 EIA1489, if they are required they would be require separate statutory consent, at which time a “full evaluation” of their effects would be undertaken.

1485 Technical Advisor Saunders Final Report 8 December 2020 §9.8.1 p85. 1486 Sic. Roancarraig is a MOWI site. 1487 Technical Advisor Saunders Final Report December 2020 §9.2 p71 Emphasis added. 1488 Technical Advisor Saunders Final Report December 2020 §9.13. 1489 EIA, §8 Principal Issues of Environmental Relevance, p14 §2.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Dr Saunders concludes “We have therefore not considered the use of ADDs in this report, beyond acknowledging that they may be necessary in the future.”

Notably here, Dr Saunders acknowledges that ADDs may be requited and his precise rationale for his not considering them further is the prospect of their full evaluation in a separate statutory consent process.

988. Where:

• a decision-maker’s Technical Advisor has refrained from assessment of a significant risk for want of information and has advised that information be sought,

• the decision-maker decides not to seek that information, one would expect the decision-maker to articulate and give reasons for its rejection of that advice and its assessment of the risk on the information available. I have not found either a request by ALAB for that information from MOWI as to its experience with seals at the Roancarrig site or an articulation of the reasons why it was not requested. It is difficult to understand why this inquiry was not made. Indeed, I have not found minutes of any consideration by ALAB thereafter of the EIA issue of the risk of seal predation causing salmon escapes – as opposed to the AA issue of risk to seals due to seal scarers.

Seals – MERC AA Report September 2020, Marine Institute Screening Matrix, MOWI’s s.47 Reply March 2021 & MERC’s briefing note May 2021.

989. MERC’s AA report dated 11 September 2020, being concerned with the risks to birds identified as requiring AA, does not in terms address risk to seals or, more specifically, risk to seals by ADDs. However it repeatedly, if incidentally, mentions seals. Those references include the following from Scottish sources:

• A 2001 review identified predators on fish farms – of which seals, were “more regularly recorded” at more sites than other predators.

• A 2004 survey of anti-predator controls at marine salmon farms stated that “seals were the most common predators, being reported at 81% of sites ..”

• A 2018 review stated that, o “… mammals, especially seals, may take, injure or frighten farmed fish, or damage nets leading to escapes. ..”

o “significant numbers of wild fish … can be attracted to the vicinity of fish.” “Although quantitative information is lacking, it seems likely that … seals, ….. could also be attracted to fish farms by increased densities of, and enhanced opportunities for feeding on, wild prey. This … may increase risks of depredation, exposure to noise pollution, entanglement or other interactions with fish farms … but further information is needed”.

So MERC records clear evidence of the interaction of seals and aquaculture sites. Indeed, if there weren’t such interaction, the issues of shooting seals and the use of anti-predator netting to deter them would not

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

have arisen. Nor would it have been the case that, as Mr Coram put it, seal scarers “are in widespread usage”.

990. MERC also cited the Marine Institute Screening Matrix as stating, that “there is no evidence in the scientific literature to suggest that aquaculture activities impact on seal species …”. As I have already observed, the legal status and the documentary basis of this matrix are unclear. Also, while such an observation may in many contexts be weighty, in the particular context of

• the low threshold and obligation of certainty in AA Screening,

• Mr Coram’s opinion that “Temporary hearing damage caused by … more especially acoustic deterrent devices cannot be dismissed” and that “The possibility of acoustic deterrents causing hearing damage to individuals from the Glengarriff Harbour and Woodland SAC cannot … be excluded”,

• the 2018 Scottish review, as to, inter alia, “exposure to noise pollution”, that “further information is needed”, one is reminded of the aphorism that absence of evidence is not evidence of absence. Though I am not to be taken as elevating that aphorism into a legal principle, it would be for ALAB to consider the content of the matrix on a full understanding of its legal and substantive significance, weight and purpose and that in the context of the other materials before it.

991. ALAB on 15 February 2021 issued a s.47 Notice to MOWI, seeking information on effects on birds. Unsurprisingly, given the AA Screening decision of February 2018, it did not address the issue of seals. MOWI replied on 2 March 2021 in terms which incidentally included a fact sheet briefly and generally describing the Harbour Seal but not describing its predation, either generally or on farmed salmon. The fact sheet was not specific to Bantry Bay, the SAC, or MOWI’s experience of “the current impact of seals at the Roncarraig Site”.1490 That is not to criticise MOWI given the terms of the s.47 request. It is merely to make clear that the information provided was not that as to MOWI’s experience of seals at Roancarrig contemplated by Dr Saunders as required.

992. MERC’s briefing note of 19 May 2021, in responding to An Taisce concerns as to Gannets, incidentally, observed that its literature review identified “those species which interacted frequently with cage farms in the north east Atlantic were seals and cormorants”.

The Plea and Evidence as to Aquaculture Stewardship Council Certification

993. As stated, MOWI pleaded that “at this point in time” (i.e. July 2022), 6 of its fish farms, including 2 in Bantry Bay were “Aquaculture Stewardship Council certified fish farms and a condition of such certification is

1490 The weblink on the fact sheet suggests it is an extract from the “Encyclopaedia of Life” – eol.org.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

not to use acoustic devices given the threat they may pose to the hearing of mammals.” Precisely, this is correct. It also authoritatively articulates “the threat they may pose to the hearing of mammals” such that it is difficult to see that any such threat to seals of the SAC could be screened out for AA.

994. The 2018 sEIS said1491 that MOWI subscribes to the Aquaculture Stewardship Council (“ASC”) Audit process, for all its sites. MOWI’s Ms McManus1492 identified Roancarraig as one of the 6 and exhibits the

Council’s “ASC Salmon Standard” version 1.3 of July 2019.1493 §2.5.1 of that Standard records that “a large

variety of acoustic deterrent and harassment devices is used in salmon aquaculture. Based on available research, it appears that the effectiveness of these devices in reducing farmed salmon predation by marine mammals can vary widely, including by location, marine mammal species, period of use, etc.” It also suggests that initial effectiveness is lost over time. The Standard does say that from July 2022 certification will require that certified farms use no such devices. However, and notably given the phrase “at this point in time”, the ASC Standard provided that until July 2022 the standard “encouraged” that acoustic scarers not be used but did not prohibit their use. Rather it required that their use be limited to non-continuous use for less than 40% of the days in the production cycle. So, it is clear that up to and at the date of ALAB’s determination, ASC certification of Roancarraig did not prohibit the use there of seal scarers.

995. In any event, it is clear that at the date of Dr Saunders’ final report in December 2020 he had no knowledge whether seal scarers had been used in Roancarrig, as he advised that ALAB enquire in that regard. I have found no evidence that ALAB did so or whether seal scarers had been used in Roancarrig.

However, the NIS in July 2020,1494 though it did not address risk to seals, did in a different context – the issue

of aquaculture waste – commit to ASC certification of the Shot Head Farm if licensed. Given the ASC ban on the use of seal scarers from July 2022, I accept that this obliquely implies a statement of intention by MOWI to not use seal scarers. But it is not apparent that:

• the “ASC Salmon Standard” version 1.3 of July 2019 or the certification status of Roancarraig were before ALAB when it made its final determination in July 2021.

• an earlier version of the ASC Salmon Standard or the certification status of Roancarraig were before ALAB when it screened out seals from AA in February 2018.

• MOWI at any point had amended its statement in its 2011 EIS that assessment of the possible need for seal scarers “to protect the stock from seal attack” would be “necessary …… early in the development of the site”1495

• MOWI at any point had articulated to ALAB its intention to not use seal scarers.

1491 pp 8 & 99. 1492 Affidavit 5 July 2022 §4. 1493 Exhibit CMcM1 Tab 5. 1494 p31. 1495 Emphasis added.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Seals – ALAB’s AA Conclusion Statement May 2021 & ALAB’s Determination June 2021

EIA of risk of Farmed Salmon escape by Seal Predation

996. ALAB’s AA Conclusion Statement dated 28 May 2021 and ALAB’s Determination dated 29 June 2021 considered the issue of impact on seals for AA purposes but not the risk of impact by seals on the risk of fish escape for EIA purposes.

997. As to EIA, the point is not that ALAB was necessarily required to conduct the further inquiries, recommended by Dr Saunders, as to MOWI’s experience of seals at Roancarrig – it was not necessarily so required given the view of the significance of uncertainty in EIA noted above as articulated in Thakeham1496 and Jones.1497 Whether those enquiries were required for EIA purposes would turn on whether ALAB considered that it had enough information already to assess in EIA the risk of impact by seals on the risk of fish escape. However, ALAB was obliged to assess that risk in EIA. As I have said, one would also have expected ALAB to explain why they rejected their own expert advice to seek that information.

998. And as to EIA of the risk of impact by seals on the risk of fish escape, the position remains as described by Dr Saunders December 2020 – as to which see above. In short, it has not been assessed. That is not a ground for certiorari as it was not pleaded. But, as the Aquaculture Licence is to be quashed for other reasons relating to EIA, in any remitted process ALAB will be obliged to conduct a comprehensive EIA, including an assessment of that risk.

ALAB’s AA Conclusion Statement May 2021

999. ALAB’s AA Conclusion Statement starts by recording that it did “An Appropriate Assessment” with contributions from external technical experts including, Alex Coram. It also records that its AA “assessed the potential ecological impacts of aquaculture activities on Natura features in and adjacent to the Natura sites in Bantry Bay” including two SACs, one of which was Glengarriff Harbour and Woodland SAC. As Mr Coram considered only seals, this looks like an assertion that ALAB did AA of risk to seals. As has been seen, that is not so. In fact, ALAB screened out risk to seals for AA purposes in February 2018. And ALAB’s AA thereafter did not include risk to seals. For example, it did not require the NIS to address risk to seals and it did not do so and there was no public participation in AA as to that risk.

1496 R (Thakeham Village Action Ltd) v Horsham District Council [2014] EWHC 67 (Admin). 1497 R (Jones) v Mansfield District Council [2003] EWCA Civ 1408, [2004] 2 P & CR 233.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1000. It is sometimes difficult to get one’s bearings in the AA Conclusion Statement. In part I think that is a matter of terminology. Not incorrectly, ALAB states “The Appropriate Assessment process is divided into a Screening stage and Appropriate Assessment proper”. This may have informed the confusing aspect of the

introduction to the Statement which, as I have said, reads as if ALAB did AA of risk to seals. While it is not

obligatory, I confess to thinking it simpler and clearer if the term “Appropriate Assessment Screening” or “AA

Screening” is used to describe the process of deciding whether to do “Appropriate Assessment” or “AA” – with the term “Appropriate Assessment” or “AA” being reserved for what is sometimes called “full AA” or

“AA proper” or “Stage 2 AA” or the like. This would be rather than using the term “Appropriate Assessment”

or “AA” as referring to a two-stage process encompassing both “Stage 1” AA Screening as well as “Stage 2 AA”. In the AA Conclusion Statement while there is a heading “Screening”, it lies under the heading “Appropriate Assessment” and there is no heading “Stage 2 Appropriate Assessment” or “Appropriate Assessment Proper”1498 or the like.

Only at the conclusion of the Report is there a purported Screening Determination that “Stage 2 Appropriate Assessment” of risk to seals was unnecessary.

1001. ALAB accepted what it summarised as Mr Coram’s conclusion that there was “no evidence of significant negative impacts predicted from the proposed development” and so ALAB determined that the proposed Salmon Farm “has no potential for significant effects and it is not likely to have a significant effect on the Glengarriff Harbour and Woodland SAC ….” For reasons I have noted, this is an inadequately bare summary of Mr Coram’s report. ALAB did not note or consider the significance of Mr Coram’s report that “The possibility of acoustic deterrents causing hearing damage to individuals from the Glengarriff Harbour and Woodland SAC cannot … be excluded.” Nor did ALAB state how, in that light and given the light trigger in AA Screening, it can have been certain beyond reasonable scientific doubt that there would be no adverse effect on seals of the SAC.

1002. ALAB noted the content of the Marine Institute AA Screening Matrix1499 and its “overall conclusions … that the culture of shellfish and finfish, as it is currently constituted and proposed, in Bantry Bay does not pose significant risk to the conservation features (SCIs) of the adjacent sites and as such existing and proposed aquaculture activity does not require a full appropriate assessment.” ALAB also stated that it had “had regard to the June 2018 and September 2020 Screening Reports produced by MI1500 and accepted their Finding of no Significant Impacts for the two relevant SAC sites only – Glengarriff Harbour and Woodland SAC and Sheep’s Head SAC – and determined that the proposed activity at the Site has no potential for significant effects and it is not likely to have a significant effect on either of the SACs either individually or in combination with other sites, plans or projects.” As I have said, the legal and substantive status of the Marine Institute AA Screening Matrix is at best unclear. And, at least in the EU Commission’s view, it would seem to be a matrix summarising a report rather than a report itself.

1498 A phrase used by ALAB in its Statement. 1499 Version 03/09/2020. 1500 Marine Institute.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1003. Under the heading “Conclusion”, ALAB expressed itself satisfied that “given the outcomes of the earlier screening process, the proposed licenced activity will have no significant effect on the integrity of the conservation objectives of” inter alia, Glengarriff Harbour and Woodland SAC (Site Code: 00090). This is

described as following from, inter alia, the “findings of the Marine Institute (2018, 2020)” (i.e. its Matrix). On

this basis ALAB “accepted” that the Salmon Farm “has no potential for significant effects and is not likely to

have an adverse effect on the listed SCI species” of, inter alia, that SAC. So, the AA Conclusion Statement of May 2021 is in fact a purported AA Screening Determination that AA of risk to seals of the SAC was unnecessary. However, it does not record that ALAB had already made that same determination in February 2018 and had done so:

• without reference to the Marine Institute AA Screening Matrix (which had not then been published),

• on foot of the Coram report which, properly interpreted, did not suffice to support a determination to that effect,

• without publishing its determination and reasons as soon as may be after making it.

Indeed, if the AA Conclusion Statement is to be regarded as an AA Screening Determination it follows that the AA Screening Determination post-dated the process on foot of which AA was performed – including, for example the preparation of the NIS.

ALAB’s Determination – as to AA

1004. ALAB’s Determination of the Appeal, dated 29 June 2021, included Part 5 headed “Appropriate Assessment”. Part 5 largely duplicates its AA Conclusion Statement of May 2021. I confess that I am unclear what ALAB intends to be the legal relationship between and difference between the two documents or why there are two. In Part 5 of its Determination, dated 29 June 2021 ALAB,1501

• Recorded that “in accordance with the Recommendations arising from the Oral Hearing Report” it had undertaken “detailed evaluation of the potential threats to seal populations” being qualifying interests of Glengarriff Harbour and Woodland SAC. Specifically, it commissioned the Coram report.

• Deemed the Coram Report a “Seal Screening Report” and accepted its conclusion1502 that the proposed fish farm is not likely to have any deleterious effect on the qualifying features of the Glengarriff Harbour and Woodland SAC and determined that there is “no potential” (not a phrase used by Mr Coram – he said “unlikely”. These seem to me to be two quite different meanings.) for significant effects and it is not likely to have a significant effect on that SAC.1503

1501 I have edited what follows to confine it to materia relating to seals of the Glengarriff Harbour and Woodland SAC.

1502 In fact the reference is to “all of the scientific literature in the material referred to in Part 2 of this Determination” but that includes Mr Coram’s report

– which report is the relevant literature as to seals. 1503 ALAB’s Determination, dated 29 June 2021 §2.16.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Recorded that the Marine Institute “carried out” AA Screening of Aquaculture Activities in Outer Bantry Bay and that ALAB had regard to the Marine Institute’s June 2018 and September 2020 “Screening Reports”1504 and accepted their Finding of no Significant Impacts for Glengarriff Harbour and Woodland SAC and determined that the proposed fish farm has no potential for significant effects and it is not likely to have a significant effect on that SAC.

1005. I observe of that Determination dated 29 June 2021 that,

• it did not record or recognise that in February 2018, ALAB had determined the scope of the AA, screened out risk to seals and required an NIS in terms which did not include consideration of risk to seals.

• its reference to “Screening Reports” is a reference to the Marine Institute Matrix – which did not exist when, in February 2018, it screened out AA of risk to seals.

• it did not refer to or address Mr Coram’s advice that acoustic effects of seal scarers causing hearing damage to seals of the SAC could not be excluded.

• it did not refer to the issue of Aquaculture Stewardship Council certification or any resultant ban on the use of acoustic devices.

The Aquaculture Licence

1006. The Aquaculture Licence does not explicitly address risk to seals or the use of seal scarers or impose by condition a requirement of Aquaculture Stewardship Council certification such as would implicitly prohibit the use of seal scarers. I cannot see on the licence any legal obligation on MOWI to procure Aquaculture Stewardship Council certification for the Salmon Farm.

1007. Condition 3.10 of the Aquaculture Licence very generally requires that “The Licensee shall ensure that any aquaculture or other activity conducted under this licence does not adversely affect the integrity of the Natura 2000 network (if applicable) through …. disturbance of the species for which the area has been designated in so far as such a disturbance may be significant in relation to the stated conservation objectives

of the site concerned.” The reference here to “the area” “designated” is clearly to any relevant Natura 2000

site and not, necessarily, to the licensed area. So the condition applies to effects on seals from the SAC occurring outside the SAC. On the other hand, Condition 4.1 requires “all steps necessary” to prevent fish escape. Very arguably, and if seal predation at Shot Head proved significant, and given they are in

1504 The phrase is that used by ALAB.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“widespread usage” this could include acoustic seal scarers. Indeed it bears observing that Conditions 3.10 and 4.1 are entirely general and say nothing explicit of seals or seal scarers.

Seals – Scarers not proposed – FitzPatrick v An Bord Pleanála & Apple

1008. However, and even in response to Mr Coram’s observation that acoustic damage to seals’ hearing by seal scarers could not be excluded, the bedrock of the defence by ALAB and MOWI in this case was not to address the substantive significance of that observation but to assert that seal scarers did not form part of the proposed Salmon Farm project1505 and so risk of their impact on seals did not require consideration in AA Screening. In other words, Mr Coram’s observation was irrelevant. MOWI submitted, “MOWI did not apply for, nor does the Aquaculture Licence authorise the use of acoustic devices at the fish farm once operational”. This might seem, but is not, inconsistent with the screening out of risk to seals more generally – for example as to other noise impacts or as to impacts within the boundaries of the SAC.

1009. Though an EIA case, FitzPatrick1506 was relied upon by analogy. The Supreme Court in FitzPatrick cited Gulmann AG in Bund Naturschutz in Bayern.1507 He considered whether EIA in a development consent application for a road link which formed part of an overall plan for a longer new road, had to encompass the entire new road of which the road link formed part. Gulmann AG gave a salutary reminder that while the environmentally optimal solution was EIA of the entire, that is not what the law requires. Finlay Geoghegan J observed that Gulmann AG “correctly …… distinguishes between what might be considered environmentally desirable and the option chosen by the EIA Directive”. She held that “The EIA Directive requires an EIA to be carried out of the project or proposed development for which the planning permission is sought.”

1010. In FitzPatrick, the Supreme Court held that EIA of a data centre proposed by Apple need not include EIA of Apple’s masterplan for multiple data centres on the lands of which the site formed part and of which the proposed data centre was to be the first phase. The Court held that the obligation in EIA Law was to perform EIA only of the proposed development for which planning permission was sought and it was permissible to treat that development as a standalone project for EIA purposes where it was not functionally or legally dependent on later phases.

1011. I accept the general analogy as between EIA and AA. But FitzPatrick merits close attention when analogising it to the present case in other aspects:

1505 The Supreme Court noted that the term “project” used in the EIA Directive is not used in the Irish legislation, being replaced in substance by “proposed

development”. It held in effect that the phrases are interchangeable – FitzPatrick §31. 1506 FitzPatrick v An Bord Pleanála & Apple [2019] 3 IR 617.

1507 Case C-396/92 Bund Naturschutz in Bayern v. Freistaat Bayeran, Judgment of 9 August 1994, [1994] E.C.R. I-3717.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• First, the principle stated above is not as stark as might first appear. The Supreme Court in FitzPatrick also held that in EIA of the proposed development account had to be taken, as far as practically possible, of potential later phases of the masterplan. Finlay Geoghegan J held that the fact that the proposed data centre was phase one of the overall masterplan affected the scope of the necessary EIA of the proposed data centre.1508

That is because, as Gulmann AG put it, the purpose of the EIA Directive is to obtain an overview of the effects of the project on the environment at the earliest possible stage in all the technical planning and

decision-making processes and to it so designed as to have the least possible effect on the environment.

• Second, FitzPatrick establishes that the precise manner in which the fact that the proposed data centre was phase one of the masterplan affected the necessary scope of the EIA of the proposed data centre depended on the individual facts and circumstances of the specific project and the overall masterplan.

• Third, and notably given the reliance in this case on the prospect of a process under s.42 of the Wildlife Act to authorise any use of seal scarers, the Supreme Court was not deterred by the prospect of later and inevitable EIA of later phases of the masterplan, from requiring that account be taken, as far as practically possible and in EIA of the proposed data centre, of potential later phases of the masterplan.

• Fourth, we are not concerned in the present case with the prospect of further or future substantive development or, to put the same thing another way, another “project”. We are concerned rather with the prospect of measures to mitigate the environmental effects of specifically the proposed development – the present “project”. It may yet prove that, for the avoidance of salmon escapes and their effects – both commercial and environmental – the development as now proposed is, in greater or lesser degree, functionally dependent on seal scarers. To pursue the analogy, it was important in FitzPatrick that, once built, the data centre could be operated as a single data hall – its operation was not dependent upon the build out of further data centres, as envisaged in the masterplan. But it is not yet clear that the Salmon Farm can be operated without seal scarers to mitigate the risk of salmon escapes due to seal predation.

Indeed, their use is envisaged as at least possibility and we are told nothing of the probability of their use.

Yet MOWI seek to defer consideration of that question until after EIA which, at least ordinarily, includes consideration of measures to mitigate the risks posed by the proposed development being subjected to EIA.

• Fifth, we are concerned, in AA, with the need for reasonable scientific certainty as to the effects of the proposed development on seals as a conservation interest of an SAC.

• Sixth, and even applying the rationale of FitzPatrick by analogy, no explanation appears in any of the papers why it is not “practically possible” to take account in AA in the present case of the effects of seal scarers on seals. There is no description or explanation of any difficulties or uncertainties as yet attending such an analysis. MOWI simply defers its decision on the issue without explaining the need for such

1508 §§52 & 54 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

deferral. Applying the principle that the purpose of the EIA Directive is assessment “at the earliest possible stage” and applying that to the EIA of the risk of farmed salmon escape in the present case and, by analogy to AA in the present case of any adverse effect on seals of steps to mitigate that risk of escape, it seems to me that it was at least necessary that MOWI explain why such analyses were not “practically possible” so that ALAB could actively scrutinise that explanation.

• Seventh, Finlay Geoghegan J in FitzPatrick says, “Just as the manner in which an EIA must be conducted is fact specific, for the reasons already stated, so too is the obligation to take into account, as far as practically possible, the remainder of the masterplan. The precise manner in which that was required to be done depended on the individual facts and circumstances of the specific project and the overall masterplan and the consequences for future phases of the masterplan of a planning consent for phase one and its development.” I form no conclusion on the issue, but the evidence to hand suggests that it was, indeed, “practically possible” to take at least some account in the EIA of the effect of the risk of salmon escape and the need for and modalities of its mitigation:

o The evidence and expert advice to ALAB is clear that seal predation at the Shot Head Salmon Farm is likely – or at very least is foreseeable as an appreciable possibility. Given the clear environmental significance of such an escape if it occurs, it follows that the foreseeable risk of farmed salmon escape must be considered in EIA.

o MOWI is, of course, entitled to decide the acceptability or otherwise of its own commercial risk. But the inference does seem to inevitably follow from the evidence that MOWI may well need, if only for commercial reasons, to take steps to prevent seal predation.

o However, the mitigation of the clearly significant environmental risks of farmed salmon escape, the quantification (insofar as possible) of the resulting residual risk of farmed salmon escape and the assessment of the acceptability of that risk are matters for ALAB’s consideration in EIA of the Salmon Farm project as at present proposed. That implies consideration of the efficacy of such mitigation and so, necessarily, of the modalities of such mitigation.

o It is clear that, in considering steps to prevent predation, MOWI will consider the use of seal scarers.

MOWI expressly says so – though that does not mean it will not reject their use in favour of other modalities such as anti-predation netting.

o MOWI’s own position is not merely that the need to consider use of seal scarers will arise in future: it is that assessment of the possible use of seal scarers will be “necessary” and, what is more, that it will be necessary “early in the development of the site, if the licence is granted.”1509

• Eighth, in EIA, the use of mitigation and assessment of its efficacy is often, even generally, an exercise in prediction. Accordingly it is notable that:

1509 EIS 2011, §3.4.11, p172.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o MOWI has not asserted – much less explained why – the prospects of the use of seal scarers cannot be assessed in the licensing process – especially as it accepts that such assessment is necessary and its deferral is only to a point “early in the development of the site”. The question arises, if so soon, why not now?

o Mr Coram says that seal scarers are in “widespread usage” and he cites a number of papers, some authored by himself, as to their use and their efficacy in deterring predation.1510 o MOWI inevitably has in its possession knowledge as to its long experience in operating its nearby Roancarrig site in the Bay. The tracking data cited by Mr Coram records seals making “extensive use of the existing Roancarrig farm site” – in a “marked association” with that site “probably due to foraging opportunities at the site” – whether predation of “aggregations of wild fish found around farms” or “direct predation on the farmed stock”. Of course, it could turn out to have been either, both or neither but MOWI has not said. But if there has been predation on the farmed stock at Roancarrig, the question would arise whether MOWI used seal scarers and to what effect as to the prevention of escapes and with what, if any discernible, effect on the seals. MOWI chose not to volunteer that information in the licensing process.

o Dr Saunders, notably citing the “ever-present risk of fish escapes” and in the cause of “more robust assessment the risk of seal damage to nets” (though I have found no assessment of that risk at all, robust or otherwise) recommends that ALAB seek from MOWI that information as to seal predation at Roancarrig. I have been unable to ascertain that ALAB considered doing so. In any event it did not happen for reasons not explained.

1012. As will appear from the foregoing, I am quite unconvinced that FitzPatrick requires dismissal of this ground as to AA Screening of any risk posed to seals from the SAC by seal scarers on the posited basis that seal scarers are not part of the “project” for which the licence was sought.

1013. In Ó Gríanna #21511 the applicants in judicial review argued that, for EIA purposes, the planning applicant developer must furnish information in relation to mitigation measures – citing Art. 94 and Schedule

6 §1(c) PDR 2001.1512 §1(c) required that an EIAR contain “A description of the features, if any, of the

1510 They are not before me. He lists various papers the titles of which suggest that they may consider the issue but those specifically entitled as to seal

scarers are:

Coram, A, J Gordon, D Thompson, and S Northridge. 2014. "Evaluating and Assessing the Relative Effectiveness of Acoustic Deterrent Devices and Other

Non-Lethal Measures on Marine Mammals." Scottish Government, 1-145.

Gotz, Thomas, and Vincent M. Janik. 2013. "Acoustic Deterrent Devices to Prevent Pinniped Depredation: Efficiency, Conservation Concerns and Possible

Solutions." Marine Ecology Progress Series 492 (October): 285-302. doi:10.3354/meps10482.

Northridge, Simon, Jonathan Gordon, Cormac Booth, Susannah Calderan, Alexander Cargill, Alexander Coram, Douglas Gillespie, Mike Lonergan, and A

Webb. 2010. "Assessment of the Impacts and Utility of Acoustic Deterrent Devices. Scottish Aquaculture Research Forum SARF044."

1511 Ó Gríanna v An Bord Pleanála [2017] IEHC 7 (High Court, McGovern J, 18 January 2017). 1512 the Planning and Development Regulations 2001 (as amended).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

proposed development and the measures, if any, envisaged to avoid, prevent or reduce and, if possible, offset likely significant adverse effects on the environment of the development.” This requirement derived from the wording of Annex IV §6 of the 2011 EIA Directive1513 save that in §6 the words “if any” do not appear. The

word “envisaged” is interesting. It is not “intended” or “proposed” or “to be effected”. It suggests that a

weaker degree of foresight of a particular mitigation measure will suffice to require its consideration in EIA.

1014. The applicants in judicial review in Ó Gríanna #2 lost, inter alia, on the basis1514 that the EIA Directive is not about formalism, but about providing effective EIA. Its objectives are to protect the environment, but not in absolute terms and the principle of effectiveness is not a mandate for construing the Directive in the most onerous manner possible. A balance is required between ensuring effectiveness and avoiding an overzealous response to cases of minor non-compliance.

1015. Reversing, but I think not unfairly, the order of McGovern J’s observation in Ó Gríanna #2,1515 I note that EIA must not be “overly pedantic” – nonetheless, the EU law principle of effectiveness requires that the courts must be “astute to ensure the objectives of the Directive are met”. I may add that one of those objectives of the EIA Directive, set by Recital 2*, is that environmental effects be “taken into account at the earliest possible stage in all the technical planning and decision-making processes.” Clearly that requirement of the principle of effectiveness applies to directives generally, such that courts must be equally astute to ensure the objectives of the Habitats Directive are met. And, just as clearly, the same obligation devolves on administrative decision-makers such as ALAB: it too must be “astute to ensure the objectives of the Directive are met”.

1016. That must include being astute to ensure that cans of risk are not kicked down the road which should be opened now and examined for worms. I emphasise that I am concerned with MOWI’s actions rather than its motives, which I do not impugn. And I do not assume or suggest that there are worms to be found. But in my view and on the papers before me, the need to assess in EIA the risk of escape of farmed salmon relates to something far from minor and to a risk the necessity to mitigate which is, at least in general terms, foreseeable. Indeed that risk has been expressly seen by MOWI as falling necessarily for consideration “early in the development of the site” and that with a view to considering if seal scarers are required – scarers which Mr Coram says are “in widespread usage”. As I say, if “early in the development of the site”, why not now? There may be good reason but MOWI has not given it or explained why, in FitzPatrick terms, examination of that issue is not “practically possible”.

1513 The 2014 wording is more extensive but need not detain us here.

1514 For which McGovern J cited Berkeley v. Secretary of State for the Environment [2001] 2 A.C. 603, Joined Cases C-128/09, C-129/09, C-130/09, C-

131/09, C-134/09 and C-135/09 Boxus et al v Région Wallonne Opinion of Sharpston AG of 19 May 2011, §79 and Tromans on EIA (2nd edn., Bloomsbury

2012) in turn citing R. v. Rochdale M.B.C., ex parte Tew [2000] Env. L.R. 1 and R. v. Rochdale M.B.C., ex parte Milne [2001] Env. L.R. 406.

1515 At §39.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Proffering of Mitigation, Aquaculture Licence Condition 4, s.42 Wildlife Act 1976 & the concept of Disturbance

1017. Leaving aside any power of the decision-maker to impose mitigation measures which the developer has neither proffered nor agreed, clearly, and generally, it is for the developer to proffer mitigation measures and to accept or reject any others suggested by other participants in the process. It is not, at least generally, open to an objector to say that a project will involve a particular form of mitigation which the developer has neither proffered nor agreed and incorporate that prospect into its opposition to the project.

1018. Of course, cakeism1516 is impermissible. In not proffering or in rejecting a posited mitigation measure, the developer takes its chances. In that event, EIA proceeds on the footing that the mitigation measure in question will not be effected, that the mitigation of risk it would produce will not be realised and that the resultant residual risk may be higher than the decisionmaker considers acceptable, such that a refusal of permission may result. On the other hand, in proffering or in accepting a posited mitigation measure, the developer prompts EIA on a basis which assumes the mitigation measure and the decisionmaker may in consequence consider the resultant residual risk acceptable when, absent the mitigation measure, it would have considered the risk unacceptable. Thus, by proffering or accepting a posited mitigation measure, the developer, ceteris paribus, increases its chance of a permission.

1019. The difficulty arises where, as here, the developer neither proffers nor rejects the mitigation measure but recognises it as potentially required and seeks to defer both a decision on that possibility and an assessment of its environmental consequences, until after the EIA. That is what has been done here. Of course, a mitigation measure may be “envisaged” (the word used in the EIA Directive) as at any point on a spectrum between certainty and very remote possibility. It will be a matter for the decision-maker to judge in EIA, and on the facts and circumstances of the particular case, where on that spectrum that prospect of mitigation lies and to judge any extent to which it is practical to take it into account in EIA on the basis envisaged in FitzPatrick. The underlying principle however is that all environmental effects of a project must be taken into account in EIA – Kokott AG in Namur-Est.

1020. As the Supreme Court said in FitzPatrick, much turns on the facts and circumstances of the particular case. ALAB will have a considerable margin of appreciation accordingly. But in this case, and on the evidence at this point to hand, the risk of seal predation is clearly foreseeable. That implies that, for weighty commercial and environmental reasons, MOWI will have to consider mitigation of that risk. Indeed, inference is unnecessary – MOWI has expressly said it will consider that issue, and soon. And amongst the significant possibilities of mitigation it has said it will consider is the use of seal scarers. That prompts the question whether the prospect of the use of seal scarers is of such degree that AA Screening of their use is

1516 The word now appears in the Oxford, Collins and Cambridge dictionaries.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

required. According to Mr Coram, they are in “widespread usage” – that tends to suggest that their use may be more rather than less likely in the present case.

1021. I may add that, as to its consideration in AA Screening (as opposed to whether AA Screening of the risk should be done), it may be that the observation that seal scarers are in “widespread usage” may be underlain by a perception of absence of adverse acoustic effect on seals. But I cannot so assume. Mr Coram could have said so but did not. He may have so assumed but his report does not disclose the “objective information”1517 on which such an assumption could be based. And, of course, AA arises only as to seals which are conservation interests of SACs as here – it may be that in many locations in which the use of scarers is widespread the seals are not conservation interests of SACs. And in any event, and more importantly, he expressly says that the risk of acoustic damage to seals cannot be excluded.

1022. Condition 4.1 of the Aquaculture Licence obliges MOWI to “take all steps necessary to prevent the escape of fish from its cages/pens”. Condition 4.2 requires compliance with the NASCO Containment Guidelines.1518 For all that they are in general terms,1519 §4.5 of those Guidelines reads: “effective predator deterrence methods shall be implemented as appropriate”. Presumably, in principle, both obligations include the possibility of seal scarers as they are in “widespread usage” for that purpose. That suggests that the possibility of their use would have contributed to a view by ALAB in EIA that any degree of risk of seal predation causing salmon escapes may was acceptable – had that risk been assessed in EIA.

1023. Further, if events prove seal scarers “necessary” within the meaning of Condition 4.1 or even “appropriate” within the meaning of the NASCO Containment Guideline and Condition 4.2, to mitigate the risk of farmed salmon escape, but a licence pursuant to s.42 of the Wildlife Act 1976 is refused (for example on AA grounds of risk of acoustic damage to seals), that may place MOWI in breach of the Aquaculture Licence. However, for present purposes and in the abstract, I need not tease out that possibility further.

1024. As to ALAB’s reassurance, and DAFM’s,1520 that if seal scarers were licenced under s.42 of the Wildlife Act 1976 any risk to seals would be “assessed in that process” and “any mitigation measures specified”, the submissions in opposition to the judicial review were strangely underdeveloped. Ss.2 and 23 and the 5th Schedule to the 1976 Act combine to designate seals as a “protected wild animal” and prohibit their injury. But s.23(7)(iv) provides that s.23 does not make unlawful anything done pursuant to a permission granted pursuant to the Wildlife Acts.1521 S.42 permits the Minister to permit such steps,

1517 Case C-127/02 Waddenzee.

1518 “the most up to date guidelines on fish containment developed by the North Atlantic Salmon Farming Industry and the North Atlantic Salmon

Conservation Organisation (NASCO) Liaison Group.” 1519 See above. 1520 EIA 12 June 2015 p14.

1521 S.23(7A) excludes from the scope of s.23(7) species strictly protected under Article 12 of the Habitats Directive – but, as has been noted, the common

seal is not so protected. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

including by scaring by specified means, as he thinks appropriate to stop serious damage being caused by protected wild animals to aquaculture installations.

1025. In passing one may observe that the amendment1522 of s.42 to provide for licenses to prevent serious damage to aquaculture installations and to do so by scaring presumably reflects an appreciation that serious damage to aquaculture installations is an eventuality sufficiently foreseeable as to require such a statutory amendment. While the amendment was not specific to seals, and might, one imagines, include birds, on the evidence before me it is not difficult to infer that, perhaps inter alia, the amendment was made with seals in mind.

1026. S.42 does not prescribe evaluation in licence applications of effects on the protected wild animals in question and, notably, such steps may include killing the wild animal in question – though that will not arise as to seal scarers. However, despite the absence of provisions in s.42 or regulations thereunder specifying the criteria by which the Minister is to consider applications under s.42, it would not seem difficult to infer that the Minister must consider whether the derogation from protection and the resultant effect on the protected animal is warranted by the seriousness of the damage sought to be prevented and is kept to the minimum required for that purpose.

1027. But even if s.42 does imply evaluation of effects on the protected wild animals in question, that is not the point. What is required, at least in the present case as to the seals of the SAC, is not a general evaluation or assessment but specifically AA Screening and, if required, AA. My attention was not drawn to any statute or regulation prescribing, as to licensing applications under s.42 of the Wildlife Act 1976, evaluation of such effects by way specifically of AA and/or AA Screening. That issue would appear to specifically arise as to animals not protected by Art. 12 of the Habitats Directive1523 but protected by the conservation objectives of an SAC – as is the case of the seals of the Glengarriff Harbour and Woodland SAC.

1028. I am far from predicting that the outcome of any AA of the use of seal scarers would prevent their use. And it may be that on the evidence before ALAB when it comes to make a remitted decision it will consider that seal scarers will not be used and impose a licence condition preventing their use. If so, it will need to consider accordingly in EIA the risk of fish escape due to seal predation in the absence of that particular form of mitigation. All that is for ALAB.

1029. But by way of obiter and analogy and while seals are not protected under Art 12 of the Habitats Directive, I note that the concept of disturbance of protected species was considered in Shadowmill.1524 I

1522 Inserted (31.07.2001) by Wildlife (Amendment) Act 2000 (38/2000), s.48(a)(iii). 1523 Animals so protected are excluded from the scope of s.42. 1524 Shadowmill v An Bord Pleanála & Lilacstone [2023] IEHC 157, §122 et seq.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

need not repeat that consideration here. Suffice it to say that not all “disturbance” colloquially so-called is disturbance within the meaning of Article 12 of the Habitats Directive and whether “negative impact” and/or “harm” are likely are relevant to the presence or absence of disturbance within the meaning of Article 12. Also, “sporadic disturbances without any likely negative impact on the individual animal or local population … should not be considered as disturbance under Article 12.” Morge1525 and Skydda Skogen1526 were cited to the effect that disturbance may properly be considered in terms of effect on “local populations” of the species concerned. These limitations on the concept of disturbance as applicable the species benefitting from strict protection seem to me applicable a fortiori to species such as seals which do not so benefit. This is not to ignore that the criterion in AA and AA screening relates to adverse effect on the integrity of the European site having regard to its particular conservation objectives. But it seems at least generally unlikely that the test of disturbance in AA should be more demanding that in the context of strict protection. The integrity of a European Site relates to its preservation at a favourable conservation status, which entails the lasting preservation of its constitutive characteristics the preservation of which was the objective of its designation as a European Site – Sweetman,1527 Balz1528 and Holohan.1529 While a demanding criterion, it is no doubt a matter of at least some degree, and any scientific doubt as to the matter must be “reasonable” for it to require refusal of development consent. Indeed, that the disturbance matters only if it is such as to adversely affect the integrity of the European site may imply a somewhat less demanding standard. Though, all that said, it is notable that the conservation objectives of the Glengarriff Harbour and Woodland SAC extend to the protection of individual seals.1530

Hellfire Massy

1030. A somewhat similar issue arose in the Hellfire Massy litigation. After the trial of the present case, the CJEU judgment1531 confirmed that where a project subject to EIA in the development consent proves “involves” the developer obtaining an Article 16 derogation from an authority other than that competent in EIA, the potential derogation must be adopted before the development consent is given as otherwise, it would be given on an incomplete basis. In other words, the EIA would be incomplete. The CJEU recited the well-known objective in EIA that effects on the environment should be taken into account by EIA by way of a “full assessment” and “at the earliest possible stage in all the technical planning and decision-making processes”, in accordance with, inter alia, the precautionary principle. So the outcome of an EIA must make it possible to determine whether the project was likely to have effects prohibited by Article 12 of the Habitats Directive.

1525

R(Morge) v Hampshire County Council [2011] 1 WLR 268. See also, Simons on Planning Law 3rd Ed’n (Browne) §15–895 et seq.

1526 Case C-473/19 Föreningen Skydda Skogen, Judgment of 4 March 2021 – as it happens a tree-felling case, though not concerned with bats.

1527 Case C-258/11 Sweetman & Others v An Bord Pleanála, Judgment of 11 April 2013, ECLI:EU:C:2012:743. 1528 Balz v An Bord Pleanála [2016] IEHC 134. 1529 Case C-461/17 Holohan v An Bord Pleanála [2019] PTSR 1054.

1530 Coram Seal Report p8. Cites the SAC conservation objectives to the effect that man-made noise should not be introduced at levels that “could result

in a significant negative impact on Individuals and/or population of harbour seal ….”. 1531 Case C-166/22 Hellfire Massy Residents Association v An Bord Pleanála, Judgment of 6 July 2023.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1031. Given the judgment post-dated trial in the present case, it is important to say that the CJEU in Hellfire Massy did not identify new law. It proceeded to judgment without an advocate general’s opinion1532 and on the explicit basis that the conclusions identified above followed from Namur-Est1533 and the caselaw cited therein.

1032. However the CJEU in Hellfire Massy also noted that the Irish High Court in 20211534 had held on the facts that, at the time when the impugned development consent was given, the need to obtain a derogation had not been “identified”. It followed that the situation in which a derogation is required before development consent is given, had not arisen. Presumably the CJEU would say that it had not been shown that the project involved actions which would require a derogation. That rationale might be thought to favour ALAB in this case. As Hellfire Massy was grounded in findings of fact made in 2021, I note that Humphreys J had at that time,

• identified a “significant distinction between matters requiring a derogation licence that “arise”1535 before

consent and after consent – between a situation in which the likelihood that a derogation licence will be required “emerges” before consent is granted and one in which before consent is granted “a derogation licence is only a possibility and in the event turns out to be needed as a result of a post-consent survey”.1536

• held that in the first situation an “argument for a procedure that should have been applied at that point”1537 could go to certiorari. The second situation did not go to certiorari.

• held, as to an argument relating to bats, that “the factual context is such that there isn’t a current established or likely requirement for a derogation licence”.1538 Accordingly, the development consent was not “premised on there being any current definite or likely intention to seek a derogation licence”. And “… the board did not rely on the prospect of grant of a derogation licence for anything that was being specifically authorised that was known about as of the time of the development consent.”1539

• rejected an argument that that the board should have anticipated possible post-consent problems in the

decision and introduced a particular condition to deal with any need for post consent derogation licences.

But Humphreys J rejected that argument not in substance but because it was not pleaded.1540

1532 Generally an indication that the point is not considered novel or difficult. 1533

Namur-Est Environnement, C-463/20, EU:C:2022:121, judgment of 24 February 2022, §§ 52 and 59 and the case-law cited therein.

1534 Hellfire Massy Residents Association v. An Bord Pleanála [2021] IEHC 424 §61. 1535 My parentheses. 1536 §77. 1537 i.e. a derogation licence application before completion of EIA and the grant of development consent. 1538 §78. 1539 §61. 1540 §61.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Found, on the other hand, that the development lands were to an extent adjacent lands the habitat of strictly protected species1541 – bats, red squirrels and otters. There was a reasonable possibility that further surveys to be done between the grant of consent and the conclusion of construction might uncover further impacts on such strictly protected species. The EIAR properly anticipated that “should any bat roost be detected during the pre-construction survey which will be disturbed or lost during construction, a derogation licence will be required.”

1033. In short, the distinction made by Humphreys J seems to turn on whether a need for a derogation at the time of the EIA is foreseeable as merely possible or as probable. I leave to another case whether this distinction is compatible with the degree of foresight required by FitzPatrick, such that EIA must take account of future events as far as practicable (it is unclear what the probability was of the masterplan in that case being fully built out), though the judgment of the CJEU in Hellfire Massy suggests it is. However, and perfectly properly, Humphreys J in Hellfire Massy did not in his judgment analyse in detail the facts which lead him to the view that the need for a derogation licence was merely possible not probable. I have no doubt that he would not wish his use of the words “arises” and “emerges” to be understood as denoting a “hands-off”, as opposed to an investigative, approach by the competent authority in EIA and AA screening or as absolving developers from providing all information reasonably required to enable the competent authority to discern whether the prospect of a derogation was a probability or merely a possibility. The least that can be said is that the need for a derogation does not usually, and generally should not if appropriate investigations are done (though it may), arise or emerge as a surprise after development consent has been granted.

1034. Accordingly, I respectfully reject ALAB’s submission that the effecting of a measure in mitigation of an environmental risk (here, of farmed salmon escape by seal predation) and which itself may pose a different environmental risk (here, of acoustic damage to seals by seal scarers) must be “certain” to happen before it arises for AA Screening before development consent is granted. In my view where, as here, the developer has explicitly envisaged a mitigation measure (seal scarers) or where that possibility is otherwise apparent, the developer is obliged to provide the information necessary, and the competent authority is obliged to investigate, to ascertain the probability that the mitigation measure will be effected. Any other view would facilitate evasion of timely and complete assessment in EIA and AA. To put it another way, a developer may not artificially define its “project” so as to exclude an element of that project likely to cause significant effect, where in reality the effecting of that element is probable (albeit on foot of a further permission process) and hence exclude that effect from assessment. That seems to me to be the logic of Namur-Est. Though I do not suggest that such was the purpose here, the effect is what matters.

1035. It seems to me that essentially the same principles must apply to the prospect of a licence under s.42 of the Wildlife Act 1976 as apply to the prospect of a derogation under Art 16 of the Habitats Directive.

1541 i.e. protected by Art 12 of the Habitats Directive.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Indeed, they are very similar as both are derogations from species protection – albeit the protection derives from different sources.

1036. As observed above, decision-makers such as ALAB must be “astute to ensure the objectives of the

Directive are met”.1542 In EIA those objectives include “full assessment” of “all significant effects” at the

earliest stage in the process, including those deriving from subsequent statutory processes such as derogations. As AG Kokott said in Namur-Est – “No exceptions are provided for certain environmental effects.” That astuteness is also required in AA Screening and AA. Accordingly, as observed in Fernleigh,1543 “the Board and the Courts must be alert to any possibility of an approach of not, …. turning over stones for fear of what might be found under them and have to be put before the Board when, otherwise, it might not reach the Board”. I do not suggest any contrivance in this case but astuteness includes positive obligations of inquiry by the decision-maker. As the regulations make clear in empowering decision-makers to seek further information and as ALAB, to its credit, repeatedly demonstrated in this case, the decision-maker is not merely the passive recipient of information which the developer choses to provide and is not bound by a developer’s choice to omit information which it is in a position to provide.

1037. Namur-Est confirms that a developer’s choice to omit information cannot be justified merely by the prospect that the environmental issues it might raise can be considered in a later process. That they can be considered in such a process does not, per se, imply that they need not be considered in EIA or AA Screening. Indeed, the requirements of early and comprehensive EIA generally imply the opposite – though, as Hellfire Massy illustrates, the question will in the end turn on the facts. MOWI gave no explanation or justification of why the prospect of seal scarers could not or should not be considered in EIA and AA Screening prior to the

decision to grant the aquaculture licence and why that “necessary”1544 consideration should be deferred to a point “early in the development of the site”. Indeed, Mr Coram clearly considered it possible to address the issue at least to the point of observing that the risk of acoustic damage by seal scarers to seals could not be excluded. It remains unknown why ALAB did not take its own technical advisor’s advice to get from MOWI the information MOWI must have as to whether it has had seal predation issues and/or has had to deter seals, at Roancarraig and, if so, by what means and with what success.

1038. Though not necessary to my conclusion, it seems to me also that the Sweetman Applicants’ citation of Art. 42(17) of the Habitats Regulations 2011 is apposite at least as reflecting the underlying principle. It provides that “A public authority shall not adopt or undertake, or grant any consent for, a plan or project containing any conditions, restrictions or requirements purporting to (i) permit the deferral of the collection of information required for a screening for Appropriate Assessment or for an Appropriate Assessment or the completion of a screening for Appropriate Assessment or an Appropriate Assessment until after the consent has been given …”

1542 Ó Gríanna v An Bord Pleanála [2017] IEHC 7. 1543 Fernleigh Residents Association v An Bord Pleanála [2023] IEHC 525. 1544 MOWI’s word.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1039. To adopt the distinction drawn by Humphreys J in Hellfire Massy in 2021, we know that seal predation and seal scarers are possible at Shot Head Salmon Farm but there does not appear to have been any investigation whether they are probable or as to what extent, per FitzPatrick, those prospects could be taken into account as far as practicable in EIA or be subjected to AA Screening or AA – as to both the risk of resultant escape of salmon from the farm and the risk of acoustic damage to seals. This is all the more surprising given Dr Saunders’ explicit advice that such inquiry be made as to MOWI’s experience of seals at Roancarraig.

Seals – Conclusion

1040. I am conscious that, in considering a ground pleaded only as to AA, I have said a lot of EIA and, as it were, hopped between the two. That proceeds in part from the Board’s reliance, as to AA, on FitzPatrick, which is an EIA case. But it proceeds also from the reality that, on the particular facts of this case, they cannot be disentangled. Unusually, though I imagine not uniquely, mitigation relevant to EIA may itself have negative effects requiring AA Screening. EIA as to risk of salmon escape due to seal predation and the alleged need for AA of any adverse impact on seals of the use of seal scarers are closely linked in that the use of seal scarers and any risk thereby to seals will arise, if it does arise, by way of mitigation of the risk of salmon escape which falls for consideration in EIA. In short, any likelihood of the use of seal scarers (the AA screening issue) proceeds from the risk of salmon escapes due to seal predation (an EIA issue). Though introduced in 2014, it seems to me that Recital (11) of the EIA Directive was merely declaratory of existing law in linking mitigation in EIA and AA. It states that “The measures taken to avoid, prevent, reduce and, if possible, offset significant adverse effects on the environment, in particular on species and habitats protected under” the Habitats Directive “should contribute to avoiding any deterioration in the quality of the environment and any net loss of biodiversity …”.

1041. As stated above, I will quash the Aquaculture Licence on other EIA grounds. My observations as to EIA of the risk of salmon escapes due to seal predation will, I hope and if only as obiter, assist ALAB in any consideration of this matter on remittal.

1042. Turning to the pleaded AA issues, the unusual circumstances, include,

• That the information before ALAB when it screened out AA of risk to seals in February 2018 clearly was, and when it made its impugned determination in substance was, that, as stated in MOWI’s 2011 EIS, “there is a likelihood that seals will visit the site. It will therefore be necessary to assess whether or not anti-predator nets or even seal scarers will be needed to protect the stock from seal attack early in the development of the site, if the licence is granted.”

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• That, as argued at trial in defence of the impugned Aquaculture Licence, very heavy reliance was placed on o the fact, which I accept, that any future use of seal scarers would require permission under s.42 of the Wildlife Act 1976 and o the assertion, which was notably underdeveloped, that any impact on seals would be “assessed” in that process.

• ALAB’s reliance in February 2018 and in screening out AA of risk to seals, on a Coram report which, as to seal scarers, did not in law bear that weight for reasons set out above.

• ALAB’s failure, in particular, to address the question of the significance of Mr Coram’s report that acoustic

effects of seal scarers on seals of the SAC could not be excluded.

• ALAB’s confusion as to the point in time at which, and identification of the decision by which, it screened out AA of risks to seals of the SAC due to seal scarers.

• ALAB’s reliance, in its later purported AA screening decision, on a Marine Institute Matrix o the legal status of and substance of which remains unclear o which had not been published at the date of its actual AA screening decision of February 2018.

• MOWI’s failure to justify deferring EIA and AA of the use of seal scarers until a point after – and only shortly after, development consent and failure to submit information in its possession bearing on the likelihood of the use of seal scarers and ALAB’s failure to investigate that likelihood – including to consider its Technical Advisor’s advice that that information be bespoken.

1043. It is important to recall that though AA Screening and AA are distinct processes, the question they pose is ultimately the same. As the EU Commission observe1545 and as was noted in Heather Hill #21546 the same level of certainty of conclusion – no reasonable doubt as to absence of likely significant effects – is required in both AA Screening and AA. They differ in essence as to the process required to arrive at that

conclusion, if it is to be arrived at. As was said by Kokott AG in Eco Advocacy,1547 in deciding in AA Screening that AA is unnecessary, a competent authority must at least provide an express detailed and comprehensible statement of reasons, recognisable as such and capable of removing all reasonable scientific doubt concerning the harmful effects of the works envisaged on the integrity of the protected site concerned. She said: “the criteria governing the screening must be just as strict as the criteria for the appropriate assessment itself”.

1545 Commission notice: Assessment of plans and projects in relation to Natura 2000 sites – Methodological guidance on Article 6(3) and (4) of the

Habitats Directive 92/43/EEC Brussels, C (2021) 6913 final, 28.9.2021, p10 Box 1. 1546 Heather Hill Management Company CLG v. An Bord Pleanála [2022] IEHC 146 §249.

1547 Case C-721/21, Eco Advocacy CLG v An Bord Pleanála, Opinion of Kokott AG of 19 January 2023 §§89, 90 & 109(5) & (6). This view was essentially

adopted by the CJEU in its judgment of 15 June 2023. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1044. Accordingly, I take the view that ALAB erred in law in screening out AA of impact of the salmon farm on seals in that it did not remove all reasonable scientific doubt as to the effects of the proposed works as

they relate, or may relate, to the effects of seal scarers on seals of the SAC – as to both the likelihood of their

use and their effect on seals if used. It was not entitled to defer consideration of that issue to a process under s.42 of the Wildlife Act 1976 and to the quite non-specifically posited assessment of risk to seals in that process.

1045. The Impugned Aquaculture Licence will be quashed accordingly. In this respect and given that it seems that explicit evidence is likely to be tendered by MOWI to ALAB in which it commits to not using seal scarers, I am particularly, even if provisionally pending submissions, of the view that the matter should be remitted to ALAB for re-decision. That will enable ALAB also to consider whether a condition prohibiting the use of seal scarers should be imposed in any aquaculture licence which may issue from the remitted process and what, if any, implications the certainty (if that is what transpires) of non-use of seal scarers may have for EIA of the risk of escape of farmed salmon due to seal predation. If, on the other hand, ALAB decide against prohibition of seal scarers, the question will arise in AA screening of consideration of acoustic risk to seals.

OTTERS

Otters – Introduction

1046. The otter1548 is listed in Annex II of the Habitats Directive as a species whose conservation requires the designation of their habitats as Special Areas of Conservation. The otter is a qualifying interest of the Glengarriff Harbour and Woodland SAC. Unlike the seal it is listed also in Annex IV as a species requiring strict protection under Article 12 of the Habitats Directive. Otters eat, inter alia, salmon.

1047. The chair of the oral hearing, in his report of 22 December 2017 noted submissions,1549 that, as well as in the Glengarriff Harbour and Woodland SAC, otters were to be found in the Dromagowlane/Trafrask catchment and in the vicinity of the Shot Head site and that a decline in the wild salmon population due to the fish farm would adversely affect them such that, in EIA, assessment was required of that effect on the Dromagowlane/Trafrask otters and AA, or at least AA Screening, was required of that effect on the SAC otters. The chair advised ALAB that, as to otters from the SAC, for a licence to be granted it must be established (in AA Screening or AA) beyond reasonable scientific doubt that the project will not impact otters to the extent that it affects the integrity of the site having regard to its conservation objectives. The standard of proof required in EIA as to the Dromagowlane/Trafrask otters was less clear. He recommended a desk-top study, which may indicate the need for supplemental AA screening, of the Dromagowlane/Trafrask otter

1548 Lutra Lutra. 1549 By SWI, An Taisce, and Save Bantry Bay.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

population, and (if necessary) assessment of potential impacts on otters, including the potential impact of declining wild salmon stocks. In passing, I observe that this is somewhat odd as effect on the Dromagowlane/Trafrask otters, not being in an SAC, is an EIA issue and an issue of strict protection under Article 12 of the Habitats Directive, not an AA issue. He did not recommend a study of the SAC otters, which would have been an AA issue.

1048. ALAB asked Dr Saunders to do the desk-top study of otters. His report concluded that otters are omnivorous generalists. They eat significant numbers of salmon but in response to availability, rather than as a dietary necessity. Given the otter's clear diet plasticity and resulting adaptability in accommodating prey availability, it would be logical to expect that during a theoretical scenario in which there is a fish farm-linked decline in wild salmonids, otters would be unlikely to be significantly affected. Indeed, throughout the period when wild salmonid populations were declining, survey data indicated either a stable or increasing trend for Irish otter populations and a papers is cited to the effect that "There is no evidence that declines in the number of salmon returning to spawn have any impact on otter numbers''.

1049. Dr Saunders noted that otters are known to cause loss of stock in marine salmon farms, but are, in general, considerably less of a problem than seals. Otters are highly tolerant of human activity and activity around Shot Head fish cages, over 200m from the shore on which otters might be present, is highly unlikely to have any impact on otter activity. (Dr Saunders’ Final Report1550 explained that “Otters have been recorded along the shoreline close to the licence area but as the marine activity of otters is generally confined to the shore and adjacent shallow waters, the disturbance from the fish farm will be insignificant.”) Given otters’ coastal range of around 4-5 km, that the SAC is 12km from Shot Head and that significant sections of the coastline between then constitutes exposed rocky cliffs or very steep escarpments unsuitable for otter foraging and which would present significant challenges for roaming individuals, there is “no discernible mechanism for a direct population impact on the SAC.”

1050. Dr Saunders’ advice is that “on the basis of the overwhelming scientific evidence, it can only be concluded that the operation of a fish farm at Shot Head is highly unlikely to have any detrimental impact on the otter population within the Glengarriff Harbour and Woodland SAC. Or, indeed, throughout the entirety of the Bantry Bay catchment. It follows that the commissioning of an otter field survey, while doubtless providing useful scientific information on the distribution of the species, is unlikely to supply any further compelling evidence in support of a refusal to grant the Shot Head licence.” When the NPWS was invited to comment, it cited the Saunders Otter report and said it had no comment.1551 While one might doubt the correctness of a standard requiring “compelling” evidence, reading the report as a whole, the conclusion clearly supported screening out effect on otters in both EIA and AA. ALAB accepted Dr Saunders’ conclusion. It is difficult to see that it was not entitled to do so. This seems to me to be an impermissible instance of an

1550 8 December 2018 §6.5.2. Similar content is found at §9.8.2. 1551 NPWS S.47 Reply to ALAB 16 November 2018.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

attempt to impose a “Johnson” burden on the Board – i.e. “Make the Board deny it” – as to a hypothetical

risk unsupported by any evidence – Heather Hill #2.1552

Otters – Pleadings

1051. SWI’s grounds narratively refer to consideration of otters in the process before ALAB but do not bring it home to specific grounds on which relief might be granted. The IFI pleads nothing as to otters.

1052. The Sweetman Applicants plead that the Saunders’ otter report was part of ALAB’s general “patching up” of MOWI’s licence application. I mention this for completeness but need not address it here.

1053. The Sweetman Applicants plead that absent “information about the likely effects on otters of an accidental escape of significant quantities of recently vaccinated or otherwise toxic fish, the screening could not have been complete. It is not good enough to say that the otter can eat fish other than wild salmon if salmon stocks decline and without also considering the impacts of consumption of escaped farm salmon.”1553 MOWI and ALAB took pleading objections that the issue as to otters’ consumption of supposedly toxic escaped salmon was pleaded in the “Factual Background” at §83 of the Further Amended Statement of Grounds and is not as a legal ground of challenge and in any event was not sufficiently particularised. Those pleadings points were conceded. I need consider the matter no further.

1054. The Sweetman Applicants also plead that the “The non-technical summary of the 2011 EIS was not amended and the opportunity for the public to question the author of this report at the Oral Hearing had passed.” As this plea flowed from the plea as to consumption of supposedly toxic escaped salmon, it probably falls with the pleading concession. But lest I am wrong, the plea can be rejected briefly.

• First, the EIS is the start of the EIA process – which is iterative (Kemper1554). It is a commonplace that further information later informs the EIA. There is no obligation to rewrite the EIS in light of that further information.

• Second, while cross-examination properly occurs at oral hearings, the process remains inquisitorial not adversarial. There is a right to be heard but no general right of cross-examination of rapporteurs outside of the oral hearing process. The various objectors were heard as to otters at the oral hearing. The Saunders Otter Report was circulated1555 to all parties for comment. Mr O’Keeffe replied disagreeing with Dr Saunders. I cannot see any unfairness here.

1552 Heather Hill Management Company CLG v An Bord Pleanála & Ors [2022] IEHC 146, §268. Also Environmental Trust Ireland v An Bord Pleanála & Ors

[2022] IEHC 540: see §247, §255 and §280. 1553 Grounds §E2.83.

1554Joyce Kemper v An Bord Pleanála [2020] IEHC 601 §237. “the EIA does not begin and end with the submission by the developer of the EIAR. Rather

the process is an iterative one ….” 1555 By letters dated 10 April 2018 under s.46 of the 1997 Act.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1055. I reject all challenges as to assessment of effect on otters.

BIRDS

1056. ALAB determined1556 in AA,1557 as to SPA SCIs1558 Fulmar, Gannet and Guillemot that the Shot Head salmon farm would not impact adversely on SCIs species or SPA conservation objectives and so would not, either individually or in combination with other plans or projects, adversely affect the integrity of the SPAs

Birds – Pleadings

1057. SWI’s grounds narratively refer to consideration of birds in the process before ALAB and frame it as part of its more general plea that ALAB bespoke information beyond that necessary to decide the licence application. But SWI plead no substantive grounds, particular to birds, on which relief might be granted. IFI pleads nothing as to birds.

1058. The Sweetman Applicants plead the following as to birds: that ALAB,

• contravened Art. 6(3) of the Habitats Directive in failing to give explicit and detailed reasons for rejecting Dr Gittings’ findings as to the effect of navigational lighting on the Storm Petrel from the Bull and The Cow Rocks SPA, outside Bantry Bay, in breach of the requirement for such reasons per Case C-461/17 Holohan.1559

• failed, without having site-specific bird counts, to perform a proper AA Screening view of best scientific knowledge and, despite the “screening in” of the Storm Petrel by Dr Gittings, failed to require an AA as to significant effect on the Storm Petrel of that SPA and so requested an NIS from MOWI which didn’t require consideration of the Storm Petrel.1560

• breached Articles 3 and Articles 5 to 11 of the EIA Directive, in failing to notify the public, prior to making its determination, of the DMP1561 statistical analysis of net entanglement mortality in Gannets from Bull and Cow Rocks SPA.1562

1556 ALAB Determination §§5.6.4, 5.6.8 & 5.7. 1557 See Article 6(3) of the Habitats Directive. 1558 Species Of Conservation Interest for which relevant Special Protection Areas had been designated. 1559 Grounds E13 & §E2.31. 1560 Grounds §E2.26, §E2.27, §E2.28 & §E2.29. 1561 DMP Statistical Solutions UK Ltd. 1562 §E2.36.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Birds – Gittings & Crowe Reports

1059. The November 2017 Oral Hearing Report recommended a desk-top study of the possibility of a need for “supplemental” AA screening of potential impacts on wild bird SCIs1563 in nearby SPAs. ALAB retained Dr Tom Gittings, whose desk-top Bird Impact Assessment in February 2018 considered1564 the 2011 EIS and the Minister’s EIA to be generally flawed as to wild birds, including non-SPA birds. He also concluded that further AA screening was required – though he considered that his assessment largely contained the information required for that AA screening.

1060. Dr Gittings noted that the EIS states that dark-coloured top nets will be used to protect the fish against bird predation. Seabirds can die by entanglement in underwater or above-water nets of marine fish farms. Therefore, there is a risk that the proposed fish farm will cause seabird mortality by entanglement in nets and there is some evidence of Gannet becoming entangled in such nets. He says that a degree of caution must be attached to anecdotal evidence that seabird mortality by entanglement offshore salmon cages in Ireland is not a significant issue absent published data on the issue. But Dr Gittings cites academic papers for a general consensus that modern well-managed facilities using appropriate equipment and predator deterrence should have no negative effects on birds.

1061. Dr Gittings noted that the EIS records that the site will have navigational lighting to prevent collisions by vessels but gives no detail of the types and positions of that lighting. Night flying birds, including the Storm Petrel can be killed, after being attracted to such lights, by grounding, or by colliding with structures. The overall magnitude and significance of any such impacts will depend on the position of the Site in relation to bird foraging areas and commuting routes and details of the lighting design and position. Killing of SCIs as a means of predation control would be illegal and, it is assumed, will not occur.

1062. Dr Gittings embarked on a detailed analysis of the vulnerability and sensitivity of Storm Petrels to marine fish farms generally and to the Site specifically given its inshore location and past survey work in Bantry Bay. I need not detail that analysis here. Dr Gittings could not rule out the possibility that the proposed fish farm Site is within the core foraging range of the Storm Petrel SCI of the Bull and the Cow

1563 Species of Conservation Interest, for which, inter alia, the SPA is designated.

1564 ALAB’s brief to Dr Gittings was for “a desktop review of the potential for adverse impact(s) from the proposed fish form on the adjacent SPAS and

their respective bird qualifying interests and … expert advice on possible requirement for an Appropriate Assessment under the terms of the Habitats

Directive. This should include:

• A review of the designated SPAS adjacent to, or within close proximity to, Shot Head, with due regard for bird mobility in respect of the distance to the

proposed fish form site;

• An assessment of the vulnerability of the species of interest, for which each identified site is designated, to salmon aquaculture activity at Shot Head,

• An evaluation of the potential cumulative or combined impacts of the wider aquaculture activity in Bantry Bay, with on assessment of the contribution

to direct and indirect adverse impacts (if any) that the additional Shot Head fish farm is likely to make on the bird resource.

• An evaluation of the existing EIS and EIA and in the context of the requirement (or not) of an Appropriate Assessment consistent with Article 6(3) and

6(4) of the Habitats Directive (92/43/EEC), providing an opinion on whether further or supplementary screening is appropriate.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Rocks SPA.1565 But he concluded for a variety of given reasons that, overall, “it is very unlikely that there is any significant spatial overlap between foraging Storm Petrels from the Bull and the Cow Rocks SPA and the proposed fish farm Site.” Necessarily, this bears upon both any risk of entanglement and any risk due to navigations lights as if the Petrel is not present it cannot fall victim to either. Dr Gittings considered the Salmon Farm unlikely to adversely affect the Storm Petrel. In considering the contribution of the proposed Shot Head farm to potential cumulative or combined impacts of the wider aquaculture activity in Bantry Bay, Dr Gittings excludes the Storm Petrel from his consideration of, as he says it is “unlikely to regularly occur within Bantry Bay”.

1063. Dr Gittings considered the Salmon Farm to be well within the likely core foraging range distance of the Gannets of the Bull and the Cow Rocks SPA. He considered the risk of Gannets entangling with the nets to be low but that Gannet populations could be sensitive to even a low number of mortalities.

1064. In Dr Gittings’ conclusions as to “AA Requirements”,1566 he recommended “further AA screening” only of the Gannet SCI of the Bull and the Cow Rocks SPA as AA of the Gannet “may be required”. He did not recommend AA or “further AA screening” of the Storm Petrel.

1065. Some confusion may have been caused by Dr Gittings’ observation,1567 after he had considered the Storm Petrels of the nearer Bull and Cow Rocks SPA, that no additional consideration need be given to the Storm Petrels of Deenish Island SPA, Puffin Island SPA and the Skelligs SPA, as the Storm Petrel was “already screened in for assessment”. However I am satisfied that while the phrasing was unfortunate, Dr Gittings was here referring to his own assessment and to the fact that, whereas he had screened out certain birds from his own assessment (Peregrine and Chough), he had not so screened out and had already assessed the Storm Petrel. This reference does not mean he had screened the Storm Petrel in for AA Screening or AA.

1066. After getting the Gittings report, ALAB sought various views, including those of the Marine Institute and the NPWS. Ultimately ALAB on 22 January 2019 approved terms of reference for an independent supplementary AA Screening. The AA screening report of Dr Olivia Crowe dated April 2019 ensued. She considered, inter alia, the Gittings report.

1067. The Sweetman Applicants1568 unfairly criticise Dr Crowe as not having addressed the possibility of AA of risk to the Storm Petrel “save in the context of her assessment that the Storm Petrel had not been

1565 The Site is around 45 km from the Cow Rock. He cites an academic paper which gives a maximum foraging range of > 65 km but gives no information

on mean foraging ranges. 1566 at p27 of his report. 1567 at p10 of his report. 1568 Affidavit of Noel Carr, 27 September 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

recorded in this part of Bantry Bay, according to a survey carried out in 1995.” This is a regrettable misquotation of Dr Crowe and mischaracterisation of her opinion. In fact, she said something appreciably different. She said that the Storm Petrel is a “highly pelagic species with very few records within Bantry Bay, the last during ESAS surveys in 1995. It is highly unlikely that there would be spatial overlap between this species and the proposed development site.”

1068. Her observation was not confined to “part” of Bantry Bay and 1995 is not the date of a survey in which “the Storm Petrel had not been recorded” . 1995 is the date on which the Storm Petrel was last recorded – and is that of the “very few” records if its presence extant. The criticism also ignores her view that the Storm Petrel is a “highly pelagic species” – that is one expected to be found in the open sea away from shore – rather than in a bay close to shore. Indeed, I note that she gives the Latin name for the Storm

Petrel – Hydrobates pelagicus – which, as to this issue, is self-explanatory. Properly characterised, her view is

readily comprehensible that “It is highly unlikely that there would be spatial overlap between this species and the proposed development site.”

1069. I should add that Dr Crowe recorded that Shot Head was, in simple distance terms, within the mean foraging range of the Storm Petrel. It is easy to infer that Dr Crowe’s rationale was related not to distance but to the record of sightings (very few and none since 1995) and to issues of direction and location. The latter is readily understood from her map of the relevant sites, an extract of which I set out below. To frequent Shot Head this highly pelagic species would have to head inland to an onshore site in virtually the opposite direction to that of the open sea.

Figure 2 Extract from Figure 3 of Dr Crowe’s Report

• The Shot Head Site, and its relatively onshore location, is marked by a small red square in the upper right quadrant of this figure.

• The Bull and the Cows Rock SPA is marked in yellow west of the tip of the Beara Peninsula.1569

1569 The site synopsis appended to the Crowe report records that the Bull and the Cows Rock SPA comprises two very small rocky islands, the Cow and

the Bull, situated 2.5 km and 4 km respectively from Dursey Head off the coast of Co. Cork and the surrounding waters to a distance of 500m.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1070. Accordingly, Dr Crowe excluded the Storm Petrel from the remainder of her study. This was, in my view, consistent with the conclusion of the Gittings report as it related to the Storm Petrel. No basis has been laid in evidence to upset that view.

1071. The subsequent AA as to birds is not impugned as to those issues it did address. For completeness, I should say that, other than as to the Storm Petrel, Dr Crowe cast the net of AA wider than had Dr Gittings. She recommended AA of not merely the Gannet (SCI of the Bull and The Cow Rocks SPA and Skelligs SPA) but also the Fulmar (SCI of the Beara Peninsula SPA, Iveragh Peninsula SPA, Deenish Island and Scariff Island SPA) and the Guillemot (SCI of the Iveragh Peninsula SPA). An AA accordingly ensued.

Birds – Decision on the Pleadings as to AA Screening and the Storm Petrel in light of the Foregoing

1072. Having regard to my understanding of the Gittings and Crowe reports as stated above I find that:

• The grounds pleaded are misconceived in that Dr Gittings, properly understood and despite some unfortunate phrasing in his report, did not screen in the Storm Petrel for AA or for AA Screening.

• Dr Gittings’ screening out the Storm Petrel is readily understood as based on his view, for which he set out his reasons, that “it is very unlikely that there is any significant spatial overlap between foraging Storm Petrels from the Bull and the Cow Rocks SPA and the proposed fish farm Site.”

• Dr Gittings’ screening out the Storm Petrel is amplified and verified by Dr Crowe: “It is highly unlikely that

there would be spatial overlap between this species and the proposed development site.” Her reasons are readily discernible and are comprehensible even to a layperson.

• The grounds pleaded are misconceived in that Dr Gittings made no findings that navigational lighting would have a significant effect on the Storm Petrel. Accordingly ALAB rejected no such findings. The plea of failure to give explicit and detailed reasons relates to a rejection of Dr Gittings’ view which did not occur.

• I may as well add that the reasons for Dr Gittings view that navigational lighting would have no significant effect on the Storm Petrel are perfectly obvious. Navigational lighting cannot affect a bird not present to be affected.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Birds – EIA – the DPM Statistical Report May 2021

1073. As to birds, that leaves only the plea that, in EIA, ALAB failed to notify the public of the DMP1570 statistical analysis of net entanglement mortality in Gannets from Bull and Cow Rocks SPA.1571

1074. The context here is of some importance. MOWI’s NIS of July 2020 and MERC’s AA report of September 2020 addressed the issues of Gannets and the risk of their entanglement. MERC’s AA report considered the Gittings report of 2018 – which had been circulated to all for comment in April 2018. MOWI’s NIS of July 2020 and MERC’s AA report were circulated to all parties and published via newspaper notices in September 2020, specifically identifying the Gannet as of concern and inviting submissions from all, including the Sweetman Applicants.

1075. MERC’s AA report of September 2020 recognised:

• in Gannets, “low inherent population growth rates and a generally poor ability to recover from factors which reduce populations, particularly if these result from additional adult mortality”1572

• Though data is lacking, Gannet predation appears to be an occasional event at salmon cages.

• So, given the plunge diving behaviour of Gannets and the use of top nets to prevent predation there is a risk of mortality by entanglement in incorrectly maintained top nets.1573

• Gannet numbers in the Bull and Cow Rock SPA and Skellig SPA are increasing – consistent with the wider trend in Ireland, the UK and Norway.

• That increase has continued throughout the period where salmon cages have been in place in Bantry Bay and the adjacent Kenmare River, suggesting that any mortality events at these sites are not currently having an adverse population level impact on the Gannet colonies at connected SPAs.

• Given the low likelihood of entanglement and in the context of an increasing Gannet population, significant effects are considered highly unlikely to occur.

• However, the Gannet population is not likely to continue to increase. Given the lack of data on Gannet predation and entanglement, this interaction requires monitoring. Should the Gannet population decline at the Bull and Cow Rock it will be important to evaluate the effect of this interaction on a declining population. A recommendation is made in this context and is an outcome of this assessment.

1076. MERC’s AA report was circulated. All were in a position, if they felt it worthwhile, to contrast Dr Gittings’ view that Gannet populations were vulnerable to even low increases in mortality with MOWI’s NIS and MERC’s AA report as they addressed risk to Gannets and to make submissions to ALAB accordingly. The Sweetman Applicants rely1574 on An Taisce’s response dated 16 November 2020 – inter alia raising concerns

1570 DMP Statistical Solutions UK Ltd. 1571 §E2.36. 1572 §5.2. 1573 §§6.3. & 7.1.1. 1574 Affidavit of Noel Carr – sworn 27 September 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

as to Gannets and the risk of entanglement of Gannets in salmon farm netting. On 5 February 2021, ALAB decided to issue a s.47 notice to MOWI seeking the most recent available data and research as to potential or observed effects on bird species at MOWI salmon farms. It issued on 15 February 2021. MOWI replied on

2 March 2021. ALAB sent that reply to MERC.

1077. Much of MERC’s Supplemental AA Briefing note of 19 May 2021 addresses An Taisce’s concerns as to Gannets. Inter alia, it observes that

• the gannet population, nationally and in the UK and in the Bull and Cow Rocks SPA, is increasing. In that SPA it is increased from 1,511 AON/S1575 in 1985 to 6,388 in 2014. It did so in the presence of existing and long-established salmon farming in Bantry Bay and in the nearby Kenmare River. So any possible fatal interactions between gannets and salmon farms in the area is not causing sufficient additional mortality to cause a population decline.

• Available literature strongly indicates that gannet mortality due to salmon farms is at worst occasional.

• For a variety of given reasons, were gannet mortality due to salmon farms more than occasional it would be known – even if not recorded by the salmon farming industry.1576 These are amongst MERC’s reasons for asserting that “the conclusion that significant mortality is unlikely to arise from the proposed Salmon Farm is supported by sufficient scientific and empirical evidence, which is considered sufficient to remove any reasonable scientific doubt concerning impacts.” This essentially repeats content of the MERC’s AA report of September 2020 – it introduces nothing new.

1078. However MERC’s Supplemental AA Briefing note of 19 May 2021 does introduce a new item – the DMP1577 statistical analysis of net entanglement mortality in Gannets.1578 It was done for MERC1579 on ALAB’s instruction1580 and addressed to MERC.

1079. DMP notes that

• Gannets are long-lived and small increases in annual mortality could cause significant population decline.

• The Bull Rock Gannet colony more than doubled in size over the period during which the existing fish farms in Bantry Bay have been operating.

1575 Apparently Occupied Nests/Sites.

1576 MERC says “High levels of interaction between seabirds and salmon cages is certain to have been noted by the statutory monitoring agencies as well

as interested stakeholders, the general public and local interests. No concerns have been raised by monitoring authorities in relation to gannet mortality

on salmon farms and no evidence of such interaction is provided in the submission by An Taisce. Wildlife stakeholders and the public generally engage

with wildlife that they can see, particularly in relation to iconic species, such as gannets. It is highly likely that injured or dead gannets would be observed

and recorded by the public, monitoring agencies, interested stakeholders as well as salmon farm operators at or close to cage farms in Ireland or other

jurisdictions that share cage farming industries and gannet populations (e.g. Norway, Scotland). No such evidence has been provided at any stage in any

submissions and desk research and specific knowledge and experience confirms that gannet mortality is a rare event on a salmon cage installation.”

1577 DMP Statistical Solutions UK Ltd.

1578 Full title: “Population Viability Analysis Of The Impacts Of Additional Mortality Due To Fish Net Entanglement In Gannets From Bull And Cow Rocks

SPA”. 1579 ALAB AA Conclusion Statement 28 May 2021 p7; MERC Supplemental AA Briefing Note 19 May 2021 p10. 1580 ALAB Minute 1 April 2021.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Gittings’ assumption that 10 additional adult Gannet deaths per year due to net entanglement could cause a potentially significant negative impact in the Bull and Cow Rocks SPA Gannet population was based on a 1% change-from-baseline threshold which Gittings described as very precautionary, and he considered further population dynamics modelling desirable to investigate which level of additional annual mortality would lead to a significant negative impact on the population.

• DMP’s assumption that all Gannet mortalities due to net entanglement would be of Gannets from the Bull and Cow Rocks SPA is precautionary and unlikely as there are other colonies nearby.

1080. The report is highly technical but it states its specifications and assumptions at some length and its summary1581 concludes, inter alia, that: 10 additional adult Gannet deaths due to net entanglement:

• corresponds in the starting year (2021) to an increase of 1% in the baseline/unimpacted annual mortality rate of Gannets in the Bull and Cow Rocks SPA.

• would by the end of the licence period (2021) likely cause an impact of low significance in the population's growth rate (0.09% reduction from unimpacted levels).

• may, by the end of the licence period (2031) and after a 10-year recovery period (2041), cause an impact of low significance in population numbers (0.99% drop from unimpacted levels).

1081. In light of the DMP Report, MERC concludes1582 that

• Based on the literature, direct knowledge of salmon husbandry at sea, as well as well-developed understanding of management of risks to wildlife on salmon farms, it is considered that Gannet mortality owing to net entanglement is highly likely to be a rare event.

• the DMP modelling shows that o at a low level of annual mortality, i.e. <10 Gannets, the effects on the Bull and Cow Rock SPA Gannet population are of low significance. o additional mortality of at least 20 Gannets per year is required before population level effects become significant.

• Mortality in excess of 20 Gannets per year (where each Gannets comes from the Bull and Cow Rock SPA) is considered to be highly unlikely. It highly unlikely that this level of mortality occurs at salmon farms without detection and without concerns being reflected in the literature.

• As such1583 significant effects on the Bull and Cow Rock SPA population as a result of this effect are unlikely.

1082. In my view, the DMP report is best understood as additional evidence as to a matter which had been live for some time between the protagonists. While its method was a novelty in the argument, the argument itself had been well canvassed. The DMP report was, indeed, at least in part, a response to An Taisce’s citing

1581 p5 et seq. 1582 pp6 & 8. 1583 Sic.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Gittings to the effect that “1.7 Gannets lost per SPA would be significant”. The DMP report introduced new evidence but raised no new issues and merely supported the view MERC had already taken. On these issues all had either had their say or had the opportunity to have it. Not merely did the Sweetman Applicants fail to take that opportunity when circulated with the NIS and MERC AA Report, they have failed to say now what it is they would have said had they been sent the DMP report.

1083. This is essentially a fair procedures argument. As is clear from cases such as Haverty,1584 Wexele,1585 and McCaffrey,1586 what is required in a process such as ALAB’s is not perfect fairness but reasonable fairness and that on the application of broad principles of commonsense and fair play having regard to all the circumstances, including the nature of the interests at stake. Reasonable fairness in all the circumstances is what is required – not perfect or the best possible justice and often it is a matter of impression as to whether or not there was unfairness. As Murphy J said of the complainants in Haverty, “I can appreciate their concern that they might have wished to expand upon their argument or to raise counter-arguments to those made in reply by the developers but I have no doubt that the real substance of their case was before An

Bord Pleanála and duly considered by it.” In Wexele, Charleton J said: “The Board is not obliged to bring every

fresh submission to the attention of a party to the appeal and to ask for further observations. The first principal applicable is that of utility. The scheme under the Act is not to be replaced with a mechanical application of the notion derived from civil law that everything before the decision maker must also be before the parties and that everything which is submitted must be known to all sides and that they must be given a reasonable opportunity to counter to with submissions of their own. In the end, someone must have the last word.

1084. And a complainant must show what, of value (s)he would have said if given that last word. In Wexele, Charleton J said:

“Fundamentally, if a complaint is made that an applicant was shut out of making a submission, that party must show that they have something to say. What they have to say must not be something that has already been said. Nor can it be a reiteration in different language of an earlier submission.

If a party is to meet the onus of alleging unfairness by the Board in cutting them out for making a submission they must reveal what has been denied then, what they have to say and then discharge the burden of showing that it had been unjust for the Board to cut them out of saying it.”

1085. I do not see that ALAB breached these principles on the facts as I have just analysed them. Even if it had, the Sweetman Applicants have failed to show that they would have said anything new or substantive in reply. Neither do I see that the Sweetman Applicants’ reliance on the public participation articles of the EIA

1584 State (Haverty) v An Bord Pleanála, [1987] I.R. 485. 1585 Wexele v An Bord Pleanála [2010] IEHC 21.

1586 McCaffrey v Minister for Agriculture [2017] IECA 246 (Court of Appeal, Hogan J, 26 July 2017) citing International Fishing Vessels Limited v Minister

for the Marine (No. 2) [1991] 2 I.R. 93. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Directive advances their case in this regard. Leaving aside any specific regulatory obligations, those articles seem to me entirely consistent with the Irish law principles of fair procedures.

Birds – Conclusion

1086. It follows from my analysis above that I dismiss all pleaded grounds as they relate to EIA, AA Screening and AA of risk to birds.

O’Toole Kelp Report 2021 – Non-Circulation

1087. It is convenient to deal here with the alleged unfairness in failure to circulate the O’Toole Kelp Report 2021 on the kelp harvesting issue. It was an AA issue as to possible risk posed by the Shot Head salmon farm to bird species cumulative with licensed mechanical kelp harvesting in the Bay.

1088. By way of background, I note from the O’Toole Kelp Report 2021 that BioAtlantis Ltd received a licence to mechanically harvest kelp at 5 locations comprising 753 Hectares in outer Bantry Bay – about 0.3% of the area of the bay1587 – for a period of 10-years from January 2014. Lengthy litigation challenging that licence ensued and by judgment of October 20221588 at least part of it1589 was resolved in favour of BioAtlantis. The O’Toole Kelp Report 2021 had referred to the kelp harvesting licence as “currently inactive” (I infer, due to the litigation). While I am unclear what became of other proceedings,1590 there was no suggestion at trial in the present proceedings that activity on foot of the kelp harvesting licence was in any degree likely, prior to its expiry in January 2024 to overlap in any environmentally relevant way with activity on foot of the aquaculture licence at issue in the present case.

1089. In truth, the point was not much pursued trial in the present proceedings. It received negligible attention in written and oral submissions but was not formally abandoned. The inactivity issue apart, I think the point was rightly not pursued.

1090. I will not consider her report in depth here but note that Dr O’Toole was reassuring and that her literature review “did not list the three SCI species of concern as species which utilise kelp as a foraging

1587 Casey v BioAtlantis Aquamarine Ltd [2022] IECA 222, §3. 1588 Casey v BioAtlantis Aquamarine Ltd [2022] IECA 222. 1589 A planning injunction application. 1590 In judicial review.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

resource”. In any event, the proposed kelp harvesting sites1591 represented a yearly and insignificant loss of

0.2 - 0.4% of available kelp foraging in Bantry Bay. Also, all three SCI species have large foraging ranges. She concluded that kelp harvesting will not have a significant impact on the foraging ability or available prey resource for those three species. And, as to effect of kelp harvesting on other foods, such as fish, consumed by the three SCI species, she considered fears of “a trophic level collapse” to be unreasonable and not borne out by the literature and previous experience in other countries. Her conclusion as to direct mortality due to and disturbance by kelp harvesting and impact thereby on breeding success and productivity was similar and her “reasoned conclusion” was that in combination/ cumulative effects of the proposed Salmon Farm and kelp harvesting on these SCI species or the conservation objectives for any designated site were highly unlikely.

1091. It is notable that:

• this report was in reaction to an issue raised by An Taisce – in other words the opportunity to ventilate the issue in objection to the proposed salmon farm had been both available and availed of by an objector other than the Applicants in these proceedings.

• By reference to the Wexele criterion, there has been no attempt to demonstrate what would have been said in reply to the O’Toole Kelp Report 2021 had it been circulated.

Accordingly, the challenge on this ground fails.

WATER FRAMEWORK DIRECTIVE

WFD – Introduction

1092. IFI does not plead the Water Framework Directive1592 (“WFD”). SWI and the Sweetman Applicants raise WFD issues relating to Dissolved Inorganic Nitrogen (“DIN”), Emamectin Benzoate (“EmBz”) and effect on the benthos.1593

1093. ALAB’s Impugned Aquaculture Licence Determination accepted the RPS Water Modelling Report 2015 results as indicating that the Shot Head Salmon Farm “will not negatively affect Outer Bantry Bay's …. classification under the Water Framework Directive”.1594 ALAB:

• cited the “High” classification of the Bay’s WFD water quality status. This is to be understood as relating

specifically to its ecological, as opposed to its chemical or overall, WFD water status.1595

1591 to be harvested on a four-year rotational basis. 1592 Directive 2000/60/EC establishing a framework for Community action in the field of water policy.

1593 The benthos, or benthic community, is the community of organisms living on, in or near the bottom of a water body.

1594 §6.6.2 It accepted the same conclusion as to Berehaven but that waterbody did not feature in argument.

1595 There is no higher Chemical Status than Good and, as overall status is the lesser of Ecological and Chemical Status, no higher overall status than Good.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• accepted that nutrient releases (DIN) “will not breach the EQS”.1596 (“EQS” stands for “Environmental Quality Standard”.)

• imposed a condition that EmBz would not breach its EQS and conditions restricting its use accordingly.1597

• found that adverse impact on the benthos would be acceptable as localised to the unremarkable and locally common benthos beneath the cages.1598

Dangerous Substances in Aquaculture Regulations 2008

1094. Though it is one only of the myriad of relevant regulatory instruments, it is useful as an overview to note that Art. 6(1) of the Dangerous Substances in Aquaculture Regulations 20081599 provides that ALAB shall not grant an aquaculture licence save subject to conditions,

“(a) (i) limiting the discharge of a dangerous substance,1600 and

(ii) establishing, in relation to the licensed activity, emission standards, set in accordance with Article 6 of the Directive.1601

(b) identifying, by means of a map or otherwise, the boundaries or limits of the place or waters in relation to which discharge of a dangerous substance may take place,

(c) relating to monitoring and inspection of discharges and emission standards,

(d) specifying the amount of feed inputs,

(e) specifying operational practices, including the stock density and fallowing of sites,

(f) relating to the use and storage of chemicals and medicines,

(g) requiring compliance with such protocols, including in relation to monitoring, auditing and any aspect of managing an aquaculture site, as may be published by the Minister, and

(h) requiring the keeping of records relating to a condition to which this Regulation relates.”

Article 6(2) obliges ALAB, to base discharge or emission standards on relevant environmental quality objectives or standards1602 published by the Minister1603 in accordance with the WFD.

1596 §§6.5.10 & 6.6.3. 1597 §§6.5.7, 6.1.14.1 & 6.1.14.3. 1598 §§6.5.1 1599 European Communities (Control of Dangerous Substances in Aquaculture) Regulations 2008 (S.I. No. 466 of 2008).

1600 By Article 2 “dangerous substance” means a substance or member of a family or group of substances listed in List I or List II of Annex I to Directive

2006/11/EC on pollution caused by certain dangerous substances discharged into the aquatic environment and includes a substance that consists of or

contains a dangerous substance; List II includes biocides.

1601 Directive 2006/11/EC on pollution caused by certain dangerous substances discharged into the aquatic environment.

1602 “EQO” or “EQS”. 1603 Minister for the Environment, Heritage and Local Government.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

WFD – Sweetman Applicants’ Pleadings & Conclusion

1095. The Sweetman Applicants plead1604 that the Aquaculture Licence and the Foreshore Licence are

invalid for ALAB’s breach of Art. 4(l)(a)(i) to (iii) WFD and Art. 4(a) of the Surface Waters Regulations as to the

Adrigole Harbour transitional surface water body – for failure to

• consider whether the project may cause a deterioration of its status or otherwise jeopardise its attainment of good status.

• take all reasonable steps to ensure that protected areas1605 achieve compliance with applicable WFD standards and objectives.

1096. However, beyond that Dr Saunders notes1606 that the WFD status of the Adrigole Harbour transitional water body is recorded by the EPA as “Unassigned”, no other particulars or facts are pleaded in support of this plea. Mr Sweetman’s affidavit says no more of Adrigole Harbour than that he is anxious to protect it and that it be accorded appropriate WFD Status.

1097. The Sweetman Applicants’ pleas as to Adrigole Harbour are entirely abstract and theoretical. While they correctly identify matters not considered by ALAB, they do not state any basis for the implied premise that those matters required consideration. They do not lay any basis in particulars or facts for any case for certiorari. Accordingly, I need not consider the application to this case of the reasoning of Hyland J in Breádan Beo1607 as to consents for developments affecting water bodies to which no WFD status is assigned. I reject their case in this regard.

1098. So the case as to the WFD is reduced to that made by SWI. Before I turn to it, I will address some issues which may assist in understanding what follows.

1604 Ground E1.14 & §E2.33. 1605 They are identified by Arts 6 & 7 & Annex IV as follows:

• Areas designated, under specific EU legislation, as requiring special protection of their surface water and groundwater or the conservation of

habitats and species directly depending on water.

• designated drinking water abstraction areas

• areas designated for the protection of economically significant aquatic species,

• designated recreational waters, including bathing waters,

• nutrient-sensitive areas, including:

• vulnerable zones designated under the Nitrates Directive.

• sensitive areas designated as under the Urban Waste Waters Treatment Directive.

• areas, including Natura 2000 sites designated under the Habitats Directive & the Birds Directive, designated for the protection of habitats or species

where the maintenance or improvement of the status of water is important in their protection. 1606 Dr Saunders Final Report 8 December 2020 §5.4.4.

1607 Sweetman v An Bord Pleanála & Breádan Beo Teo. [2021] IEHC 16, [2021] IEHC 777. See also Case C-301/22 Sweetman v An Bord Pleanála & Breádan

Beo Teo, Opinion of Rantos AG of 21 September 2023. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

WFD – Mixing Zones

1099. Stating their purpose generally and imprecisely, mixing zones, first introduced by the EQSD,1608 are areas of a water body in which the WFD regime allows localised exceedances of permissible pollutant concentrations.1609 Domestically, Art. 51 of the Surface Waters Regulations 2009 provides for mixing zones. Their premise is that such exceedances may necessarily occur close to a pollutant source but will dilute to permissible concentrations outside the mixing zone and will not affect the WFD status of the water body generally. It is not at all clear that the concept of mixing zones even applies to DIN. Art. 4 EQSD introduces them in the specific, and it seems exclusive, context of exceedances as to substances listed in Annex I, Part A EQSD and in the specific, and it seems exclusive, context of applying EQSs to those substances. Nitrogen and Nitrates are not so listed.

1100. Mixing zones had been canvassed in papers before ALAB but were not explicitly provided for in the Aquaculture Licence. Nonetheless, as SWI saw it when pleading, only a mixing zone could have rendered the expected DIN discharges of the Salmon Farm permissible. So, SWI in its pleadings inferred them as so provided – including for DIN – and made various pleas accordingly. ALAB made clear in its pleadings that it had not provided for mixing zones in the Aquaculture Licence and it was content to have the validity of the Licence assessed without reliance on any mixing zones. Accordingly, SWI’s pleas as to mixing zones don’t advance SWI’s case: absent mixing zones, SWI can simply rely on any exceedances for which mixing zones might (in its view) have been, but were not, provided. To put it another way, mixing zones could have been relevant only to ALAB’s defence of the licence but it does not rely on any. So, I can ignore the issue of mixing zones. This enables me to describe SWI’s complex pleadings a little more simply.

WFD – The DIN Limit & the “Worst Case Scenario” Approach

1101. Inorganic Nitrogen is the principal growth rate limiting nutrient for marine plants – especially phytoplankton. Salmon farms excrete both insoluble and soluble Nitrogen.1610 Insoluble Nitrogen is excreted via faeces as a component of indigestible feed and is also released from uneaten feed.1611 It is assimilated into benthic sediments. Soluble Nitrogen is excreted from the metabolism of feed.1612 In the water column it dissolves to DIN which, by such dissolution, is biologically available for use by phytoplankton. Generally, the more DIN available, the greater phytoplankton growth. Significant DIN increases are associated with

1608 Environmental Quality Standards Directive: Directive 2008/105/EC on environmental quality standards in the field of water policy. Updated by

Directive 2013/39/EU, Art 4.

1609 They were introduced to the WFD regime by Recital 19 and Article 4 EQSD. Member States may designate mixing zones adjacent to points of discharge.

Concentrations of one or more substances listed in Part A of Annex I of the EQSD may exceed the relevant Environmental Quality Standard within such

mixing zones if they do not affect the compliance of the rest of the body of surface water with those standards. Part A of Annex I lists and sets

Environmental Quality Standard for Priority Substances and certain other pollutants. 1610 See generally, RPS Report §5.2. 1611 Approximately 3% of the feed portion. 1612 Primarily as ammonia released in urine and through the gills.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

undesirable eutrophication1613 and algal blooms in waters. Accordingly, the WFD seeks to limit DIN discharges to water bodies. Much of SWI’s case as to the WFD relates to the question whether the DIN generated by the Salmon Farm would result in a breach of WFD-inspired limits1614 as to DIN.

1102. In part, the difficulties in this case as to these issues have been terminological and methodological. Terms applied by the parties and in relevant legislative and other materials to quantitative standards for DIN

have included Environmental Quality Standard (“EQS”),1615 “limit value”, “threshold”, “boundary value”,

“boundary condition”, “criteria” “Quality Element Data” and “Quality Standard”. SWI’s case has been

pleaded in terms of the “EQS” for DIN. Much was discussed at trial as to the concept of the EQS for DIN. Dr Bass was certainly understood by the chairman of the oral hearing to invoke the concept of EQS as applying to nutrients, including DIN.1616 As has been seen, ALAB determined that the Shot Head farm would not breach the EQS for DIN. But a question arises whether the concept of EQS applies to DIN at all or whether, as applied to DIN, it is a misnomer for a different descriptive term. In my earlier judgment,1617 I confessed to a lack of clarity on that issue. I am not at all confident I am much the wiser now. That is not to suggest that, for DIN, there are no WFD-inspired quantitative standards (to use a neutral descriptor – though at risk of adding to the list of descriptors). There clearly are. But the terminological issue, as applied to a complex legal architecture, has caused confusion – if only to me. I am not at all critical of any of the legal teams in this regard as the problem seems to have derived from terminological inexactitude1618 by others earlier.

1103. Lest I be accused of similar inexactitude, I should say that the terms “DIN”, “Dissolved Inorganic

Nitrogen”, “Nitrogen”, “Nitrates” and “Nutrient” are used somewhat interchangeably in what follows. 1619

1104. As to the methodological issues, it proved necessary as late as during the trial to make inquiries of the EPA1620 as to how in practical terms they assess questions of compliance with quantitative standards for DIN. I remain somewhat unclear how that practice relates to legal requirements – though that may be an admission rather than a complaint.

1613 In the Nitrates Directive, eutrophication is defined as “the enrichment of water by nitrogen compounds, the causing an accelerated growth of algae

and higher forms of plant life to produce an undesirable disturbance to the balance of organisms present in the water and to the quality of the water

concerned” and ‘pollution’ means the discharge “of nitrogen compounds from agricultural sources into the aquatic environment, the results of which are

such as to cause hazards to human health, harm to living resources and to aquatic ecosystems, damage to amenities or interference with other legitimate

uses of water.”

1614 Inspired in the sense of being requirements quantitively set not by the WFD but by Member States as required by the WFD.

1615 Environmental Quality Standard. Defined in Art 2 WFD. 1616 Oral Hearing report pp18 & 19 1617 Salmon Watch v ALAB [2023] IEHC 129, §27. 1618 I emphasise, in the literal sense of that phrase.

1619 The Nitrates Directive does not define “Nitrate”. It does define “nitrogen compound” as “any nitrogen-containing substance except for gaseous

molecular nitrogen”. Nitrates are characterised by a chemical composition including NO3. Nitrites are characterised by a chemical composition including

NO2. Ammonia also contains Nitrogen. Much of the Nitrates Directive is framed in terms of pollution and eutrophication by and controls on “nitrogen

compounds.” Phosphorous is also a nutrient but irrelevant to the Proceedings. 1620 Environmental Protection Agency. See further below.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1105. As stated, in the WFD regime, the Outer Bantry Bay coastal water body has high ecological status. While I will consider it in more detail later, it will assist to note at this point that the DIN limit set by the Surface Waters Regulations 20091621 (which transpose the WFD) for high ecological status coastal waters, such as Outer Bantry Bay, is 170 µgDIN/l.1622 The plea that this limit was invalidly set at 170 µgDIN/l – based on an assertion that it does not represent undisturbed conditions in Bantry Bay – was not pursued. Though it is repeatedly referred to as such in the papers before me, I harbour some doubt whether this limit is properly termed an EQS. But that doubt may ultimately not be all-important as all are agreed that, however the question of breach of this limit is approached, its breach is impermissible.

1106. The Surface Waters Regulations 2009 do not prescribe any basis of calculation on which the DIN Limit of 170 µgDIN/l is to be applied. It is clear however that everyone approached the question of risk of DIN limit exceedances on an asserted “worst case scenario basis” and on the basis that, as a matter of maths, the DIN produced by the fish farm must be added to the ambient DIN in the bay and the resultant total determined whether the DIN Limit of 170 µgDIN/l was breached. This implied an acceptance by all that an exceedance localised to the vicinity of the fish farm would be a breach. As Dr O’Toole put it, 170 µgDIN/l is the “ceiling”.1623 However I remain unclear as to the legal basis for such a proposition. One can readily imagine that a DIN exceedance localised at fish farm cages occupying a mere 8.82 ha may, perhaps considering also its duration and frequency, make little difference to the general nutrient status of water body of 23,000ha such as Bantry Bay. On the other hand, one may also imagine a precautionary view being taken, perhaps especially in light of in-combination effects with other nutrient sources (for example, other fish farms), that there might be low, or no, tolerance for even such localised exceedances. However, I remain unclear what the proper approach is and the legal basis for that approach. On the evidence and submissions in this trial I can only say that everyone approached the question, or purported to do so, on the “worst case scenario” basis I have described. And though RPS, Dr Saunders and Dr Bass do not invoke them, as the precautionary and preventive principles are fundamental to the high level of environmental protection and improvement which is a “base” of EU Environmental policy, for which the EU “shall work” and at which EU law “aims”,1624 it does seem to me that the “worst case scenario” approach conforms to those principles at least in general terms. On the arguments and evidence before me, I can only approach the matter in the same way – nor was it suggested that I should not. But I confess to wondering what difference different evidence and arguments might have made. I am not to be taken as critical of anyone on this account: I find the area as obscure as, no doubt, did the legal teams involved.

1621 Schedule 5, Table 9 of the European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009 as amended) –

including by the European Union Environmental Objectives (Surface Waters) (Amendment) Regulations 2012 & 2019.

1622 1mg = 1,000µg. A microgram - 1µg - is one millionth (1×10−6) of a gram. This standard is variously written in the papers as 0.17mg/l, 170µg/l, 0.17mgN/l,

170µgN/l, 0.17mgDIN/l, and 170µDIN/l. In substance and at least as relevant to this case, they all mean the same thing. I will generally use 170µDIN/l.

1623 Day 5 p52.

1624 Treaty on the European Union Article 3, Treaty on the Functioning of the European Union Articles 114(3,) and 191(2), Case C-127/02, Waddenzee,

Judgment of 7 September 2004 §44. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

WFD – SWI Pleadings

1107. SWI CG9a1625 pleads ALAB’s breach of:

• Article 4 WFD – which prohibits deterioration of surface water bodies – by allowing discharges to waters that would o breach “emission limits” for DIN and o breach “emission limits” for EmBZ set under the Dangerous Substances in Aquaculture Regulations.1626 o eliminate the benthic community beneath the fish farm,

1108. SWI also plead:1627

• Irrationality: the evidence before ALAB was incapable of supporting its conclusion that there would be no breach of the 170 µg/l EQS for DIN.

• Invalidity as based on a clear misrepresentation of fact, going to ALAB’s jurisdiction and evident on the face of the decision: MOWI used inconsistent and contradictory figures to argue that there would be no breach of the 170 µg/l EQS for DIN.

• Inadequacy of reasons: ALAB failed to explain how it concluded that there would be no breach of the 170 µg/l EQS for DIN.

1109. SWI further plead that:

• Dr Saunders, ALAB’s Technical advisor,1628 erred in considering the DIN level compliant “at” 1km from the Site.1629 He ignored the area nearer the Site – at which it would be non-compliant. o I reject this plea as it misquotes Dr Saunders. In fact he says that ▪ the worst-case scenario adds a maximum of 40 μg/l DIN within (not “at”) 1 km of the Site.

▪ adding ambient DIN yields a maximum total of 165 μg/l DIN within 1 km of the Site. o This must be read in the context of his citation of the RPS hydrographic modelling report asserting a

DIN elevation “at the Shot Head site itself” of “between 10 μg/l to 40 μg/l” which “falls 10-fold within

1km of the site”.

1625 Also Grounds §E1.9a. 1626 European Communities (Control Of Dangerous Substances In Aquaculture) Regulations 2008. 1627 Grounds §E1.16. 1628 Dr Saunders’ Final report 8 December 2020 pp79, 80. 1629 My emphasis.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o It is clear that Dr Saunders asserted compliance with the 170 µg/l limit for DIN at all points within a radius of 1m from the site – including at the Site itself. Whether that assertion was factually correct is not the point for present purposes.

• ALAB “adopted but misunderstood” Dr Saunders’ advice as being that the nutrient EQS will not be breached so that the Site will not constitute a significant additional nutrient burden to Bantry Bay. o I observe that, in fact, ALAB correctly understood Dr Saunders’ advice as I have described it above.

1110. As to Benthos, SWI plead that Dr Saunders advised of1630 and ALAB found1631 localised adverse impact by settleable solids from the fish farm on the unremarkable and locally common benthic community

– confined to the area beneath the proposed fish cages. This effect would be similar to that of other finfish farms of equal density and involved no concerns for rare or vulnerable species.

1111. As to EmBz, SWI pleads that Dr Saunders advised ALAB that:

• Rotation of use of sea lice treatments can minimise resistance issues.

• Sea lice treatments such as EMBz, administered via feed, can protect for up to 10 weeks – as opposed to waterborne treatments which act only at administration.

• The RPS Report 2015 indicates a low residual current velocity at the Shot Head Site such that the use of EmBz at standard concentrations post-year 1 stock biomass may breach the relevant EQS.1632

• Accordingly, to avoid breaching the EQS, ALAB prohibited use of EMBz after the 7th month of salmon growth.

WFD – SWI Pleadings – Errors of Law

1112. SWI’s pleas of legal error as to the WFD and associated legal instruments are very lengthy indeed. I attempt to summarise them as follows.

1113. As to the WFD, SWI pleads “failure to recognise”:

• Recital 11 WFD – requiring the application of the precautionary and preventive principles.

1630 Dr Saunders’ Final report 8 December 2020 §6.5.4, pp51-52. 1631 ALAB Determination §6.5.1, pp23-24

1632 24 hour EQS of 0.22 ng/l waterborne EmBz within 100 m of the site. The point here is that the greater the biomass of salmon the greater the feed

required and hence the greater the amount of EMBz used at standard concentrations. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Article 4 WFD – requiring that, before authorising discharge to waters, the authorising authority must satisfy itself, on a preventive and precautionary basis, that it will not cause a deterioration in water quality.

• that deterioration occurs where the status class of any one of the criteria listed in Annex V WFD is reduced.

• Annex V, Table 1.2.4 WFD stipulates as to high status coastal waters, o of general conditions that “the physico-chemical elements correspond totally or nearly totally to undisturbed conditions”.

o that “nutrient concentrations remain within the range normally associated with undisturbed conditions.”

o of benthic invertebrate fauna that ▪ all fauna associated with undisturbed conditions are present.

▪ diversity and abundance are in the range normally associated with undisturbed conditions.

o of specific synthetic pollutants, that “concentrations [are] close to zero and at least below the limits of detection of the most advanced analytical techniques in general use.”

1114. As to the Surface Waters Regulations 2009, SWI pleads:1633

• Art. 37(1) – a surface water body shall be classified as high ecological status if, o the calculated Ecological Quality Ratio (“EQR”) values for the biological quality elements1634 show no or only very minor evidence of distortion from undisturbed or reference conditions, o there are only very minor anthropogenic alterations to the status of the hydromorphological quality elements,1635 o the values for the biological and physico-chemical1636 quality elements satisfy the criteria stated Schedule 5 of the Regulations.

1633 Grounds 20/12/22 §9.12.21.

1634 i.e. Composition, abundance and biomass of phytoplankton, Composition and abundance of other aquatic flora, Composition and abundance of

benthic invertebrate fauna. (Schedule 4, Table 5, Surface Waters Regulations 2009). 1635 Hydromorphological elements supporting the biological elements. These are:

Morphological conditions: depth variation, structure and substrate of the coastal bed, structure of the intertidal zone.

Tidal regime: direction of dominant currents, wave exposure. (Schedule 4, Table 6, Surface Waters Regulations 2009) 1636 (Schedule 4, Table 7, Surface Waters Regulations 2009)

General conditions: Transparency, thermal conditions, body oxygenation conditions, salinity and nutrient conditions

Specific pollutants: Pollution by synthetic or non-synthetic substances listed in Table 10 of Schedule 5 of these Regulations, not being for the time being

identified as priority substances, which are discharged in significant quantities into the body of water. (Table 10 does not list Nitrogen or Nitrates. It sets

an “Environmental quality standard (EQS)” for each such specific pollutant in terms of either or both of annual averages and maximum allowable

concentrations calculated in accordance with certain technical specifications. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1115. In a “catch-all” manner, SWI pleads1637 that ALAB,

• Failed to consider, o if there was a point of discharge and if so, where it was. o any of the matters set out above. o whether it had evidence on foot of which it could be satisfied that those legal requirements were fulfilled.

• Had no evidence in relation to those requirements or capable of satisfying them.

• Failed to have regard to up-to-date Bantry Bay water quality data which shows that the quantity of nitrogen that “could be added”1638 is far outside the normal range for that chemical parameter in high status water.

• In all the circumstances, failed to obtain and to consider relevant information, applied the wrong legal test, misdirected itself in law, erred in law and in fact, failed to give any or adequate reasons for its

decision and acted unreasonably. Its decision is invalid, ultra vires and bad in law.

WFD – SWI case – Opposition

1116. Traverses apart, ALAB and MOWI plead that

• SWI’s particulars are inadequate.

• The sEIS did not propose a mixing zone for DIN.

• Dr Saunders’ Final Report did not find that DIN levels would breach the DIN Limit of 170ug DIN/L within 1km of the Salmon farm. He concluded that the limit would not be breached at any point within 1 km of the proposed fish farm. (I have upheld this plea above.)

• Dr Saunders’ Final Report (relying on the RPS hydrological modelling of 2015) stated:1639

“Ambient levels of nitrogen … reach a natural maximum in January of around 125 µg/l …. Worst-case scenario hydrological modelling1640 … (RPS, 2015) suggests nitrogen … levels within the site boundary will increase to 165 µg/l …. … these nutrient increases are predicted to remain below the … EQS for nitrogen (170 µg/l), …”1641

1637 Grounds 20/12/22 §9.13-9.16. 1638 I presume this means added to the waters by the proposed salmon farm. 1639 I have edited the content pleaded. 1640 Based on a Mean Allowable Biomass (MAB) of 2,800 tonnes of salmon. 1641 Saunders, Technical Advisor’s Final Report 8 December 2020, p53.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“The maximum ambient N concentrations in Bantry Bay occur between January and March reaching a maximum of 125 µg/l. The worst-case scenario adds a maximum of 40 µg/l within 1 km of the farm site, which amounts to a maximum of 165 µg/l within 1 km of the farm site. ….. It is modelled from a worst-case ….. 22 days in …. January … and is below the EQS. ….”1642

“in worst-case scenarios invoking synchronous bay production, the release and dispersal of nitrogen … will not breach EQSs. ….”1643

• The evidence was that the EQS limit for DIN will not be breached – at the Site, close to the Site or further away from the Site.

In short and simplifying, ALAB say the limit is 170 µgDIN/l and they accepted Dr Saunders’ advice that the total at the Shot Head Site would not exceed 165 µgDIN/l

WFD – Introduction

WFD – Complexity

1117. The broad thrust of the WFD’s approach to protection of water quality, via a river basin management system, is readily comprehensible. Also it is, in many respects and as expected of a framework directive, not a harmonising directive. Much detail and wide discretion are left to both many other directives1644 and to Member States – see Weser and Schwarze Sulm.1645 This position is further complicated by the later replacement of many such other directives.1646 But that does not mean the 93-page WFD1647 confines itself to general principles. Despite its being a framework directive the devil is in the detail. That the WFD is very complex is widely recognised. “From the start, it was recognised that the Directive was very complex and would clearly pose many challenges” – Giakoumis & Voulvoulis.1648 Those authors have endorsed the considerable claim, in an area of law not noted for its simplicity, that it is “the most ambitious and complex piece of legislation on environment ever enacted in the EU”.1649 I am relieved to note that Jääskinen AG in the Weser case very usefully described the scheme of the WFD and described it as “a

1642 Saunders, Technical Advisor’s Final Report 8 December 2020, pp79, 80. Similar content from p81 is also pleaded.

1643 Saunders, Technical Advisor’s Final Report 8 December 2020, p84.

1644 For example, see Article 10 WFD on the combined approach to control of discharges and emissions into surface waters. And see also Annex IV.

1645

Case C-461/13 (Weser) Bund für Umwelt und Naturschutz Deutschland eV v Germany, Judgment of 1 July 2015, §34; Case C-346/14 (Schwarze Sulm)

Commission v Austria, Judgment of 4 May 2016, §70. Also EU Water Law (era-comm.eu).

1646 E.g. the 5 Directives listed in Annex IV have been repealed since 2012 but remain listed in the consolidated version of the WFD issued in 2014.

1647 Consolidated version as amended to 31.10.2014 and as published by the EU.

1648 The Transition of EU Water Policy Towards the Water Framework Directive’s Integrated River Basin Management Paradigm Environmental

Management (2018) 62:819-831, Springer, The Authors were at the time of the Centre for Environmental Policy, Imperial College. London.

1649 Citing Prieto MM (2009) Facing the challenges of implementing the European Water Directive in Spain. In: Garrido A, Llamas MR (eds). Water policy

in Spain. Boca Raton: CRC Press, Taylor & Francis Group. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

complex and particularly elaborate measure which is unusually difficult to understand”.1650 Not least, he says of the “fundamental” article of the WFD that “the structure of Article 4 of the WFD does not allow easy analysis of its provisions”. The Court agreed – referring to “ … a complex process involving a number of extensively regulated stages,1651 for the purpose of enabling the Member States to implement the necessary measures, on the basis of the specific features and the characteristics of the bodies of water identified in their territories.” Some indication of complexity is provided by the listing of no fewer than 41 defined terms in Article 2 WFD. Many more definitions, normative and quantitative, also appear1652 – and many terms are not defined which might have been. The Commission says that the WFD is

“…… a complex piece of legislation because it includes a large number of listings and of crossreferences from one provision to the other, a high number of definitions, sometimes on basic geographic notions such as those of river or lake, as well as on a series of concepts pertaining to the “status” of water bodies, which concepts are hardly self-explanatory although essential to its underlying logics, of half a dozen “characteristics” (Article 5) and of nine paragraphs listing various and diverse “objectives”, each one of which has several sub-levels (Article 4).”1653

1118. The EPA has expressed similar views.1654 I briefly described aspects of the “complex and nuanced” WFD regime in my ruling in this case on the issue of cross-examination.1655 I gratefully adopted that phrase as used by Hyland J in a Sweetman case1656 – in which she helpfully described the WFD regime and Ireland’s transposition of it. Rantos AG in a reference to the CJEU in that Sweetman case has described the WFD as “establishing a complex process involving a number of extensively regulated stages”.1657

1119. I confess that my consideration of the WFD in these proceedings has not lead me in any degree to doubt these sentiments. Given the complexity of the WFD, I have thought a degree of repetition tolerable – even valuable – in what follows. That is particularly so as that complexity makes a rigidly linearly sequential description of the WFD unattainable.

1650 Case C-461/13 (Weser) Bund für Umwelt und Naturschutz Deutschland ECLI:EU:C:2015:433. Also, Sweetman v An Bord Pleanála [2021] IEHC 16

(Hyland J, 15 January 2021). The doubt and confusion which that complexity generated is described in Paloniitty, Journal of Environmental Law (2016) 28

(1): 151. 1651 i.e. regulated at EU level. 1652 Annex V WFD. 1653 EU Water Law (era-comm.eu).

1654 “…. the pressures and impacts analyses, and the linkages between the physical characteristics, status, impacts and the risks of not meeting the WFD

objectives (i.e. risk characterisation), are more complex and there is less clarity on how to carry out the assessments of the risk of not meeting WFD

objectives. This is indirectly acknowledged in Principle 492 of the Directive which states that the Commission ‘may adopt guidelines to promote a thorough

understanding and consistent application of the characterisation criteria’. – “An approach to characterisation as part of implementation of the Water

Framework Directive” – EPA 2016. (The reference should be to Recital 49 of the Directive). 1655 Salmon Watch Ireland v ALAB [2023] IEHC 129.

1656 Sweetman v An Bord Pleanála [2021] IEHC 16 (Hyland J, 15 January 2021). Hyland J recited the description of the WFD Scheme by Jääskinen AG in the

Weser case and adds an account of part of the process to be followed under Article 5 & Annex II WFD.

Donal Grant of the Water Policy Division of the Department of Housing, Local Government, and Heritage has also deposed to a part-description of the

WFD regime as it relates to setting EQSs and designation of mixing zones pursuant to Art 51 of the Surface Waters Regulations 2009 as amended.

1657 Case C-301/22, Sweetman v An Bord Pleanála, Opinion of Rantos AG of 21 September 2023.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1120. The complexity of the WFD regime is such that detailed analysis is difficult to avoid. In that context, it is important to ensure that a close examination of the decision as to WFD issues does not become an appeal on the merits rather than a review of legality of the Impugned Decisions.

WFD – Introductory Description

1121. An early point in applying the WFD regime is the identification of River Basin Districts1658 (which include adjacent coastal waters) as management units for WFD purposes and, within them, the identification of discrete water bodies with identified boundaries1659 and characterised by type.1660 In the present case, the coastal water body in question, to which the WFD applies, is Outer Bantry Bay.1661

1122. Each water body is assigned, by the Member State, a “surface water status” “class”. At a very general level:

• The first task in doing so is to describe and quantify the conditions which represent what is, or ought to be, the “undisturbed condition” of the water body. I have failed to find a definition of “undisturbed

condition”. Jääskinen AG in the Weser case1662 described it as the “pristine or historical status” of a water

body. As “pristine” and “historical” status are not necessarily the same, and as very few indeed must be the European waterbodies entirely and completely unaffected by human presence or activity, my broad understanding of that is that it represents the natural condition of the waterbody absent recent significant human influence. Some understanding of what undisturbed condition means can be gleaned from contrasting it with the definition of “heavily modified water body’ as a surface water body which “as a result of physical alterations by human activity is substantially changed in character”.1663

• It is for the Member States to describe and quantify the “undisturbed condition” of the water body (but in accordance with the WFD and under Commission supervision). Not least, that is because the undisturbed conditions of water bodies vary considerably. That of a bay in Donegal will naturally differ appreciably from that of a bay in Greece.

1658 The WFD defines “river basin districts” as “the area of land and sea, made up of one or more neighbouring river basins together with their associated

groundwaters and coastal waters, which is identified under Article 3(1) as the main unit for management of river basins.” Bantry Bay is in the South

Western River Basin District which comprises most of the land area of Counties Cork and Kerry and small areas of Counties Waterford, Limerick and South

Tipperary. 1659

The implementing European Communities Environmental Objectives (Surface Waters) Regulations 2009 define “body of surface water” as a discrete

and significant element of surface water such as a stretch of coastal water. They define “coastal water” as surface water on the landward side of a line,

every point of which is at a distance of one nautical mile on the seaward side from the nearest point of the baseline from which the breadth of territorial

waters is measured, extending where appropriate up to the outer limit of transitional waters. 1660 In accordance with the method set out in Annex II. 1661 Depicted in Figure 3.1 of the Supplemental EIS of April 2018. 1662 Weser Case, Jääskinen AG §§92 & 93. 1663 Art 2.9 WFD.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• The second broad task is to assign a surface water status class to the water body on the basis of any extent to which its actual condition conforms to, or varies by deterioration from, its assigned undisturbed condition. This involves assigning 3 statuses: ecological, chemical and general or overall. The WFD does not use the words “general”, “overall” or any similar descriptor but they seem to me useful as general or overall status is deemed to be the lesser of ecological and chemical status.

1123. Recital 25 WFD states that “Environmental objectives should be set to ensure that good status of surface water and groundwater is achieved throughout the Community and that deterioration in the status of waters is prevented at Community level.” Article 1 WFD identifies its purpose as being to establish a

framework for the protection of, inter alia, coastal waters,1664 inter alia from “deterioration”. Reflecting both

these objectives, Article 4 WFD1665 – described by Jääskinen AG in the Weser case as the “fundamental” article of the WFD – requires Member States to achieve two “environmental objectives”:

• To protect, enhance and restore all surface water bodies with the aim of achieving good status in accordance with Annex V, by the end of 20151666 – known as the “obligation to enhance”.

• To implement the necessary measures to prevent deterioration of the status of all surface water bodies1667

– known as the “obligation to prevent deterioration of status”.

1124. By Article 2.17 WFD, the “surface water status” of a water body (i.e. what I have called its “general”

or “overall” status) is the “general expression” of its status and, is the poorer of its ecological status and its

chemical status.1668 There is potential for confusion here as the highest possible chemical status class is

“Good”, whereas the highest possible ecological status is “High” and “Good” is one lower. So the general

surface water status class of a water body can never be better than good. This is consistent with the obligation to enhance, which aims for good status. However, the obligation to prevent deterioration requires

1664 Also inland surface waters, transitional waters, and groundwater.

1665 Art 4(1) reads in part: In making operational the programmes of measures specified in the river basin management plans:

(a) for surface waters

(i) Member States shall implement the necessary measures to prevent deterioration of the status of all bodies of surface water, subject to the application

of paragraphs 6 and 7 and without prejudice to paragraph 8;

(ii) Member States shall protect, enhance and restore all bodies of surface water, subject to the application of subparagraph (iii) for artificial and heavily

modified bodies of water, with the aim of achieving good surface water status at the latest 15 years after the date of entry into force of this Directive, in

accordance with the provisions laid down in Annex V, subject to the application of extensions determined in accordance with paragraph 4 and to the

application of paragraphs 5, 6 and 7 without prejudice to paragraph 8;

(iii) Member States shall protect and enhance all artificial and heavily modified bodies of water, with the aim of achieving good ecological potential and

good surface water chemical status at the latest 15 years from the date of entry into force of this Directive, in accordance with the provisions laid down

in Annex V, subject to the application of extensions determined in accordance with paragraph 4 and to the application of paragraphs 5, 6 and 7 without

prejudice to paragraph 8;

(iv) Member States shall implement the necessary measures in accordance with Article 16(1) and (8), with the aim of progressively reducing pollution

from priority substances and ceasing or phasing out emissions, discharges and losses of priority hazardous substances

without prejudice to the relevant international agreements referred to in Article 1 for the parties concerned; Similar obligations are set out as to groundwaters and protected areas.

1666 Subject to the application of extensions determined in accordance with paragraph 4 and to the application of paragraphs 5, 6 and 7 without prejudice

to paragraph 8.

1667 Subject to the application of paragraphs 6 and 7 and without prejudice to paragraph 8. For present purposes I ignore the obligations as they relate to

Groundwater and to Protected Areas.

1668 The WFD also stipulates a concept of ecological potential of water bodies but as it relates only to heavily modified artificial bodies of water, with

which this case is not concerned, I will omit generally references to it from what follows. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

that, if high, ecological status be maintained. As to DIN, we are concerned with ecological status and, as stated, the ecological status of the Outer Bantry Bay waters is high – the best of the 5 possible classes of status.1669 So, as to Outer Bantry Bay, the obligation to restore is redundant and we are concerned only with the obligation to prevent deterioration of status.

1125. The WFD requires Member States to make River Basin Management Plans with a view to achieving these objectives. But, to the surprise of some given the WFD is a framework directive, the CJEU has held the obligation to prevent deterioration is not merely a programmatic one to implement measures: it is binding as to the result of actually preventing deterioration and it binds in development consent procedures such as aquaculture licensing – see Weser and Schwarze Sulm.1670 Indeed, in those cases, the CJEU has said that “it is impossible to consider a project and the implementation of management plans separately” – though it must be said that the relevant management plan did not loom large in this case.1671

1126. Article 4.6 WFD stipulates that “Temporary deterioration in the status of bodies of water shall not be in breach” of the WFD – but only in limited listed circumstances not here relevant.1672 Article 4.7 WFD stipulates, inter alia, that Member States will not be in breach where failure to prevent deterioration of body of surface water from high to good status is the result of new sustainable human development activities and certain conditions are met.1673 It follows that outside those circumstances listed in Articles 4.6 and 4.7 – i.e. generally – deterioration from high to good status is in breach. No-one in this case has suggested reliance on the exceptions provided by Articles 4.6 and 4.7.

1669 High, Good, Moderate, Poor & Bad. Schedule 4, Table 4 of the Surface Waters Regulations defines High Ecological status as follows: There are no, or

only very minor, anthropogenic alterations to the values of the physico-chemical and hydromorphological quality elements for the surface water body

type from those normally associated with that type under undisturbed conditions. The values of the biological quality elements for the surface water

body reflect those normally associated with that type under undisturbed conditions, and show no, or only very minor, evidence of distortion.

Article 37 of the Surface Waters Regulations stipulates that Ecological status shall be calculated as follows — (inter alia) (1) A body of surface water for

which the calculated ecological quality ratio values for the biological quality elements show no or only very minor evidence of distortion from undisturbed

or reference conditions, and where there are only very minor anthropogenic alterations to the status of the hydromorphological quality elements and

where the values for the biological and physico-chemical quality elements satisfy the relevant criteria established in Schedule 5 of these Regulations, shall

be classified as being of high ecological status.

1670 Case C-461/13 (Weser) Bund für Umwelt und Naturschutz Deutschland eV v Bundesrepublik Deutschland §43; Case C-346/14 (Schwarze Sulm)

Commission v Austria §53-56.

1671 The River Basin Management Plan for Ireland 2018 – 2021 is exhibited. As is regrettably fairly common in the case of government documents, it

does not bear, or at least obviously bear, its publication date. However, www.gov.ie records its publication date as April 2018.

1672 “… circumstances of natural cause or force majeure which are exceptional or could not reasonably have been foreseen, in particular extreme floods

and prolonged droughts, or the result of circumstances due to accidents which could not reasonably have been foreseen, when all of the following

conditions have been met ….” 1673

(a) all practicable steps are taken to mitigate the adverse impact on the status of the body of water;

(b) the reasons for those modifications or alterations are specifically set out and explained in the river basin management plan required under Article 13

and the objectives are reviewed every six years;

(c) the reasons for those modifications or alterations are of overriding public interest and/or the benefits to the environment and to society of achieving

the objectives set out in paragraph 1 are outweighed by the benefits of the new modifications or alterations to human health, to the maintenance of

human safety or to sustainable development, and

(d) the beneficial objectives served by those modifications or alterations of the water body cannot for reasons of technical feasibility or disproportionate

cost be achieved by other means, which are a significantly better environmental option. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1127. Article 8 WFD requires Member States to establish water status monitoring programmes – I consider these further below.

1128. Article 11 WFD requires Member States to establish River Basin Management Plans for each river basin district – to include, at a minimum, “basic measures” some of which are described at Annex VI, Part A WFD. It lists, inter alia, measures required under the Nitrates Directive.1674 Notably, Article 10 WFD cites the Nitrates Directive as concerning the protection of waters against “pollution caused by nitrates”.1675 Nitrates are also identified as a “Main Pollutant” in Annex VIII WFD. So, nitrates are pollutants.

1129. The WFD defines1676

• “Emission Limit Value” (“ELV”) as the mass of an emission which may not be exceeded during any one or more periods of time. ELVs apply at the point of emission – dilution thereafter being disregarded.

• “Environmental Quality Standard” (“EQS”) as the concentration of a particular pollutant or group of pollutants in water, sediment or biota which should not be exceeded in order to protect human health and the environment.1677

Though it is obvious, it bears observing that an ELV limits quantum of emission whereas an EQS limits the resultant concentration of the emission in the receiving water, sediment or biota. It may assist to think of an ELV as addressing cause and an EQS as addressing effect.

WFD – Ecological Status, General Definition of Ecological Quality & Ecological Quality Elements

1130. As I have said, by Article 2.17 WFD, the general surface water status of a water body is determined by the poorer of its ecological status and its chemical status.1678 By Article 2.21 WFD, the ecological status of a water body is defined as an expression of the quality of the structure and functioning of its aquatic ecosystem classified in accordance with Annex V WFD. As I have also said, Member States have discretion in setting detailed ecological standards suited to their particular circumstances. But they must do so in accordance with the normative requirements of Annex V as to ecological status. In Ireland that has been done by regulation and the EPA.

1674

Directive 91/676/EEC concerning the protection of waters against pollution caused by nitrates from agricultural sources.

1675 Albeit from agricultural, not aquacultural, sources. 1676 Defined in Art 2 WFD.

1677 This definition is particular to the WFD. For example, Article 3 of the Industrial Emissions Directive 2010/75/EU defines “environmental quality

standard” differently: as “the set of requirements which must be fulfilled at a given time by a given environment or particular part thereof, as set out in

Union law”.

1678 The WFD also stipulates a concept of ecological potential of water bodies but as it relates only to heavily modified artificial bodies of water, with

which this case is not concerned, I will omit generally references to it from what follows. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1131. Annex V WFD is important. Inter alia and broadly, it:

• identifies the “Quality Elements” for classification of Ecological Status.1679 The Quality Elements differ as

between different types of water body but for all types are grouped respectively as o Biological elements (i.e. living things – biota).

o Chemical and physico-chemical elements supporting the biological elements (this should not be confused with chemical status).

o Hydromorphological elements supporting the biological elements.

o General quality elements.

• States normative definitions1680 of each Ecological Status class – High, Good and Moderate – by reference to the water body type and the state of its Quality Elements.1681

• Requires that Ecological Status be classified High, Good, Moderate, Poor or Bad – as the lower of the similarly classified1682 values for the Biological and Physico-Chemical monitoring results for the Biological and Physico-Chemical Quality Elements.1683

• Requires that the status of the Biological Quality Elements be expressed in the form of an Ecological Quality Ratio (“EQR”) calculated via an “intercalibration” process designed to achieve comparability as between Member States. Member States also decide the numerical “boundary” EQR values distinguishing the High, Good and Moderate Biological Status classes. In the intercalibration process, the Commission also verifies that those boundaries conform to the normative definitions in Annex V.1684

• States normative definitions of procedures for the setting by Member States of EQSs for pollutants listed in Annex VIII, points 1 to 9 as they bear on ecological status – for the protection of biota.1685 Notably, Nitrates are listed in point 11 of Annex VIII and so are excluded from coverage by these definitions of procedures for setting EQSs.

• States standards for design and frequency of surveillance, operational and investigative monitoring of ecological status and chemical status for surface waters and their quality elements. 1686

1132. Annex V §1.1 WFD identifies “Quality elements for the classification of ecological status”. Annex V §1.1.4 WFD identifies those quality elements for Coastal Waters:

• “Biological Elements” include “Composition and abundance of benthic invertebrate fauna”.

1679 WFD Annex V §1.1. 1680 Normative definitions are descriptive, qualitative and general rather than quantified and specific. 1681 WFD Annex V §1.2 & tables 1.2.1 to 1.2.4. 1682 i.e. High, Good, Moderate, Poor & Bad. 1683 WFD Annex V §1.4.2.

1684 WFD Annex V §1.4.1. i.e. the Commission seeks to ensure that national boundary EQR values are not too forgiving.

1685 WFD Annex V §1.2.6. 1686 WFD Annex V §1.3.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• “Hydromorphological elements supporting the biological elements” include tidal regime and morphological conditions.1687

• “Chemical and Physico-Chemical Elements supporting the biological elements” include, o as a “General Condition”, “Nutrient Conditions” such as DIN. o “Specific Pollutants” – inter alia, pollution by all priority substances1688 and pollution by other substances discharged in significant quantities into the water body.

1133. Annex II §1.3 WFD requires, inter alia, that Member States establish, specific to each surface water body type, including Coastal Waters, conditions representing the values, for High Ecological Status, of Physico-Chemical Quality Elements specified in Annex V §1.1 WFD – including DIN as a “nutrient”.1689

1134. The WFD provides general normative qualitative definitions for each ecological status class, and for each surface water category provides more detailed normative qualitative definitions for the high, good and moderate ecological status classes. Annex V §1.2 & Table 1.2 WFD accord normative qualities to the Quality Elements for the purpose of classifying ecological status. Each ecological status class, including high status, is determined on the basis of the extent of deviation from the “reference conditions” for the surface water body type in question. The “Reference Conditions” for High Ecological Status are values normally associated with the surface water body type under “Undisturbed Conditions”.1690 The greater the deviation of a water body from its Reference Conditions, the lower its ecological status and the more that status will be considered to have deteriorated.1691

1135. Headed “Normative definitions of ecological status classifications”, Annex V §1.2 & Table 1.2 WFD provide a general definition of ecological quality for, inter alia, “High Status” coastal waters, to the effect that:

“The values of the biological quality elements for the surface water body reflect those normally associated with that type under undisturbed conditions, and show no, or only very minor, evidence of distortion.”

“There are no, or only very minor, anthropogenic alterations to the values of the physico-chemical …..

quality elements for the surface water body type from those normally associated with that type under undisturbed conditions.”

1687 Morphological conditions relate to the depth variation, structure and substrate of the coastal bed, and the structure and condition of the inter-tidal

zones. 1688 Identified in Annex X as related to chemical status. 1689 Annex V §1.1 WFD and Annex V §1.1.4 WFD as to Coastal Waters. 1690 WFD Annex II §1.3(i). 1691 Weser Case (infra) Jääskinen AG §92 & 93.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Remembering that “Nutrient Conditions”, including DIN, are a physico-chemical element supporting the biological elements,1692 it will be apparent from the foregoing that “very minor” anthropogenic alterations to DIN values from those normally associated with undisturbed conditions in Coastal Waters will not result in deterioration from High Ecological Status.

1136. In addition to this general normative definition of ecological quality, Annex V Table 1.2 WFD states that, for classification purposes, the ecological status quality element values for each surface water category are given in Tables 1.2.1 to 1.2.4. Annex V, Table 1.2.4 WFD describes “Definitions for high, good and

moderate ecological status in coastal waters” for the “classification of ecological status”.1693 These are more

detailed normative/qualitative textual definitions rather than quantitative standards. As to High Status Coastal Waters, Table 1.2.4 prescribes in part as follows:

Physico-chemical quality element Quality element High Status General Conditions Nutrient concentrations remain within the range normally associated with undisturbed conditions. Specific synthetic Concentrations close to zero and at least below the limits of detection of the pollutants. most advanced analytical techniques in general use. Specific non-synthetic Concentrations remain within the range normally associated with undisturbed pollutants. conditions (background levels = bgl).

1137. The absence of quantification reflects the fact that “undisturbed conditions” and “the range normally associated with undisturbed conditions” may vary from place to place in the EU such that the ranges are for Member States to quantify. Burnett-Hall1694 observes that what constitutes good status may vary from water

body to water body – “good” in Brussels may be “poor” in Bristol. To make the WFD ecological classification

system operable in a practical sense, whilst allowing for geographical, environmental and like differences across the EU, it is for Member States to, as it were, put numbers, on these normative/qualitative textual definitions, subject to EU oversight. So, and as stated earlier, by Art. 37 and Schedule 5 Table 9 of the Surface Waters Regulations 20091695 the range of DIN representing High Ecological Status for Irish coastal waters was set at ≤ 170 µgDIN/l. To add to the complexity, Art. 39 of those Regulations requires the EPA to allow statistically limited deviations from the physico-chemical quality element values specified in Schedule 5 (i.e. including that as to DIN) “to ensure ecological relevance and so as to avoid a mismatch between the monitoring results for the biological and the general physico-chemical quality elements”. However, and thankfully, that provision for deviation did not feature in argument.

1692 See above and Annex V §1.1.4 WFD. 1693 Heading 1.1. 1694 Environmental Law 3rd edition 2012 §11-013.

1695 Schedule 5, Table 9 of the European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009) as amended –

including by the European Union Environmental Objectives (Surface Waters) (Amendment) Regulations 2012 & 2019. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1138. By Annex V §1.4.1 WFD and to enable comparison of Member States’ Biological Quality Elements standards and monitoring results, such as those for benthic invertebrate fauna, those results are expressed

as “Ecological Quality Ratios” (“EQR”) for the purposes of classification of ecological status. These ratios are

between the observed values and the reference values of the Biological Quality Elements.1696 High Ecological Status is represented by EQRs close to 1 and bad ecological status by EQRs close to 0. The Member States divide their respective EQR scales between 0 and 1 into five classes ranging from high to bad ecological status by assigning, consistently with the normative definitions in Annex V §1.2, a numerical EQR value to the boundaries between classes.1697 As stated, Part V of the Surface Waters Regulations 20091698 provides for EQR calculation and status class boundaries. The intercalibration process carried out at EU level seeks comparability as between Member States and to ensure that that the status class boundaries conform to the normative definitions in Annex V.1699 In Ireland and by the Surface Waters Regulations and as to Benthic Invertebrate Fauna in Coastal Waters, the High/Good EQR boundary is set at 0.75.1700

WFD – Chemical Status, EQSs and the EQSD

1139. Given the acronym “EQS” (Environmental Quality Standards1701) is repeatedly invoked in the Impugned Aquaculture Licence Determination and the judicial review papers and has caused confusion as related to DIN as a Quality Elements of ecological status, it is necessary to attempt to explain it. By the WFD, chemical status is determined by reference to EQSs.1702 But,

• EQSs also relate to ecological status as affected by “Specific Pollutants”.

• it seems1703 that meeting an EQS for Specific Pollutants achieves only Good Ecological Status. High Ecological Status requires the presence of no more than background levels of non-synthetic pollutants and undetectable levels of synthetic pollutants.1704 So, as to a water body with high ecological status, merely meeting an EQS for a Specific Pollutants could represent a deterioration in status impermissible by Article

4 WFD.

• in any event, DIN was excluded from the process for setting EQSs for Annex VIII pollutants for the protection of Biological Quality Elements of Ecological Status.1705

1696 It may perhaps have aided comprehension had the ratios been termed “Biological Quality Ratios” as they relate only to the Biological Quality Elements

contributing to Ecological Status. 1697

The values for the boundaries between the classes of high, good and moderate are established via an EU intercalibration exercise.

1698 European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009) as amended, including by the European

Union Environmental Objectives (Surface Waters) (Amendment) Regulations 2019 (S.I. No. 77 of 2019). 1699 WFD Annex V §1.4.1. I.e. the Commission ensures that national boundary” EQR values are not too forgiving.

1700 Schedule 5, Table 8 as substituted by the European Union Environmental Objectives (Surface Waters) (Amendment) Regulations 2019.

1701 Art 2.35 WFD states ‘Environmental quality standard’ means the concentration of a particular pollutant or group of pollutants in water, sediment or

biota which should not be exceeded in order to protect human health and the environment.”

1702 Article 1.24 states that ‘Good surface water chemical status’ means the chemical status required to meet the environmental objectives for surface

waters established in Article 4(1)(a), that is the chemical status achieved by a body of surface water in which concentrations of pollutants do not exceed

the environmental quality standards established in Annex IX and under Article 16(7), and under other relevant Community legislation setting

environmental quality standards at Community level. 1703 WFD Annex V §1.2 and tables therein. 1704 E.g. as to Coastal Waters WFD Annex V table 1.2.4. 1705 WFD Annex V §1.2.6.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1140. As to chemical pollution, Article 16 & Annex X WFD1706 provide for and list “priority substances” (including “priority hazardous substances”) for which the EU Commission must propose, and the EU Council must set, EQSs. This they did by the EQSD1707 – which addresses only chemical pollution. It requires that Member States, to achieve good chemical status in accordance with Article 4 WFD, meet EQSs1708 for “priority substances” and for 8 other pollutants listed in Annex I Part A EQSD.

Nitrogen is not listed either in Annex X WFD or in the EQSD.

1141. The question arises whether, for substances other than those for which the EQSD sets EQSs, the Commission and/or Member States are to set EQSs. The only source of such a requirement or power which I have found in the WFD is Annex V §1.2.6 WFD1709 – but it does not encompass DIN.

WFD – Monitoring

1142. Monitoring is the principal tool for determining the status of water bodies1710 and must suffice to

permit status classification consistent with the “normative definitions of ecological status classifications” set

out in Annex V §1.2 WFD. Article 8 and Annex V §1.3 WFD require Member States to establish monitoring programmes – surveillance, operational and investigative1711 – designed inter alia to provide a coherent and comprehensive overview of ecological and chemical status of surface water bodies. To supplement WFD requirements, Member States are to set, subject to Commission scrutiny, technical specifications and standardised methods for analysis and monitoring of water status. Normative definitions of general criteria are set for selection of monitoring points. Sufficiency to provide a coherent and comprehensive overview and to permit status classification implies that the monitoring points must be adequately representative of the water body and of any risks to its status. Frequency of monitoring is stipulated1712 – as is compliance with listed technical standards (though as to nutrients1713 the entry merely stipulates “Any relevant CEN/ISO

1706 Art 2.30 WFD states, ‘Priority substances’ means substances identified in accordance with Article 16(2) and listed in Annex X. Among these substances

there are ‘priority hazardous substances’ which means substances identified in accordance with Article 16(3) and (6) for which measures have to be taken

in accordance with Article 16(1) and (8).

1707 Directive 2008/105/EC on environmental quality standards in the field of water policy. Updated by Directive 2013/39/EU.

1708 EQSs are set as both annual averages and maximum allowable concentrations.

1709 Annex V §1.2.6 - Procedure for the setting of chemical quality standards by Member States. “In deriving environmental quality standards for pollutants

listed in points 1 to 9 of Annex VIII ……………… Member States shall …. ” This text excludes Annex VIII point 11, which relates to “Substances which

contribute to eutrophication (in particular, nitrates and phosphates).” So Annex VIII point 11 includes DIN and DIN is excluded from Annex V §1.2.6.

1710 Jääskinen AG in the Weser case. 1711 If required.

1712 Annex V §1.3.4 sets minima as to surveillance monitoring: Guidelines as to operational monitoring. Frequencies shall be chosen so as to achieve an

acceptable level of confidence and precision. Monitoring frequencies shall be selected which take account of the variability in parameters resulting from

natural and anthropogenic conditions. The times at which monitoring is undertaken shall be selected so as to minimise the impact of seasonal variation

on the results, and thus ensure that the results reflect changes in the water body as a result of changes due to anthropogenic pressure. Additional

monitoring during different seasons of the same year shall be carried out, where necessary, to achieve this objective.

1713 Sub nom Standards for physico-chemical parameters.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

standards”).1714 As to EQSs set by the EQSD, Articles 8, 8b and Annex 1 Part B EQSD reflect the requirement that monitoring points be representative of the water body.

WFD – DIN as a Nutrient & Pollutant, “EQS” for DIN and its application in Development Consent Applications

1143. As stated, there was ultimately no challenge in the Proceedings to the quantitative standard for DIN set by the Irish Surface Waters Regulations1715 at <170 µgDIN/l. Indeed, that standard was a premise of all analysis by all involved in the Licence applications and was a premise of all involved in the Proceedings. It was pleaded and repeatedly referred to in argument as an EQS for DIN. As I have said, I am not at all sure that EQS is a correct descriptor for that 170 µgDIN/l standard.

1144. Generally, EQSs apply to chemical pollutants of water. The prescribed relationship established by the WFD between ecological and chemical phenomena is appreciably complex. DIN seems to me to be an element of that complexity. It is both a chemical substance and a nutrient. Per se, that is unremarkable as all nutrients are chemicals. In my earlier judgment in the SWI Proceedings1716 I noted that DIN is a nutrient for the purposes of the WFD, the presence of which informs the ecological status of waters and also is a “Main Pollutant” of waters listed in Annex VIII WFD. I also expressed a lack of clarity as to methodology of application of what has been described as the EQS for DIN set by the Irish Surface Waters Regulations 2009.1717

1145. As I have said, nitrates are identified as a “Main Pollutant” in Annex VIII, Point 11 WFD. And Art. 10 WFD cites the Nitrates Directive as concerning the protection of waters against “pollution caused by nitrates”.1718 Art. 2.33 WFD defines “pollution” as the “… introduction as a result of human activity, of

substances .. into the .. water .. which may be harmful ….”.1719 DIN falls within that definition of “pollutant” in

Article 2.31 WFD as “any substance liable to cause pollution, in particular those listed in Annex VIII”. Nitrates

are pollutants as they cause eutrophication of waters and so proliferation of algae damaging to the environment. Annex II §1.4 WFD requires estimation and identification of significant diffuse source pollution from urban, industrial, agricultural and other installations and activities; based, inter alia, on information gathered under, inter alia, the Nitrates Directive. As information gathered under the Nitrates Directive relates to nitrates, the unsurprising implication of Annex II §1.4 WFD is that nitrates cause pollution. Art. 6

1714 Annex V §1.3.6. 1715

European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009) as amended, including by the European

Union Environmental Objectives (Surface Waters) (Amendment) Regulations 2019 (S.I. No. 77 of 2019). See below.

1716 Salmon Watch v ALAB [2023] IEHC 129 as to whether there should be cross-examination of witnesses. See §§22 & 27.

1717 European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No 272 of 2009) as amended, including by the European

Union Environmental Objectives (Surface Waters) (Amendment) Regulations 2019 (S.I. No. 77 of 2019). See below. 1718 Albeit from agricultural, not aquacultural, sources.

1719 The definition continues: “to human health or the quality of aquatic ecosystems or terrestrial ecosystems directly depending on aquatic ecosystems,

which result in damage to material property, or which impair or interfere with amenities and other legitimate uses of the environment”

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

and Annex IV WFD require that the areas to be registered as protected areas include nutrient-sensitive areas, including areas designated as vulnerable zones under Nitrates Directive. Art. 11 WFD and Annex VI, Part A WFD require Member States to include measures required under the Nitrates Directive in their WFD River Basin Management Plans.

1146. In that context of co-ordination between the Nitrates Directive and the WFD it seems reasonable to note for present purposes that, in the Nitrates Directive,1720 eutrophication is defined as “the enrichment of water by nitrogen compounds, the causing an accelerated growth of algae and higher forms of plant life to produce an undesirable disturbance to the balance of organisms present1721 in the water and to the quality of the water concerned” and ‘pollution’ means the discharge “of nitrogen compounds from agricultural sources into the aquatic environment, the results of which are such as to cause hazards to human health, harm to living resources and to aquatic ecosystems, damage to amenities or interference with other legitimate uses of water.”

1147. I hope to have demonstrated by the foregoing, at no doubt excessive length, that Nitrogen is both a nutrient relevant to ecological status and a chemical pollutant within the meaning of the WFD.

1148. Art. 2.35 WFD defines EQSs as including “the concentration of a … pollutant .. in water, … which should not be exceeded in order to protect human health and the environment.” This definition suggests that it would be unsurprising if the WFD required or permitted that an EQS be set for Nitrogen as a pollutant. All this seems to have contributed to the lack of clarity whether the <170 µgDIN/l standard is properly to be termed an EQS.

1149. Art. 37 of the Surface Waters Regulations 2009 describes Schedule 5 as setting “criteria”. As one would expect, Schedule 5, Table 8 sets Biological Quality Element boundary conditions in terms explicitly of EQRs. But Schedule 5 Table 9 ascribes no descriptor to the standards it sets for “Physico-chemical conditions supporting the biological elements” – including the <170 µgDIN/l standard. More specifically, it does not describe them as EQSs.

1720 The same definition appears in the Urban Waste Water Treatment Directive 91/271/EEC.

1721 Case C-280/02 Commission v France – considered the meanings of “Eutrophication” and “undesirable disturbance to the balance of organisms”.

Eutrophication is characterised by the confluence of four criteria: the enrichment of water by nutrients, especially compounds of nitrogen and/or

phosphorus; the accelerated growth of algae and higher forms of plant life; an undesirable disturbance of the balance of organisms present in the water;

deterioration of the quality of the water concerned. It is also characterised by cause and effect relationship between enrichment by nutrients and the

accelerated growth of algae and higher forms of plant life on the one hand and, on the other hand, between the accelerated growth and an undesirable

disturbance of the balance of organisms present in the water and to the quality of the water concerned. The equilibrium of an aquatic ecosystem is the

result of complex interactions among the different species present and with the environment. Any proliferation of a particular species of algae or other

plant therefore constitutes, as such, a disturbance of the balance of the aquatic ecosystem and, accordingly, of the balance of the organisms present in

the water, even when other species remain stable. As to whether a disturbance of the balance of the aquatic ecosystem is “undesirable” species changes

involving loss of ecosystem biodiversity, nuisances due to the proliferation of opportunistic macroalgae and severe outbreaks of toxic or harmful

phytoplankton constitute an undesirable disturbance. Harmful effects on ecosystems, but also deterioration of the colour, appearance, taste or odour of

the water or any other change which prevents or limits water uses such as tourism, fishing, fish farming, clamming and shellfish farming, abstraction of

drinking water or cooling of industrial installations, constitute deterioration of the quality of the water. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1150. In the present case, the matter was further confused by MOWI’s deployment of Scottish rather than available Irish standards. The Shot Head Salmon Farm EIS in 2011 stated that “the data must be examined to make sure that the resultant elevation of ambient Dissolved Inorganic Nitrogen (DIN) does not breach the established Environmental Quality Standard EQS for DIN.”1722 It later refers to an “EQS” set by the Scottish

EPA (“SEPA”) for the winter value for “nitrate nitrogen” of 168μgNO3N/l – which EQS, the EIS said, would not

be breached by the proposed fish farm.1723 This reliance on a SEPA EQS in a 2011 EIS is surprising as Ireland had by 2011 already set, by the Surface Waters Regulations 2009, the limit of 170 μgDIN/L.1724 But the 2011 EIS does not, as far as I can see, refer at all to the Surface Waters Regulations 2009. The RPS Water Quality Modelling Report 20151725 adopted the “EQO/EQS1726 …. approach” which it attributed to SEPA. It also cited SEPA's EQS for DIN at 168μgN/l. It did cite the Surface Waters Regulations 2009 as having set a “DIN standard” for high status coastal waters of 170 µgN/l which, it said, agreed with SEPA’s EQS. But, puzzlingly given its acknowledgement of the Irish standard, the narrative of that section of the RPS Report thereafter is in terms of the applicable SEPA “EQS” – which it explicitly adopts as its analytical standard.1727 The RPS Report later1728 refers to the Surface Waters Regulations 2009 “quality standard” but the acronym “EQS” predominates.

1151. Dr Saunders’ interim report1729 describes the Surface Waters Regulations 2009 as having set an “EQS” for nutrients and cites the “quality standard” of 170 µgDIN/l. The September 2017 Bass/RPS presentation to the oral hearing1730 likewise uses “the EQS for winter DIN concentration of 168 µgN/l” – why the Scottish, as opposed to the Irish, standard is invoked is not explained but again the concept is of an “EQS” for DIN.

1152. MOWI’s Supplemental EIS1731 (“sEIS 2018”) refers to “The Quality Element standard” for High Ecological Status waters as a winter DIN concentration of 0.17mgDIN/l (i.e. 170 µgDIN/l), at a median salinity of 34.5% – this is clearly a reference to the Surface Waters Regulations 2009, Schedule 5, Table 9. The sEIS 2018 later1732 commences its consideration of WFD requirements with the puzzling heading “3.1.

Introduction; EQS or WFD?” as if they were alternatives and the WFD had not, which it had, explicitly adopted the concept of EQS. The text also reads as though the EQSD was a departure from the WFD, rather than being, which it is, the effecting of elements of the WFD. Just as puzzlingly, the sEIS 20181733 opts to cite

1722 EIS 2011 Volume 1 p97. 1723 EIS 2011 Volume 1 p204.

1724 Surface Water Regulations 2009 Schedule 5, Table 9. Physico-chemical conditions supporting the biological elements.

1725 5 Water Quality Parameters; 5.1 Overview; §5.1.1 Introduction. 1726 Environmental Quality Objectives / Environmental Quality Standards. 1727

§5.1.1 states “In respect of the water quality parameters assessed in the following sections, (§§5.2-5.6), the EQS’s used are taken directly from SEPA’s

Manual on Marine Cage Fish Farming, where further information on each can be found.” 1728 §5.2.1 Simulation Group 1 – Maximum & Average Nitrogen Concentration. 1729 31 December 2016 – p70.

1730 “Numerical modelling of the dispersion of wastes, medication and salmon lice ….. from the proposed MHI Shot Head site in Bantry Bay.”

1731 April 2018 – p7. 1732 §3. 1733 §3.1.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

the OECD 2003 definition of EQS as determinant of “maximum allowable degradation of environmental media” and for good measure adds the authors’ own gloss to that definition by adding the phrase “for the maintenance of environmental stability” – this rather than adopting the definition of EQS explicit in the WFD.1734 The authors then cite the Surface Waters Regulations 2009 before asserting that “EQS is as valid now as it ever was” (seemingly posing EQS in contradistinction from the WFD) “and indeed many nutrient and physicochemical Quality Elements required for the derivation Ecological Status depend on EQS values or something very close to them”. While the words “or something very close to them” are opaque and puzzling in such a closely regulated context, the general thrust is to apply EQSs to Nutrients, including DIN.

1153. However, the sEIS 2018 then asserts its intended “conversion of earlier findings, expressed relative to EQS limits, and their re-evaluation, relative to the current Ecological Status of Bantry Bay, under the terms of the WFD and SI 272 2009”1735 – though it does not explain why the WFD and S.I. No. 272 of 2009 had not been applied in the first place in an EIS published in 2011. The sEIS 2018 confirms1736 that the EIS and RPS report, as to DIN, had compared their results “with established Environmental Quality Standards (EQS) for

DIN … to establish where the resulting elevated ambient concentration lies relative to the EQS level.” As best I

can discern this seems to have been a reference to the SEPA EQS. It records that the EPA, as to DIN, check

“median winter values” “against the Quality Element data for DIN, set out in Schedule 5, Table 9, Part A of SI

272”.1737 This seems a misnomer as Schedule 5, Table 9, Part A contains not “data” but, to use a neutral word, standards. However, it is notable that what is in Schedule 5, Table 9, Part A is not here described as an EQS.

1154. Another notable oddity is the repeated reliance in the sEIS 2018, for its understanding of EPA practice, on personal communication with the EPA Scientific Officer, rather than on EPA documents in the public domain. Nor, in the proceedings, were any such EPA documents exhibited. The reason for this reliance is not stated but it includes an assertion of an informal EPA practice of avoiding “sampling in what might be regarded as a “reasonable mixing zone” in their monitoring exercises, in their assessments of Ecological

Status”1738 – i.e. as to DIN. The legal basis for such a practice is not identified. I will return to the sEIS in due

course.

1155. To recap,

i. Annex VIII, Point 11 WFD lists as a “Main Pollutant” “Substances which contribute to eutrophication (in particular, nitrates and phosphates).” But those substances are not listed as “priority substances”

in Annex X.

1734 Article 2.35 WFD – though it is not clear that anything turned on the choice. 1735 The Surface Water Regulations 2009. 1736 §3.3.3. 1737 The Surface Water Regulations 2009. 1738 sEIS 2018. §3.3.4.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

ii. Nitrates, as nutrients, are Pollutants. They are also Quality Elements for classifying the ecological status of water bodies including coastal water bodies – WFD Annex V §1.1.4 and §1.2.4.

iii. EQSs are tools used for assessing both the chemical status and the ecological status of water bodies

– the latter as affected by Annex VIII pollutants – but not including nutrients.

iv. Annex V §1.2.6 WFD sets out procedures for the setting by Member States, for use in determining ecological status of water bodies,1739 of EQSs for pollutants listed in Points 1 to 9 of Annex VIII1740 for the protection of aquatic biota.1741 Notably, this excludes Point 11 – nutrients.

v. Whatever else may be said of the legal source of a power or requirement to set an EQS for DIN, for WFD purposes, it does not derive from Annex V §1.2.6 WFD.

1156. This omission from the scope of Annex V §1.2.6 seems significant given,

• the very purpose of Annex V §1.2.6 is to identify chemical pollutants relevant to ecological status of water bodies,1742

• the general reference point of Annex V §1.2.6 is the list of main pollutants set out in Annex VIII – that is how it identifies the pollutants with which it is concerned. And it refers to only 9 of the 12 categories of pollutant listed in Annex VIII – excluding nutrients.

1157. Accordingly, if it was intended to prescribe EQSs for nutrients, including nitrates, relevant to ecological status of water bodies, it is very arguably obvious that it is in Annex V §1.2.6 that one would expect to find that prescription. Yet §1.2.6 explicitly excludes nutrients, including nitrates, from its scope. That was clearly not accidental. It is very arguably to be inferred from that omission that the WFD intended that EQSs would not be set for nitrates. While of itself that would not prohibit Ireland from doing so, nor does it invalidate the limit of <170 µgDIN/l, it does make it difficult to locate EQSs for nutrients, including nitrates, as they might bear on the concept of deterioration of waters, within the WFD regime.

1739 That use is not readily apparent in the title to §1.2.6, which reads, “Procedure for the setting of chemical quality standards by Member States”. One

might readily imagine that §1.2.6 related to chemical status of water bodies. However, §1.2.6 is a sub-paragraph of §1.2. which is entitled “Normative

definitions of ecological status classifications”. 1740

Article 2.31 WFD states that ‘Pollutant’ means any substance liable to cause pollution, in particular those listed in Annex VIII. Annex VIII is an “Indicative

List Of The Main Pollutants”. Items 1 to 9 are: 1. Organohalogen compounds and substances which may form such compounds in the aquatic

environment;, 2. Organophosphorous compounds; 3. Organotin compounds; 4. Substances and preparations, or the breakdown products of such, which

have been proved to possess carcinogenic or mutagenic properties or properties which may affect steroidogenic, thyroid, reproduction or other

endocrinerelated functions in or via the aquatic environment; 5. Persistent hydrocarbons and persistent and bioaccumulable organic toxic substances;

6. Cyanides; 7. Metals and their compounds; 8. Arsenic and its compounds; 9. Biocides and plant protection products.

1741 Biota comprises animal and plant life.

1742 §1.2.6 is a sub-paragraph of §1.2 which is headed “Normative definitions of ecological status classifications”.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1158. Interestingly, while the Nitrates Directive sets 50 mg/l of nitrates as a benchmark for listing vulnerable surface freshwaters and groundwaters, it sets no numerical benchmark for coastal waters. Rather it sets a qualitative criterion whether they are eutrophic or in the near future may become eutrophic.1743

1159. As the foregoing demonstrates I have failed, I hope not through error on my part in considering the highly complex architecture of the WFP, to identify a warrant therein for setting an EQS, properly so-called, for DIN as informing the ecological status of surface waters. The <170 µgDIN/l standard certainly applies, but I am dubious that it is properly to be called an EQS – at least as that term is used in the WFD.

What Constitutes Breach of the <170 µgDIN/l Standard?

1160. Importantly, there was dispute as to what constituted a breach of the <170 µgDIN/l standard. Issues of localised breach, as opposed to breach in the water body generally and issues of transient or temporary breach were disputed. In that context, it is at very least desirable to identify in the WFD, if possible, the legal basis and significance of that <170 µgDIN/l standard. It is no disrespect to the parties to suggest that this proved, for all, a highly elusive task. In part, that may have been because the transposing Surface Waters Regulations set the <170 µgDIN/l standard and that is the law to be applied. But it did produce the difficulty just described and discussed at trial but not much progressed in argument, of locating that Irish <170 µgDIN/l standard in the system of the WFD and in the application, by reference to that standard, of the requirement that the prospect of its breach may require refusal of development consent for a project.1744

1161. My earlier judgment in this matter stated1745

“However, Table 9 itself sets no standards other than “≤ 0.17 mg N/l” for assessing whether the EQS is breached or met. From that position, a number of questions may arise. Is that standard to be applied as an average and if so over a specified time or over a specified number of samples or as a rolling average over a specified time or number of samples? Or will any one sample > 170 µgN/l constitute a breach? Is there a tolerance for one or a specified number of samples > 170 µgN/l in a greater number of samples? Over what unit of geographical area is the sampling to be applied?

(MOWI and ALAB seem to say the entire of Bantry Bay: Salmon Watch disagrees). What is the number of samples to be taken in a specified geographical area? Is the result to be measured against the EQS an average for the area. Under what environmental conditions are samples to be taken – for example, flood tide, ebb tide? Indeed these are the questions of a layman. They may be misconceived, inapplicable and/or irrelevant. Perhaps the answers are found elsewhere. However it does seem likely that, in order to lawfully apply that EQS, those or some such methodological

1743 Annex I A. 1744 See below as to the “Weser” case. 1745 Salmon Watch v ALAB [2023] IEHC 129, §27.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

questions will require to be answered. ……… However, I have not been directed to any source of answers to any such questions.”

1162. I confess that, as to the legal position, I do not consider myself any further on in that regard. However, as I have said, the parties all proceeded on the basis that an exceedance of that <170 µgDIN/l standard local to the salmon farm and considered on a “worst case scenario” basis was the relevant criterion such that ambient DIN should be added to farm-produced DIN and the result measured against the <170 µgDIN/l standard.

WFD – Cross-Examination of Dr Bass and Dr O’Toole - #1

1163. Cross-examination of deponents shed some, but poor, light on these issues. Dr Bass for MOWI and Dr Ciar O’Toole, for ALAB, were both deployed in defence of the impugned ALAB decision. I think I can fairly consider their evidence in this respect together.

1164. As I have observed, merely stipulating, as the Surface Waters Regulations 2009 do, “≤ 0.17 mg N/l”, does not state a standard capable of practical application. It also seems that, whereas other States’ transposing legal instruments do stipulate the methodological matters – specifically the use of medians of monitoring results to determine water quality status class canvassed above, Ireland’s Surface Waters Regulations do not. Nor could anyone – witness or counsel – direct me to any legal instrument which does. Indeed, Dr Bass said that “the strange thing is” that, though it appears in the Scottish, Northern Irish and

English legislation, “It is almost as if it's been omitted from the Irish legislation”.1746 I conveyed to Dr Bass my

lay impression that the relevant Irish statutory instruments state no method “to tell you how you can judge

success or failure” to meet the <170 µgDIN/l standard. He replied: “No, I agree entirely with you, Judge”1747

And again I observed: “It just doesn't seem to me that there is any statutory provision or legal provision which tells you how to measure the figure?” Dr Bass again agreed.

1165. In this context, Dr Bass thought it “worth pointing out” that when this “EQS” (by which he meant the “≤ 0.17 mg N/l” standard) was written (i.e. by the 2009 Regulations) “the EPA weren't up and running and doing all this analysis. … The only source of this type of data at that time, believe it or not, … were aquaculture businesses who collected it, obviously, for their own needs”. Indeed, I observe that Table 3.3 of

Dr Bass’s 2018 sEIS lists “all water control site samples taken in Outer Bantry Bay since the introduction of SI

272 2009, Includes Median values, used for Quality element determination, if and where required”. 83

1746 Day 5 p125. 1747 Day 5 p126.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

samples are listed. 2 were sampled by the EPA. The remainder were sampled by MOWI (12 samples at Roancarraig) and “Hensey Glan Uisce Teo”1748 at Gearhies.

1166. When asked why medians, as opposed to maximum values would be used in assessing compliance with the <170 µgDIN/l standard, Dr Bass was unable to offer an opinion of his expertise: he said he would

“rather refer to the EPA who are the statutory body … all their data is reported as median data”.1749 I found it

surprising that these experts could not direct me to the relevant instrument as method of application can greatly affect the rigour or otherwise of any such regime and the water quality status determinations deriving from its application and I had explicitly flagged the question in my earlier judgment. For similar reasons, I was surprised that Dr Bass could not explain why, as he had put it “everything from the EPA comes as median” – presumably a question which would have both a regulatory and scientific answer.

1167. I asked Dr O’Toole about this.1750 I suggested that the <170 µgDIN/l standard was meaningless unless “you know what the standard says about whether it is breached, if it occurs once in ten years, or whether it has to occur most of the time in a month, where will I find that statement of the standard?” Dr O’Toole’s reply was, to my mind, surprising:

“So, that statement of the standard would be an EPA document which isn't publicly available.”

I again asked Dr O’Toole:1751 “Where is the standard that says it's all right as long as it doesn't commonly occur? Where is the standard that says it's all right if it only occasionally exceeds 170?”

Dr O’Toole replied: “There isn't one, Judge …”

If that is indeed the situation – indeed the situation as understood by ALAB’s own technical advisor – it is difficult to see how aquaculture licence applicants, other stakeholders and objectors and ALAB can reliably address issues of compliance or non-compliance with the <170 µgDIN/l standard.

1168. I asked, perhaps unfairly repeating my question, where I would find the instrument establishing the EPA’s method. Dr O’Toole replied that she didn’t think there was one – she was reliant, for her understanding of the EPA’s method of determining compliance with the <170 µgDIN/l standard, on an oral enquiry of an EPA chemist who said that the EPA can determine the protocol. While that is so, it is not an answer to the question whether that protocol is published so all concerned, licence applicants, ALAB and objectors, can understand what, in practical terms, compliance with that <170 µgDIN/l standard actually means and requires of them.

1748 Referred to also in the 2011 EIS. 1749 Day 5 p125. 1750 Day 5 p75 et seq. 1751 Day 5 p91.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1169. Whether or not Dr O’Toole’s answers were correct in point of fact that the relevant criteria were unavailable – and, importantly, Dr Bass was no more able to assist – the net position is that the relevant experts who gave evidence in these proceedings and who had, in the case of Dr Bass, prepared the EIS and sEIS and, in the case of Dr O’Toole, acted as technical advisor to ALAB, were unaware of the criteria applicable in law for the application of the <170 µgDIN/l standard in individual development consent applications against a legal backdrop in which Weser makes clear that breach of such standards may require refusal of development consent.

1170. I should say in fairness to Dr O’Toole that she was genuinely doing her best to assist the court and also that she generally understood the general WFD ecological status monitoring approach of the EPA as based on medians of monitoring results – though that understanding was based on her consultation with EPA staff members as opposed to consideration of their published documents.

1171. I was told that water body status for WFD purposes is determined by the EPA.1752 It is determined with respect to an entire water body1753 as opposed to merely a part of it. That seems consistent with the general method of the WFD – though that may or may not imply a similar approach to development consent applications – see below as to Nordrhein-Westfalen1754 and Association France Nature.1755 As far as the witnesses described the EPA’s method, it is to identify “representative monitoring points”1756 in each water body and periodically take samples from each. As to each set of samples,1757 the evidence was that a mean of DIN values is calculated. In turn, whether or not the <170 µgDIN/l standard is met is determined by identifying the median of those means1758 – assessed over what time period was not stated. I do not suggest, much less find, that there is no legal basis for this method, but none was drawn to my attention by the expert witnesses.

1172. In these circumstances, I asked that the EPA be requested to shed light on its method. I consider its response below. But it will assist first to consider some caselaw as to the WFD and the relevant Irish Regulations.

1752 Environmental Protection Agency. 1753 Day 5 p98. 1754 Case C-535/18 IL et al v Land Nordrhein-Westfalen, Judgment of 28 May 2020, §111-118. 1755 Case C-525/20 Association France Nature v Premiere Minsitre, Opinion of Rantos AG of 13 January 2022, §59-61.

1756 European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009) – defined in Article 3.

1757 Which I take to mean a set of samples taken at or about the same time. 1758 Day 5 p98.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

WFD Caselaw – Weser 2016, Protect Natur 2017, Nordrhein-Westfalen 2020, Association France Nature 2022, Sweetman 2021 & Reasons.

Weser 2015 – Deterioration

1173. The Weser case1759 concerned a planning permission to dredge and deepen the navigable channels of the river Weser from the North Sea to allow larger ships access the port of Bremen. The river was divided for WFD purposes into various water bodies. In addition to the direct effects of dredging and of discharging dredged material, the project1760 would have other hydrological and morphological effects on sections of river. Tidal current speeds, tidal range and salinity would increase in the lower Weser, the brackish water limit would move upstream and silting-up of the river bed would increase outside the navigable channel. As to the WFD obligation to prevent deterioration of the status of water bodies,1761 the planning authority concluded that certain Weser water bodies would tend to be adversely modified by the effects of the project

– but without changing their Annex V WFD ecological status classes. It considered that deterioration within a status class was not deterioration of the ecological potential or the ecological status of a water body and so did not require refusal of development consent on WFD grounds. An NGO1762 challenged that decision and the German court referred questions to the CJEU.

1174. The CJEU held that the WFD imposed obligations on planning authorities asked to approve individual projects. In particular, prevention of deterioration of the status of water bodies is binding and applies to every surface water body type and status. Any deterioration of water body status must be prevented. (As Jääskinen AG had observed “Member States are required not only to prohibit any deterioration, but also to implement that prohibition effectively”.) Member States must refuse authorisation for projects which “may cause deterioration” of water body status.1763

1175. That view required the CJEU to consider the meaning of ‘deterioration of the status’ of a surface water body – a concept not defined in the WFD. It has been said1764 that the Court had two options:

• “the status class theory”/“the classes theory”1765 – that status deteriorates only when effects would relegate the water body to a lower status class; or

• “the status quo theory” – that all detrimental changes are to be prevented.

1759

Case C-461/13 Bund für Umwelt und Naturschutz Deutschland eV v Bundesrepublik Deutschland. Opinion of Jääskinen AG of 23 October 2014;

Judgment of 1 July 2015. 1760 In fact there were three projects but nothing here turns on that and they can be taken as one. 1761 See above, Recital 25, Article 1 and Article 4.1.a.i WFD.

1762 Bund für Umwelt und Naturschutz Deutschland eV (Federation for the Environment and the Conservation of Nature, Germany).

1763 §51 & Ruling §1 – Unless a derogation applies but that is irrelevant here. 1764 Paloniitty, Journal of Environmental Law (2016) 28 (1): 151. 1765 Jääskinen AG §90.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1176. The CJEU opted for the latter.1766 It held that ‘deterioration of the status’ of a surface water body is a

concept of general scope as opposed to one related to the classification of status.1767 It goes beyond the technicalities of the WFD.1768 Were the concept related to classification of status, and as classification depends on the poorest value of the applicable parameters, all the other parameter values could be reduced without legal consequence and Member States would be deterred from preventing deterioration of a surface water body within a status class. The court agreed with Jääskinen AG 1769 that the concept of ‘deterioration of the status of a body of surface water’ must be assessed in relation to any individual substance or any quality element used in the assessment of ecological status within the meaning of the WFD, without this always resulting in a classification change. Accordingly, status deterioration includes deterioration which does not result in re-classification of the water body to a lower class. There is ‘deterioration of the status’ of a surface water body, within the meaning of Article 4(1)(a)(i) WFD, as soon as the status of one quality element,1770 (for example nutrient conditions) falls by one class, even if that fall does not result in a fall in classification of the surface water body as a whole.1771

1177. The Court also adopted the ‘one out all out’ rule whereby a water body must be re-classified down a class “as soon as” the value of one substance or quality element falls below the level for the current status class of the water body.1772 That rule is a specific expression of the precautionary principle.1773

1178. As to criteria for identifying deterioration of the status of a body of water, transitory deterioration is permissible only subject to strict conditions1774 and the threshold for preventing and identifying deterioration is low.1775 The Court rejected both a criterion of ‘serious impairment’ and a criterion based on weighing adverse effects on waters in the balance against water-related economic interests. The Court emphasised that the application of the obligation to prevent deterioration involves no weighing and balancing of interests.1776 Such a balancing only occurs in applying a derogation under Article 4(7) WFD.

1179. As one commentator has said of the WFD, by the Weser case, the alleged management planning instrument (or, to put it another way, a “framework” directive explicitly so entitled) was turned into a more

traditional, formalistic legal tool, affecting individual permitting procedures all across the EU.1777 In that light

and from a practical point of view, while they cannot be expected to be lawyers, experts reporting or

1766 §55 et seq.

1767 i.e. High, Good, Moderate, Poor, Bad. These theories had their origins in academic debate – Jääskinen AG §90.

1768 Jääskinen AG §99. 1769 Jääskinen AG §100. 1770 Within the meaning of Annex V to the Directive.

1771 This could occur, for example, where the moderate status of the water body was determined by the status of one quality element but the status of

another quality element fell from high to good. Also, though not relevant here, if a quality element is already in the lowest class, any deterioration of that

element constitutes a ‘deterioration of the status’ of a body of surface water. 1772 Jääskinen AG §§109 & 110. 1773 Jääskinen AG §101. 1774 Citing the scheme of Article 4 WFD, in particular Article 4(6) and (7). 1775 §67. 1776 §68 Paloniitty, J Environmental Law (2016) 28 (1): 151. 1777 Paloniitty, J Environmental Law (2016) 28 (1): 151.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

advising on environmental effects on water bodies must have regard to the clarifications Weser provides – perhaps especially as to the obligation to prevent deterioration. Weser post-dated the 2011 EIS but preceded by some months the RPS Report 2015 and by some years preceded the 2018 sEIS.

Protect Natur 2017

1180. Protect Natur1778 was an NGO shut out at Austrian Law from challenging the extension of a permit for a river water-fed ski resort snow-making facility granted pursuant to legislation governing water-related matters. Protect Natur said the WFD ecological status of the river had already deteriorated as a result of the facility. It was shut out on the basis that it had not claimed that its rights protected under that legislation been impaired. The CJEU cited the Weser case extensively, and also LZ1779 as to the obligation to ensure effective judicial protection of the rights conferred by EU environmental law, in holding that the effectiveness of the WFD requires that individuals or, where appropriate, a duly constituted environmental organisation be able to rely on it in legal proceedings. EU environmental law rules are usually in the public interest and the objective of environmental NGOs is to defend the public interest. Standing criteria requiring impairment of individual rights must not prevent such NGOs verifying that EU environmental law rules are being obeyed. So a duly constituted environmental organisation operating in accordance with the requirements of national law must be able to contest before a court a decision granting a permit for a project that may be contrary to the WFD obligation to prevent water body status deterioration.

Sweetman v An Bord Pleanála & Breadán Beo

1181. Hyland J in Sweetman1780 considered a permission for water abstraction from a lake. The EPA had identified the lake as a water body but had not classified its water quality status. Hyland J, citing Weser, considered that the concepts of deterioration and good surface water status are inextricably tied to the complex evaluation framework identified in the WFD. The only method of determining WFD status is by the application of WFD methodology. The notice party’s environmental consultants, by hydrological feasibility assessments, had attempted to examine the likely WFD implications of the abstraction and the Board’s Inspector accepted that analysis.

1778 Case C-664/15, Protect Natur-, Artenund Landschaftsschutz Umweltorganisation v Bezirkshauptmannschaft Gmünd, Judgment of 20 December

2017. 1779 Case C-243/15 Lesoochranárske zoskupenie VLK, Judgment of 8 November 2016, EU:C:2016:838.

1780 Sweetman v An Bord Pleanála [2021] IEHC 16 (Hyland J, 15 January 2021). Hyland J referred questions to the CJEU on which, in Case C-301/22, Rantos

AG opined on 21 September 2023 and the CJEU gave judgment on 25 April 2024 However its answers are not relevant to the present proceedings. They

relate to the fact that the EPA had identified that the small lake in question was a water body for WFD purposes but had not classified its water quality

status. To the extent that the CJEU records legal principles relevant to the present proceedings it does so in reliance on established authority which was

argued before me and which I have considered. Notably, Bund für Umwelt und Naturschutz Deutschland, C-461/13, EU:C:2015:433,

(Weser), Commission v Austria, C-346/14, EU:C:2016:322 (Schwarze Sulm), Association France Nature Environnement, C-525/20, EU:C:2022:350 and

Protect Natur-, C-664/15, EU:C:2017:987. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1182. Notably, Hyland J held that the Board was obliged to ensure that the obligations articulated by Article 4(1)(a) WFD (inter alia, the obligation to prevent deterioration) were fully applied in individual authorisation decisions using the detailed and complex framework of the WFD. She observed:

“What is quite striking about the report and the reliance upon it by the Inspector is that it is impossible for me to understand the relationship between either the first or second methodology of screening the proposed abstraction, and the detailed framework and structure of the WFD described above. Nor is there any way of understanding by reference to the WFD why the initial approach is considered no longer an ecologically relevant metric.” 1781

“Rather, what the Inspector is forced to do is to carry out some kind of proxy evaluation which makes reference to the WFD but does not in fact employ the architecture of the WFD. In truth, the Inspector is simply reciting the conclusions of the Ryan Hanley report without carrying out any independent evaluation of the conclusions of those reports by reference to the criteria set out in the WFD.”1782

“The reliance by the Inspector and the Board on some type of proxy evaluation referring to concepts said to stem from the WFD but which did not follow the steps identified by the WFD, does not constitute compliance with the WFD.”1783

“But I cannot evaluate the efficacy or appropriateness of those mitigation measures by reference to the WFD for the reasons set out above. If I were to accept the appropriateness of the Board’s approach to the mitigation measures, I would be doing so without reference to the WFD. This would be to ignore my obligations under the Directive.” 1784

1183. What seems to me significant (if unsurprising) in these remarks by Hyland J are the necessities in development consent procedures of:

• methodology of analysis clearly grounded in the detailed framework and structure of the WFD.

• independent evaluation by the decision-maker, by reference to WFD criteria, of the expert reports of the applicant for development consent.

• a resultant development consent decision comprehensible in its reasoning so as to permit confirmation by the reader – including the court – of the correct application of WFD requirements.

1781 §117. 1782 §120. 1783 §135. 1784 §134.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Reasons

1184. The judgment of Hyland J in Sweetman is a helpful example of the requirement that the reasons for a decision as to WFD issues must enable the reader to “understand the relationship between” the methodology of analysis employed “and the detailed framework and structure of the WFD.”

1185. More generally, that reasons for an impugned decision must disclose its essential rationale and reasoning on the main issues – the main reasons on the main issues – and must in that regard be clear, cogent, and meaningful, in the sense of capable of persuading (they need not actually persuade), is trite law

– e.g. YY, Connelly, Damer, Balz, NECI, and Ballyboden TTG.1785 The recent case of Sherwin1786 made no new law but is an example illustrating these requirements. The degree of reasoning required is not “too exacting” and lies in the “middle ground”,1787 on the spectrum between discursive judgment (not required) and perfunctory anodyne box ticking1788 or “boilerplate” and “generalised and enigmatic statements”1789 (not

sufficient). The courts must apply a “rigorous” and “exacting” scrutiny of reasons but “avoid over-refined

scrutiny” avoid “search for any error, or for some doubt” and generally “attempt to understand fairly what the decision maker has decided”.1790 What is required is a “broad and common sense approach” in asking whether the reasons enable the court (and others interested) to “genuinely understand the reasoning process”.1791 What that standard requires in a given case depends on the circumstances of that case. There is no uniform standard of reasons required in judicial review – in identifying the correct spot on the spectrum much depends on the circumstances, including the applicable statutory regime, the nature of the decision, the nature of the decisionmaker, the importance of the issue at stake and whether the decision is of a kind which must be transparent not merely to those involved but to the public at large. So, while the principles applicable are general, their application inevitably results in nuanced judgments in judicial review as to the requirements of reasons – as perhaps the account of the law set out above reveals.

1186. As a general proposition, though not one to be overapplied, in planning and environmental decisions the relevant spot on the spectrum of reasoning tends to lie towards its somewhat more demanding end. That view is taken in light of, no doubt inter alia,

• the high public and private importance of planning and environmental decisions as to their substantive content and effect.

• the high public importance of transparency to the legitimacy of planning and environmental decisionmaking processes which decide matters of appreciable public and national, as well as private and local, controversy and, often, matters of high monetary consequence.

1785 YY v Minister for Justice [2017] IESC 61, [2018] 1 I.L.R.M. 109; Connelly v. An Bord Pleanála [2018] IESC 31; [2018] 2 ILRM 453, [2021] 2 IR 752 §30,

Damer v. An Bord Pleanála [2019] IEHC 505 §§5 – 7, 36, 39; Balz v An Bord Pleanála [2020] 1 I.L.R.M. 637; Ballyboden Tidy Towns Group v An Bord Pleanála

& Shannon Homes [2022] IEHC 7 §272. 1786 Sherwin v An Bord Pleanála [2024] IESC 13. 1787 Connelly §84. 1788 Connelly §30, 82, 96. 1789 YY §79. 1790 YY §80. 1791 YY §80.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• the high level of protection of the environment at which EU environmental policy and law aims via the precautionary and preventive principles – Article 191 TFEU and Waddenzee.1792

• the high stakes created by the legal fact that the precautionary and preventive principles may require the preference of environmental to economic interests, with substantial economic consequences, even where the environmental risks remain uncertain – Bayer.1793

• the fact that such decisions are typically made by expert, specialist, quasi-judicial tribunals of whom appreciable cogency may reasonably be expected.

1187. The observation that reasons must enable the reader to “genuinely understand the reasoning process” is important in its requirement of disclosure not just of conclusionary reasons but of the “rationale”, 1794 a “reasoning process”. A “standard of reasoning” applies – which chimes with the requirement of “engagement”. Reasons must disclose that the decision-maker has “truly engaged” with the evidence before it and submissions made to it. It is a fundamental and basic element of any decision-making affecting the public that relevant submissions should be addressed, and an explanation given why they are not accepted.1795 Of some relevance to this case, Connelly makes it apparent that the required reasoning may be “complicated or scientific if the issues which arose in the context of the grant or refusal of permission required engagement with such issues.”1796 While Clarke CJ was speaking of engagement by the member of the public seeking to understand the reasons, his implicit but necessary premise is such engagement by the

decision-maker in expressing them.

Nordrhein-Westfalen 2020 & Association France Nature 2022 – Localised & Temporary Effects

1188. Nordrhein-Westfalen1797 concerned permission for a road project which authorised the developer to discharge rainwater running off the road surfaces into surface water or groundwater.1798 Hogan AG cited the Weser case extensively and observed that Article 4(1) WFD is similar to Article 6(3) of the Habitats Directive in that both integrate the precautionary principle and both may require refusal of a development consent.1799 And Article 4 WFD imposes a general obligation to prevent surface water deterioration – without referring to Annex V classification.

1792 Case C-127/02, Judgment of 7 September 2004 §44.

1793 Cases T-429/13 and T-451/13, Bayer CropScience AG, et al v European Commission, Judgment of the General Court, 17 May 2018.

1794 Damer §5. 1795 NECI §155 citing Balz. 1796 §74. 1797 Case C-535/18. IL et al v Land Nordrhein-Westfalen, Judgment of 28 May 2020.

1798 The Court held that the WFD objectives and Article 4 obligations for surface water and groundwater are largely identical.

1799 Opinion of 12 November 2019: “45. … the authorisation criterion laid down in the second sentence of Article 6(3) of the Habitats Directive integrates

the precautionary principle and thus seeks to prevent adverse effects on the integrity of protected sites as the result of the plans or projects being

considered. For this reason, the competent national authorities can authorise the activity at issue only if they have established that it will not adversely

affect the integrity of the site involved. This must also be the case for Article 4(1) of the Water Framework Directive since this directive is based on Article

175 TEC (now Article 192 TFEU). As such, it contributes to the achievement of the objectives of Union policy on the environment, which is based — as

expressly required by Article 191(2) TFEU (ex-Article 174(2) TEC) and indicated in recital 11 of the WFD — on the precautionary principle.”

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1189. The CJEU in Nordrhein-Westfalen also cited Weser and held that:

• Checks to establish whether the project may have adverse effects on water – whether Article 4 WFD requirements to prevent status deterioration are met – must precede permission of a project.

• So, and for EIA purposes, the information made public in the permission process by the developer must include the data necessary to assess the effects of the project on water, in light of Article 4 WFD and within the limits of what a developer may reasonably be required to compile.

• That information must allow the public an accurate impression of the impact of the project on the status of water bodies so the public can verify compliance with, inter alia, Article 4 WFD.1800

• The data must show whether, by WFD criteria, the project is liable to result in a deterioration of a water body.

• Members of the public must be able to litigate a breach of the obligation to prevent water body deterioration if the breach concerns them directly.1801

• The threshold for preventing and finding water body status deterioration “must be as low as possible”1802

– the words “as possible” are an added gloss on the similar finding in the Weser case.

1190. The court also found that, though chemical status classification of groundwater is done by aggregating the results of the monitoring points in the water body, this does not mean that, to find status deterioration, the whole groundwater body must be affected. Even if the chemical status of the body generally remains good, failure to comply with one quality element at a single monitoring point is deterioration of chemical status of the water body as representing deterioration of at least a significant part of the groundwater body.1803

1191. I observe that it may be that this finding as to deterioration at a single monitoring point is particular to groundwater. The WFD is arguably more demanding as to monitoring and selection of groundwater monitoring points than as to surface water. The Court cited the role and the importance of each monitoring site in the system for monitoring groundwater. And the finding proceeded in part from the obligations of Directive 2006/118 as to Groundwater Protection.1804 But, in appreciable part, the CJEU’s view proceeded from requirements, applicable to both groundwater and surface water, that monitoring provide a coherent and comprehensive overview of status, such that location of monitoring points be representative. I infer that the idea that monitoring locations be representative envisages their being representative of both the water body in general and of its WFD status.

1800 An incomplete file or data that scattered incoherently, across a multitude of documents will not suffice.

1801 So, well users drawing from a groundwater body had standing as to risk of its deterioration. Hogan AG had suggested that since any requirement

imposed by a member state to be ‘directly concerned’ restricts access to justice, it must be interpreted restrictively.

1802 §101. 1803 §111 - 118.

1804 Under Article 4(5) of that directive, Member States are, in accordance with Article 11 of Directive 2000/60, required to take such measures as may

be necessary to protect aquatic ecosystems, terrestrial ecosystems and human uses of groundwater dependent on the part of the body of groundwater

affected by that exceedance. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1192. And Hogan AG considered that the notion of ‘deterioration of the status’ of water bodies for WFD purposes “should be rather similar”, irrespective of whether it is surface water or groundwater. The CJEU agreed and

• repeated Weser to the effect that the objectives of the WFD and the obligations deriving from Article 4(1)

WFD as to surface water and groundwater, are largely identical and that “the concept of ‘deterioration of the status’ of waters is a concept of general scope” and

• held that the same principles determine the scope of the concept of ‘deterioration of the status’ of water, irrespective of the type of water concerned.

Also, and though the CJEU does not say so, it is not difficult to infer that this view is informed by the view that the threshold for preventing and finding water body status deterioration “must be as low as possible” and from the precautionary and/or preventive principles and perhaps, given the finding relates to localised deterioration, to the prioritised principle of rectification of environmental damage at source.1805

1193. In Association France Nature1806 Rantos AG both

• proposed to apply Nordrhein-Westfalen such that failure to comply with one quality element at a single monitoring point indicates a status deterioration for at least a significant part of a surface water body1807 and

• seemed to suggest that this could be decided on a single measurement.1808

The CJEU held that

• the threshold for breach of the obligation to prevent status deterioration must be as low as possible.

• temporary, short-term impacts without lasting consequences can be ignored only where it is established that they have little effect on water body status and so can’t lead to a deterioration of status.

• but temporary, transitory or short-term impacts which cause even a temporary status deterioration require refusal of development consent. But the CJEU did not address, positively or negatively, the proposal by Rantos AG that failure to comply with one quality element at a single monitoring point indicates a status deterioration for at least a significant part of a body of surface water.

1194. However, as I have said, MOWI and ALAB did not approach the issue of WFD status deterioration by reference to effects likely at EPA monitoring points or effects on the bay as a whole. In the impugned

decision they approached it, in a sense, more rigorously – asking whether the limit of 170 µgDIN/l would be exceeded at the Shot Head Site. I am concerned with ALAB’s decision as made – not to rewrite it as a

decision it might have made but did not make. Accordingly, I need not decide whether deterioration of

1805 For all these principles see Article 191 TFEU. 1806 Case C-525/20 Association France Nature v Premiere Minsitre, Opinion of Rantos AG of 13 January 2022, §59-61. 1807 §26. 1808 Nordrhein-Westfalen had stated that principle solely as to groundwater.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

water status as to DIN is to be decided by reference to the bay as a whole or by reference to a particular location if so, what location – other than at the Shot Head Site. Nor need I decide whether, in any remitted

decision-making process, a particular view might be taken on that question.

WFD – Irish Regulation

1195. As stated above, Ireland’s Water Policy Regulations1809 and Surface Waters Regulations1810 transpose and implement WFD, EQSD and DSD1811 requirements. I have referred to the Surface Waters Regulations above as to the question whether it is correct to refer to an “EQS” for DIN. The Surface Waters Regulations transpose and implement WFD, EQSD and DSD inter alia by measures envisaged in Art. 2, headed “Purpose and scope of the Regulations”, which lists inter alia:

(a) The protection of surface water bodies whose status is high or good.

• This relates to the Article 4 WFD obligation to prevent deterioration. That obligation is reflected in Article 28(1) of the Surface Waters Regulations.

(b) Establishing EQSs for priority substances and certain other pollutants as provided for in Article 16 WFD – to apply in calculating the chemical status of surface water bodies;

(e) Establishing EQSs for pollutants listed in Annex VIII, Points 1 to 9 WFD to apply in calculating the ecological status of surface water bodies.

• §(b) and §(e) do not include nitrogen. §(e) clearly relates to Annex V §1.2.6 WFD. §1.2.6 specifically relates

to chemical pollutants potentially affecting the ecological status of surface water bodies and, as has been noted, notably excludes Point 11 of Annex VIII which reads “Substances which contribute to eutrophication (in particular, nitrates and phosphates).”

(f) Establishing EQSs for the general conditions specified in Annex V WFD to apply in calculating the ecological status of surface water bodies.

• As has been seen, the “general conditions” specified in Annex V WFD1812 do include nutrients. Both Annex V WFD and §(f) clearly relate to ecological status and Annex V §1.2.6 WFD, as has been said, excluded nutrients from its prescription that EQSs be set. It is not apparent from Annex V WFD that EQSs should be set for nutrients. Yet §(f) in effect explicitly envisages EQS for nutrients.

1809 European Communities (Water Policy) Regulations 2003 (S.I. No. 722 of 2003) and various amending regulations to the European Union (Water

Policy)(Amendment) Regulations 2022.

1810 European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009) as amended, including by the European

Union Environmental Objectives (Surface Waters) (Amendment) Regulations 2019 (S.I. No. 77 of 2019). 1811 Directive 2006/11/EC – the Dangerous Substances Directive. 1812 Specified at Tables 1.2.1 to 1.2.4 as “Physico-chemical quality elements”.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• However, the Surface Waters Regulations are not impugned in these proceedings and absent argument on the issue and given, not least, the complexity of the WFD regime, I make no finding in this regard.

(g) Establishing the EQRs for the biological quality elements specified in Annex V WFD that represent the boundaries between high, good and moderate ecological status surface water.1813

1196. Art. 2 also envisages rules1814 for the presentation and reporting of surface water monitoring results and the classification of ecological status and the chemical status of surface water bodies as required by Article 15,1815 Annex V1816 and Annex VII1817 WFD.

1197. Art. 3 of the Surface Waters Regulations 2009 defines a “representative monitoring point” as a surface water monitoring point identified by the EPA in accordance with the 2003 Regulations1818 and to be used to calculate the status of a surface water body.1819

1198. EQS is not defined in the Surface Waters Regulations 2009 and so, by Article 3(2) has the meaning given it in the WFD: “The concentration of a particular pollutant or group of pollutants in water, sediment or biota which should not be exceeded in order to protect human health and the environment.”1820

1199. Arts. 7 and 9 of the Surface Waters Regulations 2009 require licensing authorities such as ALAB to

“aim to achieve”, inter alia, the “the environmental quality standards set out in Schedules 5 and 6”. In light of

Weser and following cases, this must be understood as a binding aim – an obligation to refuse development consent for project which may result in water body status deterioration as that concept is explained in the cases.

1200. Arts. 32 and 33 of the Surface Waters Regulations 2009 reflect Art. 4.6 & 4.7 WFD. In effect, they prohibit both temporary surface water status deterioration and failure to prevent deterioration of a surface water body from high status to good status resulting from new sustainable human development activities save, in each case, on strict conditions set out in those articles and Article 34. Those conditions are not here relevant – what is relevant are the prohibitions. (And the prohibition on temporary deterioration must now be read in light of Association France Nature).

1813 In accordance with Article 4 WFD. 1814 See Article 2 – Purpose and scope of the Regulations. 1815 Reporting to the Commission of river basin management plans and updates of them. 1816 Surface Water Status. 1817 River Basin Management Plan content.

1818 Article 10 European Communities (Water Policy) Regulations 2003 (S.I. No. 722 of 2003) and various amending regulations to the European Union

(Water Policy)(Amendment) Regulations 2022. 1819 And which does not lie within the mixing zone of a point of discharge. 1820 Article 2.35.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1201. Part IV of the Surface Waters Regulations 2009 – Arts. 35 to 40 – addresses the calculation of ecological and chemical status.1821 Art. 35 requires the EPA to assign surface water status based on monitoring results. Art.36 deems surface water ecological status to be the lowest of the biological and the physico-chemical quality element status.1822 Art. 37(1) provides that a surface water body shall be classified as being of high ecological status, where all of the following are true:

• the EQRs for the biological quality elements show no or only very minor evidence of distortion from undisturbed or reference conditions, o This relates, inter alia, to the benthos.

• there are only very minor anthropogenic alterations to the status of the hydromorphological quality elements, o This is irrelevant to the issues in this case.

• the values for the biological and physico-chemical quality elements satisfy the “criteria” set in Schedule 5

of these Regulations. o This relates, inter alia, to the benthos and to DIN.

1202. In 2009, Schedule 5, Table 8, entitled “Biological Quality Elements” did not set criteria for benthos in Coastal Waters.1823 The amending 2019 Regulations did so as follows:

Ecological Quality Ratio Biological Quality Element Classification System High-good boundary Benthic invertebrate fauna IQI — Infaunal Quality Index 0.75

1203. In 2009, Schedule 5, Table 9, entitled “Physico-chemical conditions supporting the biological

elements”, in “Part A General Conditions” under the heading “Nutrient Conditions”, stipulated for DIN in high

ecological status coastal waters a criterion of “≤ 0.17 mg N/l” (≤ 170 µgN/l). The amending 2019 Regulations

substituted the following, which I have extracted from the table:

Nutrient Conditions1824 Coastal water body (winter and summer)

Dissolved Inorganic Nitrogen (mg N/L) High status Good status (1)

(0 psu ) (0 psu(1))

≤ 1.0 ≤ 2.6

1821 We can ignore here the calculation of ecological potential.

1822 Save that in assigning high ecological status the hydromorphological quality elements are considered also. But that is irrelevant to this case.

1823 Presumably pending completion of the EQR intercalibration referred to in section Annex V §1.4.1 WFD. Article 44 of the Surface Waters Regulations

2009 addressed “Knowledge gaps in classification systems” and recorded that it had not been possible to complete the intercalibration exercise and so

it had not been possible to set national classification boundaries at quality element level for all the biological quality elements referred to in Annex V §1.2

WFD. It obliged the EPA to programme completion of that work but meanwhile to “keep under review … the boundary conditions and environmental

quality standards established for the general supporting conditions and specific pollutants listed in Tables 9 and 10 of Schedule 5”.

1824 Whether or not an EQS, this is clearly not an ELV.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Nutrient Conditions1824 Coastal water body (winter and summer)

High status Good status (34.5 psu(1)) (34.5 psu(1))

≤ 0.17 ≤ 0.25

(1) Linear interpolation to be used to establish the limit value for water bodies between these salinity levels based on the median salinity of the water body being assessed.”1825 Note: the S.I. defines “psu” as a “Practical Salinity Unit”.

1204. Donal Grant1826 has deposed to the determination of this standard for DIN at 34.5 psu (≤170 µgDIN/L) on foot of cumulative monitoring results from sampling programmes at coastal water monitoring sites around Ireland in undisturbed conditions. He says that Table 9 implements the requirement of Annex V, Table 1.2.4 WFD that, for high ecological status in coastal waters, nutrient concentrations remain within the range normally associated with undisturbed conditions.

1205. The EPA has stated1827 that the ≤170 µgDIN/L standard is set by reference to “median” values. But while other standards set in the Surface Waters Regulations are explicitly set as medians1828 and, indeed, others again are set as “means”,1829 and leaving aside the question of legal basis for means or medians, of what samples, in what number, over what time periods, the ≤170 µgDIN/L standard is not explicitly set in

Schedule 5, Table 9 of the Surface Waters Regulations by reference to “median” – or for that matter “mean”,

values. I have referred above to the witness evidence in that regard. I reiterate that there is no challenge to the Regulations or to the EPA’s methods but the observation may be part of an explanation of the approach to DIN in fact taken by all concerned in the Aquaculture Licence application.

1206. Art. 39 of the Surface Waters Regulations 2009 requires that, when assigning surface water body ecological status, the EPA must be satisfied that the values of its general physico-chemical quality elements (which include DIN) are consistent with achieving the EQR specified for the biological quality elements. So the EPA must establish and publish, before classifying the waters, the permitted statistically based range within which the general physico-chemical quality elements (including DIN) may deviate from the “values” specified in Schedule 5 in order to ensure ecological relevance and avoid a mismatch between the monitoring results for the biological and the general physico-chemical quality elements. That range is not in evidence.

1825

A similar footnote in the 2009 version is difficult to understand as the 2009 version set a criterion only for waters of 34.5 psu. In any event, despite

the scope for interpolation by reference to psu, the case proceeded on the basis that the standard of ≤ 0.17 mg N/L of DIN (≤170µgDIN/L) applied –

perhaps unsurprisingly as the EIS records salinity close to 35% throughout the year – see EIS Vol 1 §2.6.4. 1826 of the Water Policy Division of the Department of Housing, Local Government, and Heritage. 1827 EPA letter 4 May 2023. 1828 For example, for Molybdate Reactive Phosphorus in Transitional Coastal water bodies.

1829 For example, for Molybdate Reactive Phosphorus in River water bodies. And EQSs for specific pollutants are set by Table 10 as annual averages (it

seems, means) and maximum allowable concentrations. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1207. Art. 40 of the Surface Waters Regulations 2009 provides that for a surface water body to be classified high or good ecological status, the concentration of “Specific Pollutants” at any representative monitoring point shall not exceed the EQS set out in Table 10 of Schedule 5. Presumably this is intended to effect WFD Annex V Tables 1.2.1 to 1.2.4 as to specific pollutants. As to specific pollutants in high or good ecological status coastal waters, Table 1.2.4 states as follows:

Physico-chemical quality elements1830 Element High status Good status Specific Concentrations close to zero and at Concentrations not in excess of the standards set synthetic least below the limits of detection of in accordance with the procedure detailed in pollutants the most advanced analytical section 1.2.6 without prejudice to Directive techniques in general use. 91/414/EC1831 and Directive 98/8/EC.1832 (< EQS). Specific non- Concentrations remain within the Concentrations not in excess of the standards set synthetic range normally associated with in accordance with the procedure detailed in pollutants undisturbed conditions (background section 1.2.61833 without prejudice to Directive levels = bgl). 91/414/EC and Directive 98/8/EC. (< EQS).

1208. It is not readily apparent from WFD Annex V Table 1.2.4 that mere satisfaction of the EQS as to a Specific Pollutant suffices in high ecological status coastal waters. It appears that satisfaction of an EQS merely justifies good status. However, as Schedule 5, Table 10 of the Surface Waters Regulations does not list any pollutant relevant to this case, and as WFD Annex V Table 1.2.4 lists Nutrients separately to Specific Pollutants, I need pursue that question no further. In similar vein, it is noteworthy that, as to chemical status, some EQSs are expressed in terms of “maximum allowable concentration” such that, by Article 41(2) of the Surface Waters Regulations 2009, they are breached in the event of an exceedance at “any representative monitoring point within the water body”. No such provisions appear as to DIN. Possibly, one implication of the exclusion of DIN from these provisions may be that DIN exceedances at a “single representative monitoring point” do not, per se, prevent a surface water body being classified high ecological status and that DIN exceedances have that effect only if calculated for the water body generally as opposed to for a particular location within it. However, and again given the approach to DIN taken by all parties in the Aquaculture Licence application – i.e. in terms of exceedance of the ≤170 µgDIN/L standard locally to the fish farm site – I need pursue that question no further.

1830 The following abbreviations are used: bgl = background level, EQS = environmental quality standard.

1831 Plant Protection Products Directive 91/414/EC replaced by Regulation (EC) No 1107/2009 concerning the placing of plant protection products on the

market.

1832 Biocides Directive 98/8/EC replaced by Regulation 528/2012 concerning the making available on the market and use of biocidal products.

1833Application of the standards derived under this protocol shall not require reduction of pollutant concentrations below background levels: (EQS >bgl).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1209. Art. 41 of the Surface Waters Regulations 2009 provides that the chemical status of a surface water body shall be assigned by the EPA according to the monitoring results for the chemical substances and their EQSs established in Schedule 6. Schedule 6, Table 11 sets EQSs for priority substances and certain other pollutants. Schedule 6, Table 12 sets EQSs for priority hazardous substances. Neither lists DIN or EmBz.1834

WFD – Role of the EPA

1210. For reasons I have described, I enquired during the trial as to the EPA’s role and method in classifying and monitoring the status of waterbodies – as to both of which it is the competent authority on foot of the Water Policy Regulations 20031835 which empowered it, in accordance with Articles 7(1) & 8 WFD, to establish systems for classifying water bodies1836 and monitoring water body status.1837 My inquiry resulted, in turn, in inquiry of the EPA and nobody objected to my having regard to their reply.1838 Its content includes the following – as to which my comments involve some repetition of points noted earlier:

i. The EPA is the competent authority for identifying and assigning WFD status to water bodies in Ireland. It is also the competent authority for preparing and implementing the National Water Quality Monitoring Programme.

ii. The EPA assigns ecological status to waterbodies based on the monitoring programme results and in accordance with the requirements of the WFD, national legislation and any relevant guidance.

Coastal water bodies are monitored at multiple sites/stations in each water body for a range of biological and physico-chemical parameters and the results are combined to give an overall ecological status and chemical status for that water body.

iii. The EPA assesses water body status every 3 years using data gathered over the previous 6 years.

iv. Schedule 5 to the Surface Waters Regulations 20091839 sets “standards” for general physico-chemical quality elements of ecological status (for example nutrient conditions including DIN1840).

1834 The same is true of Schedule 6 Table 13 Watch list of substances for Union-wide monitoring as set out in Article 8b of Directive 2013/39/EU inserted

by S.I. No. 386 of 2015 and in turn replaced by S.I. No. 77 of 2019. 1835

European Communities (Water Policy) Regulations 2003 S.I. No. 722 of 2003. Article 24 et seq and Part IV empower the EPA to classify waters and

govern the calculation of status. 1836 Article 9 – Systems for, as here relevant,

(a) estimating the values of the biological quality elements specified for each surface water category, ……

(b) the classification and presentation of the ecological status and chemical status of surface waters, and (c) …..

1837 Article 10 – subject to ministerial amendment and to be sent to the EU Commission. 1838

EPA to Philip Lee LLP 04 May 2023 (Philip Lee LLP is on record for ALAB). I have supplemented the description slightly for clarity where that is reliably

possible. A supplementary inquiry by SWI of the EPA by letter 5 May 2023 received a reply dated 9 May 2023 replying also to an enquiry by Dr Bass,

apparently made on his own initiative, as to the status of Outer Bantry Bay. The EPA states that it has no role in monitoring compliance with licenses

issued by other licensing authorities such as ALAB. It does not seem to me that this reply is of present relevance.

1839 European Communities Environmental Objectives (Surface Waters) Regulations (2009 S.I. No. 272 of 2009) as amended – including by the European

Union Environmental Objectives (Surface Waters) (Amendment) Regulations (2019 S.I. No. 77 of 2019).

1840 See above – Schedule 5 sets “Criteria For Calculating Surface Water Ecological Status And Ecological Potential” and Table 9 of Schedule 5, under the

heading “Nutrient Conditions”, stipulated for DIN in a High Status Coastal water body a criterion of “≤ 0.17 mg N/l = 170 µgN/l.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

v. The EPA reply does not refer to an “EQS” for DIN, or use the term EQS at all save in a footnote. In that footnote it identifies the “standards” set in Schedule 5 as the “criteria” for calculating surface water ecological status. It refers to these “standards” or “criteria” as the “environmental quality standards” (i.e. EQSs) set out in Schedule 5 Tables 8, 9 and 10. However only Table 10, as to specific pollutants, lists “Environmental quality standard (EQS)”. Table 8, as to biological quality elements, predictably lists “Ecological Quality Ratio” – and they are clearly not EQSs. And, as stated earlier, Table 9, as to Nutrients, including DIN, uses no such descriptor of the standards it sets.

vi. The EPA says it developed assessment criteria which were used to set ecological status “boundaries”

(including for DIN), according to the normative definitions from Annex V WFD.1841 (This seems, clearly enough, to relate to the boundaries between the water quality status classes as they relate to each quality element.) By this observation the EPA appears to assert that it, in effect, set the values set in Schedule 5, Tables 8, 9 and 10 of the Surface Waters Regulations 2009. Subject to the caveat of the intercalibration exercise, this is unsurprising.

vii. The EPA refers to “threshold”1842 values for DIN as, it says, detailed in Schedule 5 Table 9 of the Surface Waters Regulations.1843 I presume that “thresholds” and “boundaries” are synonymous – though why two terms are used is unclear. That for High status coastal water is <170 μgDIN/l.1844

viii. The EPA says that the “median” values for the ‘Good’ and ‘High’ status “boundaries” are set out in

the Surface Waters Regulations.1845 (As I have said, that is explicitly true of some quality elements1846 and others are set as “means”.1847 But that for DIN is not explicitly set as either.) But, both expert witnesses said that Schedule 5 Table 9, in setting the high ecological status standard for coastal waters at <170 μgDIN/l does not, if itself, provide a practically applicable standard.

ix. The EPA says that Art. 39 of the Surface Waters Regulations 20091848 requires the EPA to establish and publish a statistically-based range within which those quality elements of ecological status may deviate from standards set in Schedule 5 “in order to ensure ecological relevance”.1849 Neither the expert witnesses nor any of the materials before ALAB referred to this possibility of deviation or publication nor is it in evidence.

1841 The letter says “Appendix V” but the correct meaning is clear. 1842

It is unclear why the EPA introduces a new term – “threshold” – thereby creating the potential for confusion – at least amongst the non-cognoscenti

– but the reference is clearly to the EQS set by Table 9.

1843 European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009) as amended – including by the European

Union Environmental Objectives (Surface Waters) (Amendment) Regulations 2019 (S.I. No. 77 of 2019).

1844 In full salinity (34.5psu). The letter refers to interpolation to adjust the threshold by reference to salinity but the environmental assessments and

these proceedings proceeded on the basis that the standard of ≤ 0.17 mg N/L of DIN applied unadjusted – perhaps unsurprisingly as the EIS records

salinity close to 35% throughout the year – EIS Vol 1 §2.6.4.

1845 European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009) as amended – including by the European

Union Environmental Objectives (Surface Waters) (Amendment) Regulations 2019 (S.I. No. 77 of 2019). 1846 E.g. Molybdate Reactive Phosphorus in a river or lake.

1847 E.g. Total Ammonia and Molybdate Reactive Phosphorus in a transitional water body and Phytoplankton in coastal waters.

1848 European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009 as amended) – including by the European

Union Environmental Objectives (Surface Waters) (Amendment) Regulations 2019 (S.I. No. 77 of 2019).

1849 and so as to avoid a mismatch between the monitoring results for the biological and the general physico-chemical quality elements.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

x. The EPA says that its nutrient monitoring methodology sets out the procedure for assessing DIN – using a robust statistical assessment of ecologically relevant values. The methodology is not set out but the reader is referred to a detailed description of the assessment methodology first published in the EPA’s Water Quality in Ireland 1998-2000 (Toner et al, 2002) and used since. This document does not seem to have been known to the expert witnesses and, as to the period it covers, predates the WFD implementation deadline – though it was published after the WFD was promulgated. The frequency of monitoring for different parameters (including DIN) in each waterbody type is said to be outlined in the National Water Quality Monitoring Programme1850 and meets or exceeds the minimum requirements of the WFD. As to DIN, data is collected for all stations in the water body for each season (winter and summer).

xi. The EPA states how the DIN standard is applied in practice, including by way of adjustment for salinity.1851

However, Bantry Bay salinity has been regarded by all as constantly in the region of 34.5 psu such that the ≤170 µgDIN/L standard is applicable without salinity adjustment. So it is possible to simplify the EPA’s description of its method to the following:1852

▪ Monitoring data for DIN is collected for all stations in the water body in winter and summer.

▪ The median summer and median winter values for DIN are calculated using that data from the most recent 3-year dataset.

▪ If the median DIN value is ≤170 µgDIN/L High status is assigned.

▪ If the median DIN value is >170 µgDIN/L good or lesser status is assigned.

1211. The EPA reply is helpful. But:

• As water body status is assessed every 3 years using a 6-year dataset, I confess to remaining unclear what is the precise significance or purpose of calculating median summer and median winter values for DIN using a 3-year dataset and how often that calculation is performed.

• As I have said, Schedule 5, Table 9 of the Surface Waters Regulations does specify means and medians as to specific quality elements. But it does not do so as to DIN – as to which neither the word “mean” nor “median” is used.

• Nor, for that matter, does Table 9 specify that the DIN limit is a maximum as opposed to a mean or median value, or whether compliance is to be measured by reference to

1850 The letter does not say so but I infer it was adopted pursuant to Article 10 of the 2003 Water Policy Regulations.

1851 By linear interpolation – see Schedule 5 Table 9 as to Nutrient Conditions. This is not an issue in the present case.

1852 I have edited the text for clarity where safe to do so.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

o exceedances at a single monitoring point (perhaps based on a median of results over time at that point)

o many monitoring points (perhaps based on a median of results at a single point in time or based on a median of results at multiple points in time).

1212. I am grateful for the information provided by the EPA. They were not parties to this action and were prevailed upon at short notice to voluntarily assist. While I cannot say I fully understand it on the limited information to hand, (and if this matter is remitted to ALAB they may wish to pursue the issue) the EPA methodology was not challenged in these proceedings nor do I suggest it is in any way deficient by reference to WFD requirements.

1213. For present purposes, the important point seems to be that the EPA calculations are of medians from all the stations in the water body. In other words, as to DIN, the calculations are as to the condition of the entire water body rather than as to a particular locus within the water body. This is consistent with the evidence of Dr O’Toole that compliance with the WFD is assessed by the EPA on an average of samples across the entire water body.1853

1214. However, and in contrast, the 2011 EIS for MOWI, RPS for MOWI in 2015, the 2018 sEIS for MOWI, Dr Saunders and ALAB – and, for that matter, the objectors – did not describe the EPA’s method for calculating ecological status or consider the effect of the fish farm on ecological status. They all proceeded on the basis that, however measured, avoidance of the exceedances of the limit of 170 µgDIN/l local to the Shot Head Site, considered on a “worst case” basis, represented the standard which MOWI had to meet. As I have said, it seems to me proper to approach these proceedings on that basis. But I do so without prejudice to any arguments which might be made in a remitted or other process before ALAB for or against a different approach. More generally, it seems to me entirely possible that evidence different to that before me could appreciably affect the outcome of a different case.

1215. I respectfully express considerable sympathy with all concerned at a confusion which seems to me to proceed first from the complexity of the WFD, second from a lack of clarity in the Surface Waters Regulations 2009 as to the basis on which the limit of 170 µg DIN/l is to be applied and third from a lack of clarity (I emphasise, on the evidence before me) as to the substance and basis in law of the EPA’s monitoring regime. I should add that it would not surprise me if, in the legal thicket from which the limit of 170 µgDIN/l protrudes, there is to be found a satisfactory answer to these questions and I am far from suggesting that the Irish classification system is deficient. But on the evidence before me and submissions to me have found no answers to these questions and the parties and, notably, their expert witnesses directed me to none.

1853 Day 5 p98.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

WFD – DIN – Approach taken in Licence Application

WFD – Introduction, EIS 2011 & EIA 2015

1216. In my earlier judgment in this case I expressed puzzlement as to how modelled “typical” DIN levels could form part of a “worst case scenario” for purposes of WFD compliance analysis. Despite crossexamination of experts and lengthy argument at trial, I remain puzzled. I do not rule out such a possibility, but no one has explained it – either in the papers leading to and constituting the Impugned Decisions or at trial despite cross-examination of expert deponents.

1217. A striking features of the water quality materials submitted by MOWI, and of the subsequent analysis by the Minister and ALAB, is the belatedness of addressing the WFD at all and their simplicity of approach (if not of method) despite the complexity of the WFD regime described above. For example, there was no detailed description or assessment at any time of the WFD monitoring regime in place for Bantry Bay. Nor was there analysis whether deterioration of water quality was to be assessed at any in particular of the monitoring points or by way of an averaging of results (whether by mean or median) from those monitoring points as representative of the general status of the water body – as opposed to its condition at particular localities within the water body, such as the site of the Shot Head Salmon Farm. To some, but a limited, extent this is explicable as to materials predating clarification of the WFD provided by the Weser case in July 2015 and the other cases cited above. But it is not explicable thereafter and as to a determination made by ALAB in 2021.

1218. MOWI took the simpler – and in fairness more demanding – approach of asking whether the <170 µgDIN/l standard set by the Surface Waters Regulations would be exceeded in the immediate vicinity of the Salmon Farm. As far as that goes, it was a worst-case scenario approach as that was the locus in which, at least as a general proposition, exceedances were most likely to occur. SWI is correct in objecting that it was only in written submissions and at trial that an attempt was first made to suggest that the question of risk of deterioration of water body status ought to be assessed on the basis of the general status of the entire water body – as opposed to on the basis of conditions local to the Salmon Farm. This was not the approach taken by MOWI and ALAB in the licensing application.

1219. While the 2011 EIS addresses nitrogen discharge from the Salmon Farm, the risk of eutrophication, “whether Bantry Bay has the nutrient carrying capacity to comfortably accommodate these salmon farming activities”1854 and an “EQS” set by SEPA1855 for nitrate nitrogen at 168 µgNO3N/l (which it says is not even

1854 §4.7. Quantifying the maximum impact of soluble salmon farm discharges in Bantry Bay; dilution box model. 1855 The Scottish EPA.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

approached), it does not even mention the WFD or Ireland’s Water Policy Regulations1856 and Surface Waters Regulations.1857 The Minister’s EIA1858 in June 2015 mentions the WFD – but only once and in passing in the context of EQSs (inter alia for EmBz) imposed by the Control of Dangerous Substances in Aquaculture Regulations 2008.1859 Those regulations do not address DIN. Nor, though it briefly discounts risk of nitrate eutrophication, does the EIA address the issue of risk of deterioration of WFD water quality status. Indeed, it does not mention WFD water quality status at all. It appears that there was no real consideration of the WFD as to DIN in the process before the Minister and until the Appeal to ALAB.

WFD – RPS Report 2015

1220. The RPS Water Quality Modelling Report of November 20151860 post-dates the Appeal to ALAB. It mentions the WFD only twice – and that by way only of brief description as the source of the Surface Waters Regulations 2009. While that is de facto surprising it is not necessarily a problem de jure as it is the transposing regulations which represent the directly effective law in Ireland. Indeed, this is the more understandable as the RPS report preceded Weser – until which decision the role of the WFD, and of Article

4 in particular, in development consent applications was disputed and unclear – not least as the WFD was considered a framework directive only. The RPS Report describes its basis in regulation and standards as being the Control of Dangerous Substances in Aquaculture Regulations 2008, SEPA guidelines and EQS’s for salmon farming1861 and the Surface Waters Regulations 2009.1862 The RPS Report concludes that the impact of the proposed increases in salmon production would fall within the acceptable limits set out by the Surface Waters Regulations 2009 as to surface water quality and by the SEPA EQSs for salmon farming. 1863 The report approximates the Surface Waters Regulations DIN limit for high status coastal waters of <170 μgDIN/l with SEPA's EQS for DIN of 168μgNO3/l. But it generally adopts SEPA’s EQO/EQS approach and states that “In respect of the water quality parameters assessed ….. the EQS’s used are taken directly from SEPA’s Manual on Marine Cage Fish Farming, where further information on each can be found.”1864 RPS does not state what was lacking in the Surface Waters Regulations, the EPA’s methodology or any other relevant Irish sources which rendered reliance on the Scottish EQSs and guidance preferable. RPS states that the EQS comprises

1856 European Communities (Water Policy) Regulations 2003 ( S.I. No. 722 of 2003 ) and various amending regulations to the European Union (Water

Policy)(Amendment) Regulations 2022.

1857 European Communities Environmental Objectives (Surface Waters) Regulations 2009 (S.I. No. 272 of 2009) as amended, including by the European

Union Environmental Objectives (Surface Waters) (Amendment) Regulations 2019 (S.I. No. 77 of 2019). 1858 See generally pp35 – 37. 1859 European Communities (Control of Dangerous Substances in Aquaculture) Regulations 2008 (S.I. No. 466 of 2008).

1860 Water Quality Modelling for all existing & currently proposed salmon farm sites in Bantry Bay IBE0744/R07/Rev02/NS Marine Harvest Ireland

November 2015. The study was undertaken in two phases; first the hydrodynamic model development and validation phase and second, the water quality

modelling phase using the model established in the first phase to assess the effects of all potential discharges into Bantry Bay as a result of all existing

and proposed fish farming activities.

1861 The report says that the SEPA EQSs are well-tested, long-established and very widely adopted, not just in the context of salmon farming but as an

important international benchmark in environmental quality management as a whole. 1862 RPS Report p1. 1863 RPS Report p1. 1864 RPS Report: 5 WATER QUALITY PARAMETERS; 5.1 OVERVIEW; 5.1.1 Introduction.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

the sum of the nitrogen concentrations resulting from fish farm discharge plus the existing ambient concentration.1865

1221. The RPS Report was explicitly prepared on a worst-case scenario basis: “This study considers the worst case”; “It is emphasised that, at every point where a choice was available, the worst case input option was selected for simulation.” 1866

1222. The RPS Report also modelled effects in-combination with other salmon farms in the bay.1867 RPS asserted1868 that, overall, tidal flushing of Bantry Bay of roughly 27 billion tonnes of water per month – a

“huge volume of water” – dilutes and carries dispersing wastes out of the Bay and into the Atlantic in a slow,

counter-clockwise, circulation. RPS’s assumptions were not disputed – save for its modelling 12 and not 16+2

cages,1869 the choice and significance of its “Typical” figures and the use to which its “Typical” figures were

ultimately put – albeit not by RPS.

1223. The RPS Report1870 describes its model as dividing the 23,000 ha of Bantry Bay into a mesh of cells as small as 20m2 (about 4.5m x 4.5m) in the region of the Salmon Farm where finer detail was needed and as modelling the water quality in each cell in time-steps typically of less than 10 seconds and over model runs of up to 22 days (i.e. a full tidal cycle and > 190,000 time steps) depending on the specific quality parameter being considered. Inevitably, this produces vast amounts of data (> 8.5 billion data points1871) which will be incomprehensible unless rationalised for presentation purposes. So four types of graphical output – “Plume Envelopes” – were generated; Maximum Concentration Plume Envelope; Average Concentration Plume Envelope; Typical Ebb Concentration Plume Envelope. Typical Flood Concentration Plume Envelope. While plumes are produced for various parameters, we are here concerned with DIN.1872 The plumes represent DIN

1865 RPS Report p30. 1866 RPS Report p1, 27, 28, 30. Emphases added.

1867 RPS asserts “The in-combination effects were established using the alternating 24 month cycle for the production sites, operated by MHI”. This is a

reference to MOWI’s intended operation of its Bantry Bay sites on 2-year cycles with each site being harvested in alternate years, thereby permitting

annual production rather than bi-annual production. The ‘worst case’ month modelled was that in which the Shot Head site was dominant in its

production of DIN – the January of the 2nd year of its production cycle – and MOWI’s Roancarrig/Ahabeg site was on the alternate year in the cycle.

1868 Bass/RPS Presentation to Oral Hearing September 2017 p7 - 10. 1869 Addressed elsewhere in this judgment. 1870 p26 et seq. 1871 Bass/RPS Presentation to Oral Hearing September 2017 p6. Oral Hearing report p19. 1872 The list of figures relevant to DIN is as follows:

Figure 5.1: Maximum Plume Envelope of Nitrogen Concentration arising from the Shot Head site only. Figure 5.2: Average Nitrogen Concentration arising from arising from the Shot Head site only.

Figure 5.3: Maximum Plume Envelope of combined Nitrogen Concentration arising from all currently proposed and existing Bantry Bay sites – January,

year 2 – Shot Head / Fastnet dominant.

Figure 5.4: Average combined Nitrogen Concentration arising from all currently proposed and existing Bantry Bay sites – January, year 2 – Shot Head /

Fastnet dominant. Figure 5.5: Typical Ebb Plume of Nitrogen Concentration arising from the Shot Head site only.

Figure 5.6: Typical Flood Plume of Nitrogen Concentration arising from the Shot Head site only. Figure 5.7: Typical Ebb Plume of combined Nitrogen

Concentration arising from all currently proposed and existing Bantry Bay sites – January, year 2 – Shot Head / Fastnet dominant.

Figure 5.8: Typical Flood Plume of combined Nitrogen concentration arising from all currently proposed and existing Bantry Bay sites – January, year 2 –

Shot Head / Fastnet dominant. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

concentrations decreasing with distance from the Salmon Farm.1873 Importantly, the plumes depict dispersal only of DIN from salmon farms. They do not depict background DIN concentration deriving from other sources – though background DIN is included in total DIN calculations in order to assess compliance with DIN standards.

1224. I set out RPS Figure 5.1 below both to illustrate the general method and given the role it played in the case.

Figure 3 RPS Figure 5.1: Maximum Plume Envelope of Nitrogen Concentration arising from the Shot Head site only.

• The small black rectangle represents the Salmon Farm.

• Shot Head is immediately to its north west.

• The Dromagowlane/Trafrask river system flows into Trafrask Bay, the small inlet just north of Shot Head.

• Whiddy Island is to the east of the Salmon Farm and Bantry lies beyond it off the Figure.

• Adrigole Bay is north west of the Salmon Farm.

• Bere Island is seen west-south-west of the Salmon Farm – beyond which are the mouth of the Bay and the Atlantic.

• The maximum concentration of 0.04 to 0.07 mgDIN/l of (40 to 70 μgDIN/l – darker orange) is essentially at, and west, north and east around, the Salmon Farm.

1873 RPS explain at p26 the method of generating the Figures so as to render them visually comprehensible despite the large quantity of data informing

them and the low absolute values away from the Salmon Farm and also to render the Figures comparable with each other. In particular a logarithmic

scale was used in which each increment is approximately double that of the previous step and the same colour scheme is used in all Figures.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1225. RPS Figure 5.31874 shows a Maximum Plume Envelope for all proposed and existing Bantry Bay salmon farms with a view to addressing cumulative effects. Again, the maximum concentration of 0.04 to

0.07 mg/l of DIN (40 to 70 μgDIN/l – darker orange) is essentially at and west, north and east around the Salmon Farm. In this respect there is no discernible difference between Figures 5.1 and 5.3 – suggesting that, specifically as to maxima, in-combination effects are negligible.

1226. RPS stresses, as have MOWI and ALAB in these Proceedings, that these Maximum Concentration Plume Envelopes do not represent any single point in time. They plot the maximum DIN concentration modelled as reached in each cell at any time in the 22-day cycle. That maximum may be reached once only or repeatedly. If reached once only, values close to it may have been reached frequently or not at all. The maximum reached in one cell may occur – often will – in a different timestep to that in which another cell’s maximum occurs. However, and despite MOWI’s stressing that Maximum Concentration Plume Envelopes do not represent any single point in time, that is not a basis for dismissing them as unreal or uninformative. It is rather, a matter of being clear as to what they do convey. Every one of the maximum values they plot, for each cell, does occur in the 22 days – they just don't all occur at the same time. The central message for present purposes is that in every cell in the area coloured dark orange in Figures 5.1 and 5.3, DIN will reach the range of 40 to 70 μgDIN/l at least once in the 22-day cycle and possibly more often. RPS says that “there is little likelihood of any of the maximum values recorded occurring simultaneously.”1875 Intuitively that seems unlikely – not least as to adjacent and small cells. Elsewhere, RPS puts the matter quite differently and, it seems to me as a probability, more accurately: “the concentrations shown in the maximum plots ….. would not all occur simultaneously.”1876

1227. Likewise, the RPS Average Concentration Plume Envelopes do not represent any single point in time. They show the average concentration in each cell during the 22 days.1877 Indeed they are means – so the precise value depicted as to a given cell may never have occurred at all. The highest average DIN level areas in Figures 5.21878 and 5.41879 roughly correspond to the same area around the Salmon Farm and both are in the (light orange) range of 20 to 40 μgDIN/l.

1228. Strikingly, given the EPA deploys medians in calculating ecological status, RPS calculates, depicts and deploys no medians at all.

1874

Maximum Plume Envelope of combined Nitrogen Concentration arising from all currently proposed and existing Bantry Bay sites – January, year 2 –

Shot Head / Fastnet dominant 1875 RPS Report §5.1.2. Emphasis added. 1876 RPS Report §5.2.1. Emphasis added.

1877 This was generated by averaging all the values recorded in all ≤10 second time steps in each cell over the course of the simulation.

1878 Figure 5.2: Average Nitrogen Concentration arising from arising from the Shot Head site only.

1879 Figure 5.4: Average combined Nitrogen Concentration arising from all currently proposed and existing Bantry Bay sites – January, year 2 – Shot Head

/ Fastnet dominant. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1229. RPS states that the average plume envelopes give “an indication of how frequently the maximum values given could occur”. This somewhat vague observation is in fact underlain by the general idea that the closer the averages are to the maxima, the greater the likely frequency of the maxima or values close to them and, conversely, the further the averages are from the maxima, the lesser the likely frequency of the maxima or values close to them.1880 But interpretation of specific data to discern the specifics of that relationship is a matter for experts. And, of course, frequency of occurrence of maxima may – likely does – vary as between cells. RPS does say that “the percentage frequency of occurrence (of the maximum in each cell) will be low due to tidal oscillation”.

1230. While “low” is not quantified, later in the report RPS says that Figures 5.1 to 5.4 “all show a disparity

between the maximum and average plots of around one to two contour levels in each case, which equates to maxima being two to four times the average values. …. this suggests that the maximum values given in the Maximum Concentration Plume Envelope plots do frequently occur or persist for long periods.” 1881 It was urged that this represents a typo and should read “do not frequently” occur. While the difference the additional word would make is striking and such typos should in theory not occur in properly proofed formal reports, the fact is that they do and probably always will. A judgment of the present length likely contains at least some. To correct typos where the true meaning is clear is a legitimate exercise in interpretation of a document and should not, per se, be regarded as criticism of the document. And the mistaken omission of

“not” is by no means unprecedented. The word “do” in the phrase is slightly clumsy and is unnecessary to its

literal sense. While “do” could have been inserted for emphasis, that seems unlikely given the overall thrust of the report. In the literal sense, the phrase could as easily, and perhaps more naturally, have read: “this suggests that the maximum values given in the Maximum Concentration Plume Envelope plots frequently occur or persist for long periods.” My task is to interpret as a matter of fact the probable meaning of the report. Given the overall thrust of the report, the slight awkwardness of the word “do” in the phrase in question and, most particularly, the earlier assertion that “the percentage frequency of occurrence (of the maximum in each cell) will be low due to tidal oscillation”, I infer that a typo has indeed occurred by way of omission of the word “not”. I take the true meaning of the phrase to be that comparison of the maxima and averages “suggests that the maximum values given in the Maximum Concentration Plume Envelope plots do not frequently occur or persist for long periods.”

1231. However, the quantification and wider significance of that observation are not self-evident. That is so not least given that Weser and Association France Nature hold that:

• the threshold for breach of the obligation to prevent status deterioration must be as low as possible.

• temporary, transitory or short-term impacts, o without lasting consequences can be ignored where it is established that they have little effect on water body status and so can’t lead to a deterioration of status. o which cause even a temporary status deterioration require refusal of development consent.

1880 This is a matter of fairly simple maths and logic but was in any event confirmed by Dr O’Toole – Day 5 p71.

1881 RPS Report p31.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1232. The Typical Ebb and Typical Flood Concentration Plume Envelopes are tendered, the RPS Report says, “to give an indication of the actual1882 dispersion pattern within the Bay for each parameter, the typical flood and ebb contour plots have also been included. These are ‘snapshots’ from the model for a typical mid-flood or mid-ebb tide situation. Unlike the previous plots, these values can be related to real moments in time.”1883

These observations by RPS must – to use a current colloquialism – be “unpacked”.

• First they are merely an “indication”. That is not per se a criticism – it is a matter of knowing what is claimed for them.

• Next, it is important to say that all the Plume Envelopes, including the Typical plumes, are unreal in the sense that they depict modelled, not measured, data. Again that is not a criticism – modelling is the best predictive tool to hand. But in this respect the Typical plumes are no more “actual” than the others.

• That the Typical plumes represent a particular and single point in time across the entire surveyed area may be, but is not necessarily, a virtue. It may render them more or less useful, “actual” or “real”

depending on how those particular and single points in space and time have been chosen for depiction as “Typical” and the uses to which they are put and the questions which they are set to answer.

• As “snapshots”, the Typical plumes consist of depictions of two single 10-second timesteps out of a total of >90,000 such time steps. They represent a choice by RPS of those two timesteps as “typical” of the >190,000 such time steps over 22 days. To identify two of 190,000 datasets over 22 days as “typical” is a large claim.

• Other than that they are mid-ebb and mid-flood snapshots, we are not told why those snapshots were chosen and why they were deemed “typical”.

• Even the choice of mid-ebb and mid-flood snapshots – as opposed, for example, to slack water snapshots

– is not explained.

• In short, beyond the observation that they represent two of 190,000 “actual” modelled points in time, it is not clear what is the significance of these Typical plumes or what weight can be given to the word “actual”

in the text of the RPS Report.

1882 Emphasis in original 1883 Emphases added save where indicated.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1233. As stated, and notably, RPS do not reveal in their Report, and Dr Saunders does not infer from the RPS Report (or does not state his inference), what criteria were applied by RPS in deeming particular points in space and time to be “Typical” for depiction. We are told1884 that the Average Concentration Plume Envelope was “useful when used in conjunction with the maximum plume envelope for gauging the ‘typical’ values”. But that is not very informative.

1234. Equally notably and as will be seen, the experts Dr Bass and Dr O’Toole who were cross-examined, were unable to infer those criteria from the RPS Report.

1235. I should say that the foregoing observations as to Typical plumes are not in the end a criticism of the RPS Report given the approach RPS took to drawing conclusions. However they become significant in considering how Dr Saunders approached the matter in his final report which formed the basis of ALAB’s

decision on this issue.

1236. Crucial in the RPS report is their assumption of a background “average winter” (mean) concentration of circa 100 µg DIN/l such that a DIN discharge by the Salmon Farm “resulting in an increase of less than 70 µg/l would be acceptable”.1885 This observation is clearly made on the basis that “The EQS comprises the sum of the concentrations resulting from the discharge … plus the existing ambient nitrogen concentration.” 1886

and on the basis that the “EQS” is the “quality standard” of 170 μg/ml set by the Surface Waters Regulations.

But this observation and what follows are puzzling:

• RPS do not justify their use of an “average winter” (it seems December to March1887) background DIN as opposed to, for example a maximum and/or January DIN levels given Year 2 January is the month in the production cycle modelled as part of the worst-case scenario.

• RPS say, inter alia, that the Maximum Concentration Plume in Figure 5.1 of DIN concentrations arising from the Shot Head site only show “levels below 10 µg/l (0.01mg/l) and are typically less than a tenth of this value.” While a similar remark as to Figure 5.3 is unquantified and contains another typo,1888 as already noted the maxima are the same in both Figures 5.1 and 5.3. However that is not what is puzzling.

The assertion that the maxima in Figures 5.1 and 5.3 are “below 10 µg/l (0.01mg/l)” does not correlate to the legends to those Figures, which clearly show maxima around the Salmon Farm up to 0.07mg/l or

70 µg/l.

1884 RPS Report §5.1.2. 1885 RPS Report p31. 1886 RPS Report p30. 1887 sEIS 2018 p96.

1888 It says “Similarly, Figure 5.2 and Figure 5.3 show Maximum and Average Concentration plumes from the combined sites in the combined peak

discharge month of January of alternate years”. Clearly his should refer to “Figure 5.3 and Figure 5.4”. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• More importantly, and though this passage of the RPS report also refers to the averages depicted in Figures 5.2 and 5.4,1889 (which are up to 40 µg/l), the premise of this passage is to reassure that even adding the maxima produced by the fish farm to the average ambient DIN level, the “EQS” of 170 μgDIN/ml is not exceeded.

1237. So, and significantly RPS, in interrogating the question of acceptability of DIN, invoked1890 not the

“Typical” Figures 5.5 to 5.8 but the “Maximum” Figures 5.1 and 5.3.1891 Correcting their error in reciting the

actual maxima in those figures, it becomes clear that RPS’s assertion of the acceptability of the discharge of DIN from the Salmon Farm, by reference to the High WFD status of the waters of Outer Bantry Bay, is in substance an assertion that if one adds the maxima of 40 – 70 µgDIN/l to the ambient 100 µgDIN/l, getting

140 – 170 μgDIN/l, the quality standard of 170 μgDIN/l set in the Surface Waters Regulations is not breached. Notably, given those maxima are centred on the locus of the Salmon Farm, this is an analysis centred on that locus.

1238. On that interpretation and to this point, on its face, being based on the maxima (and ignoring the issue of the use of average background levels) the RPS analysis sufficed both as a worst case scenario and as confirmation of no deterioration of the High WFD status of Bantry Bay. While RPS did not analyse the locality of the EPA’s representative Monitoring Points, clearly if the quality standard of 170 μgDIN/l was not exceeded in the immediate vicinity of the Salmon Farm it would not be exceeded at those monitoring points whether near or far from the Salmon Farm. This approach obviated a bay-wide approach as it was a worstcase approach as to the locality of the measurement in that it asserted compliance with the 170 μgDIN/ml standard local to the Salmon Farm. Given what ensued, it is important to note that such compliance depended on an ambient 100 µgDIN/l – any higher and compliance would not be apparent on this approach.

1239. However, given what ensued, I must say some more of the “Typical” Figures. It is not clear to me exactly what role the Typical Figures played in the 2015 RPS Report analysis of compliance with the WFD obligation to prevent deterioration. In the end, I think they played little or none. While RPS said1892 they contribute to a view that “under all circumstances, dispersal of nitrogen from the sites has a limited impact on dissolved nitrogen levels in Bantry Bay as a whole”, as I hope I have demonstrated, RPS’s deployment of the maxima obviated, on their figures, the need for a bay-wide approach. The highest level areas in the Typical Plume Figures 5.51893 and 5.61894 differ a little more in their location from the other Figures – perhaps due to tidal flow given one is ebb and the other flood. However, it seems notable that the levels in those areas are the same as the highest averages in Figures 5.3 and 5.4 – all are in the (light orange) range of 20 -

40 µgDIN/l. While this apparent correlation should be for the experts to interpret, it is notable that neither

1889 Mistakenly referred to as 5.3. 1890 RPS Report p31. 1891 Also the average figures but that is not what matters here. 1892 RPS Report §5.2.2. 1893 Figure 5.5: Typical Ebb Plume of Nitrogen Concentration arising from the Shot Head site only. 1894 Figure 5.6: Typical Flood Plume of Nitrogen Concentration arising from the Shot Head site only.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Dr Bass nor Dr O’Toole could discern or infer the basis or criteria on which RPS had chosen their 2 “Typical” snapshots in time from the 190,000+ snapshots available. If Dr Saunders did infer such basis or criteria, he

did not say so and neither did RPS. It is clear that the “Typical” levels approximate to the “Average” levels.

But as I have said, the averages are means – not medians as used by the EPA.

1240. It is difficult - certainly difficult absent explanation - to see that “Typical” data – the Typical Plume

Figures 5.5 and 5.6 – can form part of a “worst-case scenario”.1895 It seems to me that this is revealed in RPS’s oxymoronic description of them1896 as based on a “high level of worst case expectation”. The word “worst” has a particular meaning and it is not merely “very bad”. I should emphasise that I do not say that concepts of reasonableness or likelihood should necessarily not be incorporated into analyses or that a highly pedantic view should be taken of the word “worst”. My point is that whatever standard is adopted should be clear and transparent. Even the experts who testified in defence of the Impugned Decision were

unable to illuminate the basis of RPS’s choice of “Typical” scenarios. To be fair to RPS, the lack of clarity on

that issue may reflect a view on their part of the Typical scenarios as merely supplementary to their analysis based on the maxima. And, as I say, on that latter basis, RPS’s view is internally coherent. But it was not the view taken by Dr Saunders or by ALAB.

WFD – Technical Advisors’ Interim Report (Saunders) 2016

1241. There was criticism of ALAB that Dr Saunders swore no affidavit. I understand he is no longer available to ALAB. That does not seem to me a basis for criticism of ALAB but any consequences of his absence must lie where they fall. His reports are in evidence before me and properly so as part of the basis on which ALAB made its Impugned Decision.

1242. Dr Saunders1897 very briefly cited the WFD and the Surface Waters Regulations in asserting that “EQSs define safe limits of nutrients or chemicals in the marine environment.” As will have been seen, I have been unable to satisfy myself that, as to the specific term “EQS”, that is correct as to nutrients. Dr Saunders describes1898 the RPS Report 2015 for a worst case scenario DIN analysis. However, though it is not express in his report, it seems to me that Dr Saunders departs in two significant ways from the RPS Report 2015.

• First, he adopts a “maximum ambient N concentration” of 125 µgDIN/l as compared to the RPS figure of

100 µg/l. Some involved considered that too stringent1899 but that is not the point. All parties accept, as I

1895 I have not ignored Figure 5.7: Typical Ebb Plume of combined Nitrogen Concentration arising from all currently proposed and existing Bantry Bay sites

– January, year 2 – Shot Head / Fastnet dominant and Figure 5.8: Typical Flood Plume of combined Nitrogen concentration arising from all currently

proposed and existing Bantry Bay sites – January, year 2 – Shot Head / Fastnet dominant. But they do not seem to me to add to the analysis

1896 RPS Report 5.2.2. 1897 p70. 1898 p72.

1899 For example as the figure relates to samples recorded by RPS as taken from a boatyard in Berehaven – a different water body.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

understand it, that Dr Saunders adopted the figure of 125 µgDIN/l for ambient levels as the basis of his analysis and so that it formed the basis of ALAB’s decision. However, it must be properly understood. It is not a maximum in the full sense of that word. It is derived from Table 5.2 of the RPS Report. So, it should properly be described as the “Maximum Monthly Mean Ambient Concentration”. Given the EPA’s methodology, it seems notable that this figure is a mean not a median. But, primarily, its significance is that Dr Saunders increased the ambient DIN figure from RPS’s 100 µgDIN/l to 125 µgDIN/l.

• Second, instead of adding the Salmon Farm maxima, the premise of which RPS had in effect accepted, he added the Salmon Farm Typical figures such that, he said, “The worst-case scenario adds a maximum of 40

μg/l within 1 km of the farm site.” As Dr O’Toole said, Dr Saunders “decided to go with the typical graphs”

– but, I observe, Dr Saunders does not tell us why. Thus, instead of adding 40-70 µgDIN/l (which would have resulted in a total of 195 µg/l and exceeded the “EQS” of <170 µgDIN/l), he added 20-40 µg/l which resulted in a compliant total of 165 µgDIN/l “within the site boundary”. He says, “This represents the maximum localised nitrogen release”1900 and that “as these nutrient increases are predicted to remain below the .. EQS for nitrogen (170 μg/l), any rise in primary production within the site is not expected to incur significant environmental effects”.1901

Notable here also is the continued acceptance by Dr Saunders of RPS’s “worst case” approach of assessing compliance with the “EQS” of <170 µgDIN/l by reference to the “maximum” DIN concentration due to and “localised” to the Shot Head Salmon farm.

1243. Dr Saunders also states that,

“The results of the comprehensive modelling conducted by RPS indicate that, even under worst-case scenarios covering a 22 day period, nitrogen and phosphorous do not exceed EQSs in close proximity to the proposed farm site (1 km) and are rapidly dispersed to ambient levels. Whilst we fully accept that modelling is a predictive tool with inherent limitations, the model used is amongst the best currently available and has been successfully used for a range of other civil and governmental applications. We would also accept that this analysis has incorporated a considerably conservative margin through their selection of “worst case” options. It is therefore concluded that nitrogen and phosphorous from the proposed fish farm site will not constitute a significant additional nutrient burden to the Bay, will not stimulate algal blooms or enhance naturally occurring blooms and therefore presents no risk to wild or cultivated shellfish.”1902

1244. However, this conclusion depends on his conclusion that “even under worst-case scenarios .. nitrogen .. do not exceed EQSs in close proximity to the proposed farm site.” This in turn depends on his second departure from the RPS Report 2015 described above. Dr Saunders does not state any reasons for his

1900 p72. Emphasis added 1901 p47. 1902 p73.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

departure from the RPS approach or how deployment of the Typical figures, as opposed to the maxima, yields “the maximum localised nitrogen release”.

1245. Dr Saunders lists1903 as part of the “case for granting the licence on the basis of no significant effects on the environment”, his view that “Nutrient discharges will not stimulate algal blooms or enhance naturally occurring bloom and therefore presents no risk to wild or cultivated shellfish”. But he does not

• express his conclusion in terms of the WFD or its Article 4 obligation to prevent ecological status deterioration (which he does not mention).

• explain, justify or even acknowledge his departure from the RPS Report 2015 by using the Typical figures rather than that maxima.

• explain how typical figures can form part of a worst case scenario.

WFD – Bass/RPS presentation to Oral Hearing – September 2017

1246. The next significant document on this issue is the Bass/RPS presentation to the Oral Hearing dated September 2017. In February 2017 it became apparent at the oral hearing that the RPS Report 2015 had not been circulated. The oral hearing was adjourned to allow its circulation and it sat again in September 2017. The Bass/RPS presentation is based on the RPS modelling and report of 2015 and it may be unfair to parse it too closely as I imagine it was intended both as supplementary to that report and as a document to be spoken to in oral evidence.

1247. Nonetheless, as to water quality, it notably cites neither the WFD nor the Water Quality Regulations. The understanding of the concept of deterioration of status by then to hand from the Weser case and its clarification that breach of the Article 4 WFD obligation to prevent deterioration required refusal of development consent are not mentioned in the presentation as the applicable legal standards.

1248. It is clear that the Bass/RPS presentation reverted to the Scottish SEPA EQS of 168 µgDIN/l rather than the 170 µgDIN/l from the Irish Water Quality Regulations – while the quantified difference is irrelevant, the preference is puzzling. More notably, the presentation asserts that “Always, only worst case discharge conditions are modelled”.1904 Beside this sentence is reproduced Figure 5.8 of the RPS Report 20151905 – but it is now re-named “Typical worst case1906 Flood Plume of combined Nitrogen concentration ……..” – the words “worst case” being added but the addition not being explained. As I have said, it is difficult to see that

1903 p91. 1904 p10 – Emphases in original.

1905 “Figure 5.8 Typical Flood Plume of combined Nitrogen concentration arising from all currently proposed and existing Bantry Bay sites – January, year

2 – Shot Head / Fastnet dominant”. 1906 My emphasis.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“Typical” data can form part of a “worst-case scenario” and RPS and Dr Bass did not here explain how it can.

Absent such explanation, the phrase “Typical worst case” seems, prima facie, oxymoronic.

1249. The important calculation follows. To the “peak winter ambient DIN of say 100 µgN/l” is added a discharge from the Salmon Farm “(peak at site centre)” of 40.0 µgDIN/l yielding a total of 140.0 µgN/l. Clearly, though they do not say so or explain why they have done so, RPS and Dr Bass,

• insist on ambient DIN of 100 µgDIN/l – not the 125 µgDIN/l preferred by Dr Saunders in his interim report.

• perhaps taking their cue from Dr Saunders’ interim report, depart from the RPS Report 2015 in that they o no longer reassure the reader that, on adding the maxima of up to 70 µgDIN/l to the ambient DIN level the EQS is not exceeded “even at site centre”.

o instead of 70 µgDIN/l now add the “Typical” value of up to 40 µgDIN/l to the ambient DIN level for a total of 140 µgDIN/l “even at site centre (very worst case)”.– such that the EQS is undershot by an even greater margin than RPS had advised in 2015.

1250. Of course, RPS and Dr Bass were entitled to reconsider and change RPS’s views of 2015. But one would have expected them to explain and justify those changes. They did not. And, given they have used typical figures not maxima, it is difficult to understand, and they do not explain, their assertion that they have depicted the “(very worst case)” or “Always, only worst case”. In cross-examination, Dr Bass was

unable to provide a rationale for, or to discern RPS’s rationale for, including “Typical” figures in a worst case

scenario.

WFD – Oral Hearing & Report thereon

1251. It bears observing here that, at the oral hearing, counsel for An Taisce submitted as to the WFD that, as a matter separate from EQS compliance, MOWI’s application had not addressed the issue of deterioration of status1907 and had specifically drawn ALAB’s attention to the Weser case1908 and the obligation stated therein to refuse development consent where the project may cause a deterioration of water body status.

1252. The Oral Hearing Report briefly referred to that citation of Weser as requiring refusal of development consent where the project “would result” in waters falling below “good water status”. But it did not mention the more general Article 4 WFD obligation to prevent deterioration as elucidated in Weser.1909 That obligation not merely prohibits a fall from “good water status”: it also prohibits a fall to

1907 Oral Hearing report p18.

1908 Case C-461/13, Bund für Umwelt und Naturschutz Deutschland eV v Bundesrepublik Deutschland. See Oral Hearing Transcript 15/2/19 10:28.

1909 Referring only to an obligation to refuse authorisation for any project which would result in water standards falling below “good water status” as

defined under the WFD. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“good water status” from “high water status”. This omission is echoed in the Oral Hearing Report’s recommendation merely of:

“An assessment of the potential impact of salmon farm waste on water quality, having particular regard to the maintenance of ‘good water status’ as required under the WFD”

WFD – sEIS 2018

1253. At the instance of the Oral Hearing Report and by s.47 request dated 20 December 2017, ALAB required MOWI to submit an sEIS addressing, inter alia, “The impact of salmon farm waste on water quality in Bantry Bay, having particular regard to the maintenance of 'good water status' as required under the Water Framework Directive”. This question was misconceived: first because the status of Outer Bantry Bay is high not good; second because, as Weser had held in 2016, the true issue for ALAB on the facts before it was deterioration of status – including from High Ecological Status to Good – as the concept of deterioration was explained in Weser.

1254. The sEIS 2018 prepared by Dr Bass responds by correcting the first error: it states that “The question to be answered in this section is therefore whether High Ecological Status will be maintained in Outer Bantry

Bay, once the Shot Head site is fully operational, if the licence is upheld”.1910 But that did not correct ALAB’s

second error and it set the legal bar too low. It in effect applied the ‘the status class theory’, rejected in Weser,1911 that deterioration occurs only when effects would relegate the water body to a lower class (for example from high to good). Weser made clear that ‘deterioration of the status’ of a surface water body is a concept of general scope as opposed to one related to the classification of status and it includes deterioration which does not result in re-classification of the water body to a lower class. Deterioration occurs as soon as the status of one quality element,1912 falls by one class, even if that fall does not result in a fall in classification of the surface water body as a whole.

1255. At cost of some repetition of earlier analysis in this judgment, I record that the sEIS analysis1913 starts with the puzzling question “EQS or WFD?” – as if they were alternatives, when the concept of EQS was defined in and is a considerable part of the WFD.1914 In that context, the citation in the sEIS of the EQSD (which post-dated the WFD by 8 years) as the source of EQSs seems to contradistinguish it from the WFD – when in fact that EQSD is specific to water policy and explicitly sets EQSs in accordance with the WFD.1915 This impression of contradistinction is reinforced by the deployment of an OECD definition of EQS in

1910 p7 & p93. 1911 As I have explained above. 1912 Within the meaning of Annex V to the Directive. 1913 p90 et seq. 1914 E.g. Article 7.2 1915 EQSD Recital 5.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

apparent preference for the different definition set out in the WFD.1916 The sEIS asserts that “many nutrient and physicochemical Quality Elements required for the derivation Ecological Status depend on EQS values or something very close to them.”1917 It is not clear what is meant by “or something very close to them.” The sEIS asserts that “EQS is as valid now as it ever was ………. Nonetheless, WFD methodologies have greatly expanded the scope of ecological assessment …” – again the impression is given that EQSs are somehow separate or to be distinguished from the WFD. The conclusion is that the s.47 request “entails the conversion of earlier findings, expressed relative to EQS limits, and their re-evaluation, relative to the current Ecological Status of Bantry Bay, under the terms of the WFD and SI 272 2009”. This seems to be an acceptance, without explanation, that the EIS in 2011 had not had regard to the WFD or the 2009 Surface Waters Regulations but had assessed water quality via EQSs generated external to the WFD regime and, it might be inferred but is not clear, generated via an EQSD process external to the WFD regime (when the EQSD is itself a creature of the WFD regime).

1256. It seems to me accordingly that, as to an understanding of the applicable legal regime, ALAB’s s.47 request and the sEIS got off to a confused and confusing start.

1257. This supposed distinction “EQS or WFD?” is relevant as the sEIS 2018 states that DIN “results are then compared with established Environmental Quality Standards (EQS) for DIN in coastal waters, to establish where the resulting elevated ambient concentration lies relative to the EQS level.” It is next stated that DIN is a parameter used to “derive Quality Elements (QE) in Coastal Waters” under the Surface Waters

Regulations 2009.1918 It is not clear what the word “derive” means here. In the WFD,1919 a “Quality Element”

is a parameter which is identified in the WFD as contributing to water status. The WFD Annex V list of “Quality elements for the classification of ecological status” includes “Chemical and physico-chemical elements supporting the biological elements – General” which in turn includes “Nutrient conditions”. DIN is not mentioned in the WFD but there is no doubt but that is a component of nutrient conditions.

1258. I have recorded above information on the EPA’s WFD practice which came to hand during the trial. The information in the sEIS on the EPA’s WFD practice is, it seems to me surprisingly, recorded as obtained by personal communication from the EPA’s Scientific Officer rather than from any published materials recording highly regulated WFD practices approved at EU level. As has been noted, a general paucity of information on the EPA’s WFD practice persisted to trial and, in my view, still persists. That is not to impugn

the EPA’s practice or to suggest the information does not exist – it is just to say that it is not in evidence in

this case, either directly or via the documents before ALAB in making its Impugned Decision.

1916 Article 2.35 WFD. 1917 Sic. 1918 Misidentified as “SI 272 2007” but the typo is clear. 1919 Notable in Annex 5.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1259. The sEIS1920 describes the EPA as pooling all monitored data from all monitoring stations and depths on Outer Bantry Bay. It says that for nutrients only winter values1921 are assessed, since this is when they are most mineralised. (In contrast, the EPA letter to the court asserted analysis of both winter and summer DIN monitoring data – though, of course, practice may have changed in the interim). The sEIS states that for DIN, “median winter values are checked against the Quality Element data for DIN, set out in Schedule 5, Table 9,

Part A of SI 272”. I do not suggest error but am not told, nor have I found, the legal basis for using medians –

as opposed, for example, to means. As noted earlier, for some quality elements the Surface Waters Regulations 2009 do specify medians – but not for DIN. And the description of Table 9 as including “data” for DIN is inaccurate – Table 9 contains standards (however they are to be properly named) not data. For all that, the sEIS description of the EPA system is clear that a median is drawn from the pooled winter DIN readings for all representative monitoring points on Outer Bantry Bay and that median is compared to the quantitative standard for DIN, in Table 9, of ≤170 µgDIN/l. However, as observed earlier, the RPS Report 2015, from which all data as to DIN informing ALAB’s Impugned Decision is drawn, contains no medians.

1260. The sEIS then introduces the concept of mixing zones as derived from “Article 19” EQSD (there is no Article 19 EQSD – the relevant article is Article 4) and Article 51 of the Surface Waters Regulations 2009. The sEIS states that “mixing zones are applied (in regulation), for solids and medication EQS standards. It is submitted that the same view should be taken with other parameters, for example for DIN.” Given Article 4.1 EQSD limits mixing zones to exceedances for substances listed in Annex I Part A and as Part A, which sets EQSs for the substances listed but does not list Nutrients, Nitrogen or DIN, it is not clear on what legal basis this submission is made to ALAB. The sEIS records the EPA’s position, as stated in the personal communication of its Scientific Officer, that “For our assessments (EPA doesn’t) have a formal consideration of mixing zones. (EPA doesn’t) sample directly beside known discharges and for an assessment of a waterbody we pool all the available data together”. This perhaps again demonstrates that it would have been desirable to have had regard to any public documents of the EPA describing their monitoring regime and method of selection of representative monitoring points. I mention the issue of mixing zones primarily as it caused some confusion in the proceedings and also as it points up the question whether, by the WFD and/or the EQSD, the concept of EQS properly applies to DIN at all. In any event, and for reasons given above, as ALAB did not adopt mixing zones I need consider this issue no further.

1261. The supposed contradistinction posited in the sEIS between EQS and WFD is repeated in the

assertion1922 as to DIN that “Its EQS is a winter limit value of 168μg/l.”, whereas “In SI 272”1923 the winter DIN

“limit value”1924 for High Ecological Status is 170 μg DIN/l. The value of 168μg DIN/l though not credited, is

clearly the Scottish SEPA EQS. The sEIS at this point addresses the issue of ambient DIN contested at the Oral Hearing and invokes the EPA’s method of pooling all data to find a median ambient of 115.2μg DIN/l.

1920 p96. 1921 December to March – sEIS p97. 1922 3.3.4. Dissolved Inorganic Nitrogen (DIN). 1923 i.e. the Surface Water Regulations 2009 – Citing Table 9. 1924 Yet further nomenclature.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1262. It will be seen that this is slightly higher than the ambient level of 100 μg DIN/l assumed by RPS in 2015 and in the Bass/RPS Presentation of September 2017. Using RPS’s 2015 methodology, adding the RPS Salmon Farm maxima of 40 - 70 µgDIN/l1925 would yield a total DIN concentration of 155.2 – 185.2 µgDIN/l. That exceeds, at its upper end, the 170 μgDIN/l set by Table 9 of the Surface Waters Regulations.

1263. However, rather than using RPS’s 2015 methodology, the sEIS uses the methodology prefigured in the September 2017 Bass/RPS presentation to the oral hearing. It does not add the median ambient level of 115.2μgDIN/l to the RPS Salmon Farm maxima of 40 - 70 µgDIN/l.1926 Rather it adds the RPS Salmon Farm Typical levels of 20 – 40 µgDIN/l1927 to yield a maximum total of 155.2μgDIN/l such that “the elevation of ambient DIN to 0.1552 DIN/l close to the site ….. is well within the set QE standard for High Ecological Status, on a worst case basis”.1928 While that claim is also made for more distant waters, it is clear that the sEIS set itself the task, as had the earlier RPS and Saunders analyses, of demonstrating compliance in the immediate locale of the Salmon Farm. It is said that “More than anything else, this demonstrates that DIN dispersing from the Shot Head site at worst case will not elevate ambient DIN to the extent that any environmental disturbance, …. will result and High Ecological Status will be maintained.” However, and again, no explanation is given of why one would add a “typical” “snapshot” value to a median value. And RPS, who chose the Typical values, had not generated any medians.

1264. Again, as in the September 2017 RPS/Bass presentation, Figure 5.8 of the RPS Report 20151929 is reproduced in the sEIS as Figure 3.31930 – but it again re-named “Typical worst case1931 Flood Plume of combined Nitrogen concentration ……..” – the words “worst case” being added. The sEIS describes this

“Typical DIN plume plot on flood tide” as “Selected from a range of DIN dispersal plots available in the” RPS

Report 2015. The sEIS does not explain

• on what basis 2 timesteps were selected from the 190,000 available and deemed typical.

• of what these 2 timesteps are alleged to be typical.

• why it was appropriate to use typical values in the calculation or

• how a calculation based on typical values constitutes one done “on a worst case basis”. It is unclear on what basis in logic or validity the sEIS asserts that “More than anything else, this demonstrates that DIN dispersing from the Shot Head site at worst case will not elevate ambient DIN to the extent that any environmental disturbance such as elevated primary production, will result and High Ecological Status will be maintained.”

1925 RPS 2015 Figure 5.1. 1926 RPS 2015 Figure 5.1. 1927 RPS 2015 Figure 5.8. 1928 sEIS p98 – §3.3.4. Dissolved Inorganic Nitrogen (DIN).

1929 “Figure 5.8 Typical Flood Plume of combined Nitrogen concentration arising from all currently proposed and existing Bantry Bay sites – January, year

2 – Shot Head / Fastnet dominant” 1930 p99. 1931 My emphasis.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

WFD – Technical Advisors’ Final Report (Saunders) 2020

1265. As to DIN, Dr Saunders’ final report1932 repeats almost verbatim the text of his interim report. He again prefers the ambient figure of a “natural maximum in January of around 125μgDIN/l” – a preference accepted and adopted by ALAB and by all at trial – including Dr Bass and Dr O’Toole. As I have observed earlier, this is in fact a maximum mean not a maximum in the full sense and also is not a median as medians are used by the EPA. He again deems the RPS “typical” 40 μgDIN/l to represent a “maximum” and a “worstcase scenario” and adds it to the 125μgDIN/l to produce a “maximum of 165 μg/l within 1 km of the farm site. This represents the maximum localised nitrogen release. It is modelled from a worst-case ….”1933 He

nowhere explains why the “typical” 40 μgDIN/l represents a “maximum” and a “worst-case scenario” – and,

of course, in its 2015 Report at least, RPS had not so described it.

1266. Dr Saunders again accepts1934 that modelling is a predictive tool with inherent limitations but notes that the model used is amongst the best available. This is not disputed – though, as I observed in my earlier judgment, it is a necessary but, by itself, insufficient reassurance. He also accepts that the RPS analysis incorporated a considerably conservative margin by selection of “worst case” options. Again, that many model assumptions were worst case is not disputed. The question is whether, as Dr Bass and RPS claimed in their 2017 presentation, and as is in any event implicit in the word “worst”, the assumptions were of the “very worst case” and “Always, only worst case”.

1267. As to the WFD/Surface Waters Regulations regime, Dr Saunders asserts that “EQSs define safe limits of nutrients or chemicals in the marine environment”. As stated, I have been unable to satisfy myself of the use of the term “EQS” as to nutrients.

1268. Dr Saunders concludes1935 that RPS’s comprehensive modelling indicates that, even under a 22-day worst-case scenario DIN does not exceed “EQSs” at and within 1km of the fish farm site and is rapidly dispersed to ambient levels. Again, Dr Saunders does not mention the obligation to prevent deterioration as elucidated in the Weser case. I accept that, ceteris paribus, in a water body already classified as high ecological status “ambient levels” imply no deterioration. However, that observation excludes effect local to the salmon farm site and is also based on representing the fish farm DIN contribution to water quality by way of the RPS “Typical” data.

1932 At p53 and p78 et seq. 1933 p80. 1934 p81. 1935 p81.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

WFD – ALAB Determination

1269. As to EIA, ALAB’s determination1936 records its having required the sEIS, inter alia, “to assess the potential impact of Salmon Farm waste on water quality, having particular regard to the maintenance of 'good water status' as required under the Water Framework Directive.” It does not mention the other fundamental Article 4 WFD obligation – to avoid deterioration from high status. However, in its determination for purposes of s.61 of the 1997 Act1937 ALAB states that it:

• accepts the findings of Dr Saunders’ Final Report and the RPS Water Modelling Report 2015 as to nutrient releases (i.e. including DIN).

• notes that Dr Saunders’ Final Report and the RPS Report state that nutrient releases will not breach the EQS.

• notes the high WFD status of Outer Bantry Bay.

• notes that the RPS water modelling indicates that the impact “will not negatively affect Outer Bantry Bay's … current classification under the Water Framework Directive”.

• “considered and accepted this outcome”.1938

1270. I observe that ALAB:

• does not give reasons for this decision as to nutrients and WFD classification. Of itself, that is not a difficulty, as it clearly adopts those found in Dr Saunders’ Final Report and the RPS Report 2015.

• does not address, prefer or give reasons for a preference as to the difference in approach as between the RPS Report 2015 which based its reassurance on its modelled data for the maximum DIN effect of the fish farm and Dr Saunders’ Final Report which deploys RPS’s “typical” data to that end.

• specifically, does not give reasons for the deployment of RPS’s “typical” data in the asserted “worst case

scenario” analysis.

WFD – DIN Issue – Affidavits & Cross-Examination of Dr Bass and Dr O’Toole - #2

WFD – DIN Issue – Affidavits & Cross-Examination – Introduction & judgment of 16 March 2023

1271. SWI CG9a asserts that ALAB, by the Aquaculture Licence and contrary to Article 4 WFD, allowed a discharge to waters by the operation of the proposed Salmon Farm that would breach “emission limits” for nitrogen. I gave liberty by judgment of 16 March 2023,1939 in the SWI proceedings only, to cross-examine as to that allegation. In doing so, I raised issues as to the evidence, which I revisit below. I refer to my doing so as that judgment, I presume, informed the cross-examination of the deponents in at least some degree and their preparation for that cross-examination. As stated earlier, 3 deponents were cross-examined: John

1936 §4.2 1937 §6.5.11, 6.6.2 & 6.6.3. I have changed the order of the conclusions for clarity.

1938 I have omitted ALAB’s reference substantive risk to shellfish and finfish as beside the point as to WFD compliance.

1939 Salmon Watch v ALAB [2023] IEHC 129.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Murphy of SWI, Dr Neill Bass consultant to MOWI and, Dr Ciar O’Toole, Technical Advisor to ALAB. It is necessary to describe their affidavits to understand the cross-examination. Before I do so I should say a little more of my judgment of 16 March 2023.

1272. I observed that,

• there was no issue but that the modelling software used by RPS was very well-established and used and relied upon by many and varied governmental and professional entities.

• software is not mechanistic – all depends on how you use it and what assumptions, judgments and choices you make in using it, what data and other inputs you use and to what use you put its outputs. That RPS used excellent software is of course a necessary reassurance as far as it goes – but it is not, per se, an answer to the real issue.

1273. Modelling involves collection of such data as is available, choice of data for input to the model and the making of assumptions on which mathematical calculations are based with a view to generating data modelled as likely to arise on the assumptions and in the circumstances modelled. The reliability and usefulness of the modelled data output depends greatly on the choice, nature, quantum and reliability of the data input, the assumptions made, the quality of the modelling software, and the quality of algorithms used to perform the mathematical calculations to produce the modelled output data. All these variables should be open to interrogation in justification of the use to which the modelled output data is put. In short, model output is a creature of the data input to the model and the assumptions made in processing it. Also, it is vital to ensure that the correct model output is put to the correct use in drawing conclusions from the model.

1274. I noted in Clifford1940 that that experts are expected to “show their workings” where that is reasonably practicable, rather than demand deference for assertions or conclusions merely because of their

position or qualification. As Collins J said in Duffy,1941 mere assertion or “bare ipse dixit” by an expert witness

is worthless. I add in this respect that, while RPS and the other experts for ALAB were not witnesses before a court when composing their reports to ALAB, I consider that in a public and controversial quasi-judicial process such as that before ALAB – at very least where the controversy was evident in the process as it readily was here – they bore the duty, identified in the Ikarian Reefer1942 and approved in many Irish cases including by the Supreme Court in O’Leary,1943 of stating the facts or assumptions upon which their opinion was based.

1940 Clifford v An Bord Pleanála, O'Connor v An Bord Pleanála [2021] IEHC 459 (Humphreys J, 12 July 2021). 1941 Duffy v McGee [2022] IECA 254 – Collins J §19.

1942 National Justice Compania Naviera S.A. v Prudential Assurance Co. Ltd (The Ikarian Reefer) [1993] 2 Lloyds Rep. 68 at 81-82.

1943 O’Leary v Mercy University Hospital Cork Limited [2019] IESC 48, MacMenamin J.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1275. In my judgment of 16 March 2023, I canvassed the difficult area of the legal test for irrationality in

decision-making. I anticipated that SWI would rely on the general Keegan1944 test of “fundamental variance from reason and common sense” to argue that it was at that variance for ALAB to accept an RPS Report which input typical rather than maximum values into a worst-case scenario – all the more so when the report did not explain the choices underlying the typical values. I also noted

• SWI’s plea of error of fact and noted Holohan1945 as authority that the Court can “quash for material error of fact”.

• Kevin’s1946 as authority that if the decision-maker in fact adopted an incorrect reasoning process, whether factually or legally, the outcome will not normally be upheld just because it could have adopted, but didn’t adopt, a different and lawful reasoning process.

• that counsel for SWI had observed that the question “How can something that is merely typical be worst case?” stated his case in a nutshell. As a matter of logic, it seems arguable that reliance on an overtly typical scenario could not, as a matter of proper reasoning, inform a worst-case scenario.

• that SWI allege that alternative materials proper to a worst-case scenario were available in the form of RPS modelling data to the effect that DIN levels in each individual cell close to the Salmon Farm would result in DIN Limit exceedances on at least one occasion in each cell in the model run of 22 days.

1276. I did not, of course, decide any such issues in that judgment. I mention that content to again demonstrate that the witnesses can hardly have been surprised to be cross-examined accordingly.

Dr Bass for MOWI – Affidavit 7 July 2022 & Donal Grant for the State – Affidavit 13 July 2022

1277. The affidavits added little to the pleadings until the first of Dr Bass for MOWI.1947 He cited the RPS Water Quality Modelling Report 2015 to the effect that, even in winter when ambient DIN levels in Bantry Bay can reach 120 μg DIN/L, the DIN from the proposed farm at Shot Head, in combination with the other operational facilities in Bantry Bay, would not breach the 170 μg DIN/L limit set for high status waters by the Surface Waters Regulations 2009.1948

1278. For the Minister, Donal Grant1949 deposed that limit of 170 μg DIN/L (0.17 mg/l) for coastal waters is based on cumulative monitoring results of sampling at various coastal water sites around Ireland – both historic results and results obtained for the purpose. Monitoring is led by the EPA.1950

1944 The State (Keegan) v Stardust Compensation Tribunal [1986] I.R. 642. 1945 Holohan v An Bord Pleanála [2017] IEHC 268. 1946 Kevin’s GAA (Flannery et al) v An Bord Pleanála [2022] IEHC 83. 1947 Affidavit of Neil Bass sworn 7 July 2022

1948 He also addressed Dissolved Inorganic Phosphorus and Biological Oxygen Demand but these were not controversial as the case ran.

1949 Affidavit of Donal Grant sworn 13 July 2022.

1950 with assistance from local authorities, Inland Fisheries Ireland, the Marine Institute, the Office of Public Works, and Waterways Ireland.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

John Murphy for SWI – Affidavit 21 October 2022

1279. Mr Murphy’s essential points are that:

• the “typical” DIN values RPS describes and Dr Saunders adopts as likely to be generated by the Salmon Farm discharges and which it added to the median ambient DIN value for the Bay 1951 to produce a total for comparison with the limit of 170 μg DIN/l, are not maximum values and so do not represent the “worst-case scenario hydrological modelling” and “multi-level worst case scenario” asserted by Dr Bass in his sEIS or the “worst case” asserted by Dr Saunders or an appropriate element of a comparator with the limit of 170 μg DIN/l.

• “typical” values have no place in the “worst case scenarios” asserted by Dr Saunders.

• the correct course was to take the maximum ambient DIN level (125μg/l or 0.125mg/l)1952 and add the maximum emission value from RPS Figure 5.1 (70 μgDIN/l), giving a total maximum of 195μgDIN/l– a clear breach, he says, of the “EQS” of 170 μgDIN/l.

• so, the Impugned Decision is irrational as there is no evidence capable of supporting it.

1280. Mr Murphy cites the RPS Report – §5.2.2 and Figures 5.5 to 5.81953 and Tables1954 – which record the highest “typical” DIN plume due to the Salmon Farm at 20-40 µg/l. RPS Tables 5.2 and 5.4 show “average” levels due to the Salmon Farm also at 20-40 µg/l. But the maxima plotted in RPS Tables 5.1 and 5.3, show levels around the site rising by 40-70 µgDIN/l. Mr Murphy cites RPS1955 to the effect that as:

“.. the average winter background level is circa 100 µg/l then a discharge resulting in an increase of less than 70 µg/l would be acceptable.”

Mr Murphy comments that this is, in effect, to add a maximum (from the salmon farm) to an average (ambient).

1281. Mr Murphy cites RPS to the effect that as:

“All concentrations for both outputs are below 10 µg/l (0.01mg/l) and are typically less than a tenth of this value.”

1951 Which he criticises as being a median value of 115μg/l “selected from a range of DIN dispersal plots available”, not a maximum, value – though that

argument receded given Dr Saunders’ adoption of a higher ambient DIN than that suggested by RPS and the sEIS. 1952 As the Technical Advisor did.

1953 Depicting typical ebb and flood plumes for Shot Head only and typical ebb and flood plumes for all Bantry Bay sites.

1954 Incorporated in the figures. 1955 RPS Report 2015 §5.2.1.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Mr Murphy omits the preceding sentence from the RPS Report without which the content he cites cannot be understood. It reads “Figures 5.1 and Figure 5.2 show the Nitrogen concentrations arising from the Shot Head site only, as Maximum and Average Concentration plumes during production.” Mr Murphy comments that

“These figures are clearly contradicted by the graph which shows maximum values which are less than 100 µg/l, but not less than 10 µg/l.”

I have considered Figures 5.1 and Figure 5.2 and conclude that Mr Murphy is right: Figure 5.11956 shows a maximum plume of 0.04 – 0.07mg/l (40 – 70) µg/l. Figure 5.21957 shows an average plume of 0.02 – 0.04mg/l (20 – 40 ) µg/l.

1282. Mr Murphy says that the sEIS, though derived from the RPS Report, adds to the confusion by suggesting mixing zones for DIN1958 and that it continues by deriving,

• from Bantry Bay monitoring results,1959 a median ambient DIN of 0.1152mg/l (115.2 µg/l).

• from adding that median ambient DIN to a highest “typical” plume from the Site at 0.04mgDIN/l (40 µgDIN/l) taken from sEIS Figure 3.3,1960 what the sEIS calls1961 a “peak elevated ambient” DIN level at the Site of 0.1552mg/l1962 (155.2 µgDIN/l) – well within the standard for High Ecological Status of 170 µgDIN/l “on a worst case basis”.

1283. Mr Murphy, in essence, disputes MOWI and ALAB’s assertion that DIN was assessed on a “worst case

basis” and the use of the word “peak” on the basis that a median is a form of average1963 – not a maximum –

and a “typical” pollutant plume is not a maximum either. And he points out that RPS had provided maximum plumes1964 representing a true worst case basis of “0.04–0.07mg/l” (40-70 µgDIN/l).

1284. Mr Murphy summarises that, as to compliance with the limit of 170 µg DIN/l,

• RPS added “the average winter background level [of] circa 100 µg/l” to “an increase of less than 70 µg/l”

(adding an average to a maximum).

• The sEIS added a median background of 115 µgDIN/l to a “worst case typical” of 40 µgDIN/l to get 155 µgDIN/l (both averages, not maxima).

1956 Maximum Plume Envelope of Nitrogen Concentration arising from the Shot Head site only. 1957 Average Nitrogen Concentration arising from arising from the Shot Head site only. 1958 §3.3.4. 1959 sEIS, April 2018, Table 3.3 at p97.

1960 sEIS, April 2018, p99. This is a reproduction of RPS Figure 5.8 “Typical Flood Plume of combined Nitrogen concentration arising from all currently

proposed and existing Bantry Bay sites – January, year 2 – Shot Head / Fastnet dominant.” 1961 sEIS, April 2018, p7 & p98. 1962 0.1152 + 0.04. 1963 The middle value in an ordered sequence of data. 1964 RPS 2015, Figures 5.1 and 5.3.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• Dr Saunders added a maximum mean background of 125 µgDIN/l to a “worst case typical” of 40 µg/l to get

165 µgDIN/l (adding a maximum to a “typical”).

1285. In other words, Mr Murphy asserts that the sEIS, though increasing the maximum background to 115 µgDIN/l and Dr Saunders though increasing the maximum background to 125 µgDIN/l, abandoned RPS’s view that “a discharge resulting in an increase of less than 70 µg/l would be acceptable” and did not apply the RPS maxima of an increase of 40-70 µgDIN/l1965 and instead applied a “worst case typical” of 40 µg/l to stay below the limit of 170 µg/l.

1286. I observe that

• adding that 40-70 µgDIN/l to the median ambient of 115.2 µgDIN/l taken from the sEIS would yield a total of up to 185.2 µgDIN/l – in excess of the 170 µgDIN/l standard for High Ecological Status.

• while that might have sufficed for Mr Murphy’s purpose, he amplifies his point by introducing another variation – that Dr Saunders had taken a higher ambient level of 125 µgDIN/l. Adding this to that 40-70 µgDIN/l would yield a total of up to 195 µgDIN/l – yet further in excess of the 170 µgDIN/l standard for high ecological status.

1287. Mr Murphy concludes that there is no evidence to support the decision or that its rationale is not apparent. Given what followed at trial, I should say that in my view Mr Murphy’s affidavit, if argumentative, consisted in the application of logical reasoning rather than expertise in drawing his conclusions and so I do not find it necessary to ignore that evidence. However he largely expresses views, born of that logical reasoning which I in any event and in appreciable degree likewise reached by reasoning.

Dr Bass – Affidavit 19 January 2023

1288. Dr Bass1966 disputed Mr Murphy’s analysis and seeks to demonstrate, that “Mr Murphy's reasoning on these points is incorrect.” He vouches for the contrary views taken by the Technical Advisor Dr Saunders – and, by implication, the view of RPS on whose report the Technical Advisor relied in large part.

1289. Dr Bass recited and explained the progression of the assumption of ambient levels from 100 µgDIN/l (RPS) to 115.2 µgDIN/l (sEIS) µg/l to 125 µgDIN/l (Saunders). To my mind he does not in any real way disagree with Mr Murphy in this regard. What primarily matters is that, in the end and as informing the

1965 Plotted in RPS Tables 5.1 and 5.3. 1966 Affidavit 19 January 2023.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Impugned Decision, Dr Saunders and ALAB assumed an ambient 125 µgDIN/l. And I would add that it was a mean not a median.

1290. Dr Bass reviewed the RPS modelling purportedly to show why he agreed with Dr Saunders’ use of the ”typical” 40 µgDIN/l as a “worst case projection” of the DIN emanating from the salmon farm. He described RPS as seeking to “ensure that the modelling identified the worst case scenario”. He records various elements of the worst case scenario which SWI did not dispute as proper worst case assumptions.1967 He describes RPS’s various modelled Maximum, Average and Typical DIN plumes – which plumes SWI did not dispute in their own terms. But their significance in a worst case scenario is disputed. Dr Bass then said:

“It is of critical importance to understand that neither the maximum, nor the average, plume modelling represented actual projections, but rather were tools employed by RPS to ensure that the typical projections at ebb and flood, which are the actual projections of nitrogen emissions expected from the Shot Head Farm once operational, were accurate.”

1291. In this regard Dr Bass cites RPS’s explanations of its plume envelope figures.1968 I have already described that system and commented thereon – including on the word “actual”. Dr Bass does not illuminate RPS’s explanations further. For example, he does not explain how RPS chose the typical snapshots, why it chose mid-flood or mid-ebb snapshots or how the maximum and average envelopes assisted in “gauging” the typical values.

1292. Dr Bass asserts as a “fact” that the "maximum plume" does not represent an actual projection and

does not identify the “worst case” DIN projections. Those “worst case projections”, he says, are depicted in

the RPS Typical plumes1969 as 40 µg DIN/L – to which Dr Saunders, properly Dr Bass says, added an ambient

125 µgDIN/L, yielding a total “worst case” of 165 µgDIN/L – below the limit of 170 µgDIN/L. Dr Bass says his

own use in his sEIS, and Dr Saunders’ use in his Final Report, of these “Typical” plumes from the RPA report

to identify “worst case nitrogen levels” was “entirely appropriate as they constitute “actual data” and ALAB

agreed with Dr Saunders.

1293. I observe that Dr Bass does not explain of the typical plumes how or why,

• they were derived.

1967 RPS based its DIN calculation on: Total inorganic nitrogen – including insoluble nitrogen, thee by artificially inflating DIN levels; The worst month of the 22 month growing cycle – when fish biomass of and feed consumption are highest; A modelled point discharge – in fact discharge will be dispersed through the cages.

An assumption that inorganic nitrogen is conserved as it disperses – in fact it is – assimilated by bacteria and fauna in the water column as it disperses.

1968 RPS Report §5.1.2.

1969 RPS figures 5.5 and 5.6. Figure 5.5: Typical Ebb Plume of Nitrogen Concentration arising from the Shot Head site only. Figure 5.6: Typical Flood Plume

of Nitrogen Concentration arising from the Shot Head site only. In fact they are mid-Flood and mid-Ebb – see RPS report p26.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• what they depict was deemed “typical”. What were the criteria for typicality and why were those criteria chosen?

• they were appropriate for inclusion in a “worst case”.

I observe also that,

• in asserting as a “fact” that the “maximum plume” does not identify the worst case projections, Dr Bass was in truth not asserting a fact but was invoking his expert judgment and opinion.

• it is unclear in what sense Dr Bass used the word “actual” of the RPS “typical projections”. They were modelled, just as were the maximum and average plumes. As Dr O’Toole pithily observed “…. the reason you model the data is that the data doesn't exist …”1970 – for example because it was not recorded or because it is sought to describe future events. The output of a model is notional data predicted by the model.

• while the maximum plume does not represent an actual projection in the sense that it does not represent the whole Bay at a single point in time, the modelled maxima for each cell which contribute to the plume/figure/plot are as much “actual” data as are the modelled data which underlie the typical plume/figure/plot.

• putting it at its lowest, including a “typical” parameter in a “worst case scenario” is counterintuitive and so

calls for explanation.

• the like may be said of the phrase “worst case typical” – what does it mean?

• It is not apparent that the choice of a single point, or even two points,1971 in time representing the entire Bay represents the choice of the worst-case point in time.

Dr O’Toole for ALAB – Affidavit 23 January 2023

1294. Dr O’Toole1972 disputes Mr Murphy’s analysis of the DIN calculations in much the same terms as does Dr Bass. She, also, does so essentially by repetition of excerpts of the RPS report. She uses the phrase, “actual or typical”1973 as if the two words are synonymous – which is not apparent to me. She states that the “typical” plots1974 “indicate that the concentrations from the Shot Head fish farm will not exceed

1970 Day 5 p42. 1971 Mid-Flood and mid-Ebb. 1972 Affidavit of Dr Ciar O’Toole 23 January 2023. 1973 Affidavit of Dr Ciar O’Toole 23 January 2023 §29 et seq. 1974 Figures 5.5 and 5.6 of the RPS Report.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

0.04mg/l.”1975 I observe that unless one reads that as “will not typically exceed”, the proposition seems to me a non-sequitur and there is a great difference – even to a non-lawyer – between the phrases “will not exceed” and “will not typically exceed”.

1295. In similar vein, Dr O’Toole cites RPS to the effect that “the maximum typical or actual value for nitrogen … is 0.04mg/l or, in other words, the maximum or worst case actual value for nitrogen is 0.04mg/l.” She essentially relies on the ipse dixit of the RPS report to conclude that it is the typical flood and ebb contour plots, not the maximum plots, that indicate the likely actual DIN dispersal pattern.

1296. Broadly, I make the same observations as I have made regarding Dr Bass’s views. I must add that, as I have said, a proposition that a “typical” plot shows concentrations which will not be exceeded seems to me a non-sequitur. Had Dr O’Toole said that a “typical” plot shows concentrations which will not typically be exceeded I could understand – but that formulation prompts the questions,

• how that “typical” scenario can be described as “worst case” or as based on a “maximum .. actual value”.

• what, in context, is meant by “typical” and how was what is said to be “typical” identified as such.

1297. Dr O’Toole1976 makes the point that the RPS Figures depict ranges of DIN values – and range of 40 –

70 µgDIN/L means that discharges of 70 µgDIN/L could occur – not necessarily that they will occur. That is undoubtedly correct as to fact. But I don’t see that it advances ALAB’s case. The ranges were chosen by RPS for MOWI and tendered to ALAB as part of the basis on which it should make its decision. Proof of certainty of discharges of 70 µgDIN/L is not the point – the risk of such discharges is the point. A “worst case”, considered on the precautionary and preventive principles, assumes the occurrence of 70 µgDIN/L on at least some occasions.

Admissibility & Weight of Evidence

1298. John Murphy’s oral evidence in cross-examination consisted largely of opinion of an expert kind. Its admissibility was challenged. As to experience, knowledge and expertise, Mr Murphy undoubtedly qualified as an expert. However, cross-examination made it very clear that he had, before swearing his affidavits and even as late as his cross-examination, no appreciation at all of the specific duties of expert witnesses – in particular duties of to the court of objectivity, impartiality and independence, as opposed to his duties to the party, SWI, of which he was a director and for which he was a deponent and witness. Those duties are

1975 Emphasis added. 1976 Affidavit of Dr Ciar O’Toole 23 January 2023.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

explained in many cases, including the Ikarian Reefer,1977 O’Leary1978 and Duffy1979 – as is the duty of legal advisors to make those duties known to prospective expert deponents and witnesses. But I hasten to say that, in substance, I found Mr Murphy an entirely straightforward, admirably frank, genuine, sincere and honest witness. I have no doubt that he did his best to honestly assist the court. He very properly and frankly said that he and SWI oppose salmon farming generally and as a matter of policy and in the interests, as they see it, of the protection of wild salmonids. In effect, he is an advocate in that regard.

1299. It is no disrespect to Mr Murphy, his advocacy, his expertise or his opinions to hold, as I must, that I cannot regard him as an expert witness whose opinion evidence is admissible in these proceedings. He lacks the necessary objectivity, impartiality and independence. I can have regard to his evidence only as to fact (he adduced no such evidence) or to any extent that it calls attention to defects in reasoning or logical inconsistencies to which I might have had regard without his evidence and merely by analysis of the materials before me.

1300. Dr Bass and Dr O’Toole were not similarly cross-examined on behalf of SWI as to their appreciation of their duties as expert witnesses, nor was the admissibility of their evidence challenged. As was said in RAS Medical,1980 the onus to raise the status of evidence lies on the party who wishes to assert that it is inadmissible. So their evidence is properly before me as expert evidence. It is no disrespect to Dr Bass to confess to some disquiet that Mr Murphy’s evidence must be excluded whereas Dr Bass’s is admitted. That is not because I doubt Dr Bass in particular but because there is something unreal in excluding Mr Murphy’s evidence while admitting that of an expert who, since 2009 has been retained by MOWI to advance the Shot Head Salmon Farm project and to that end has successively prepared numerous documents advocating the project including two EISs on which the success of the licence application in very considerable degree depended and a written submission and oral evidence to the oral hearing. Dr O’Toole’s position was less acute. She became involved as technical Advisor to ALAB only in October 2020 – though she was put in the position of an expert defending ALAB’s decision on which she had in various respects, advised ALAB. I have recently expressed similar concerns.1981 But I readily acknowledge the difficulty of reconciling principle with practicality in this respect. Such issues are inbuilt in planning and similar litigation in which expert deponents have often had long association with their client – typically as to the very project and planning/licensing process under scrutiny in the proceedings. I also acknowledge that in-house experts are frequently tendered as witnesses for their employers and their evidence is considered as to weight rather than admissibility.1982 Though particular considerations may arise where the in-house expert has advised the decisionmaker in its making the very decision impugned and while expert evidence may be excluded for conflict of interest, that

1977 National Justice Compania Naviera S.A. v Prudential Assurance Co. Ltd (The Ikarian Reefer) [1993] 2 Lloyds Rep. 68, §81-82.

1978 O’Leary v Mercy University Hospital Cork Limited [2019] IESC 48, MacMenamin J. 1979 Duffy v MaGee [2022] IECA 254. 1980 RAS Medical v The Royal College of Surgeons in Ireland, [2019] 1 IR 63, §§70, 92, 113, 114.

1981 Environmental Trust Ireland v. An Bord Pleanála & Cloncaragh Investments [2022] IEHC 540; Donegal County Council v Planree [2024] IEHC 194,

§161.

1982 Galvin v Murray & Cork County Council 2001 2 ILRM 234 & 2001 1 IR 331; Freeney v. Health Service Executive [2020] IEHC 286.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

will not necessarily be the case – Toth1983 and Factortame.1984 Nor is the issue of admissibility of expert evidence decided as to conflict of interest on the objective bias test applicable to decisionmakers – for reasons including that the rigid application of such a test would pose insurmountable problems in litigation so the ideal must bend somewhat to the practical – Factortame, O’Leary1985 and UCC v ESB.1986 There is no entirely satisfactory solution but determined insistence on the Ikarian Reefer duties of experts to the Court is at least part of it. The present case is one in which the unfortunate contrast arises as between the respective parties’ witnesses in that the ideal cannot, I think, bend so far to the practical as to render Mr Murphy’s opinion evidence admissible on matters of expertise.

1301. However, that conclusion does not oblige me to ignore observations by Mr Murphy which seem to me based in common-sense or simple logic – even if evidence were needed thereon – that the mobilisation of “typical” data is erroneous or at least counterintuitive and requires explanation (I add, all the more so where its typicality is unexplained). Neither need I ignore Mr Murphy’s evidence as to factual error in the material before ALAB – notably RPS’s erroneous assertion that its data showed concentrations at below 10 µg/l.1987

1302. Admissibility aside, the evidence of Dr Bass and Dr O’Toole was challenged in substance and I am not obliged to accept their evidence even if uncontradicted. If I do not, I must give my reasons – see Duffy v McGee1988 and Hyper Trust.1989 It has been said that judges should approach all experts with polite scepticism.1990 In Duffy, Collins J said that the court must not surrender its judgment to expert witnesses.

“The court must be able to understand and engage with the evidence, which in turn requires that experts should sufficiently explain their opinions and the basis for them… An expert witness must “provide material on which a court can form its own conclusions on relevant issues” … Mere assertion or “bare ipse dixit” on the part of the expert witness is, accordingly, “worthless””.1991

1303. Charleton & Rakhmanin1992 usefully observe that judging expert evidence will remain focused on the testimony itself. They suggest that relevant to its weight is its internal consistency, rationality, logic, precision and accuracy of thought – as demonstrated particularly in facing up to the logic of a contrary proposition – the care with which the issue was considered and any extent to which the evidence is

1983 Toth v Jarman [2006] EWCA Civ 1028, [2006] 4 All ER 1276.

1984 R (Factortame) v Secretary of State for Transport, Environment and the Regions, [2002] EWCA Civ 932, [2003] QB 381, [2002] 4 All ER 97, [2002] 3

WLR 1104. 1985 O'Leary v Mercy University Hospital [2019] 2 IR 478. 1986 University College Cork v ESB [2015] IEHC 598, §1186. 1987 See above. 1988 Duffy v McGee [2022] IECA 254 – Collins J §18, Noonan J §80. 1989 Hyper Trust Ltd v FBD Insurance PLC [2021] IEHC 78 (McDonald J, 5 February 2021), §254. 1990 Charleton & Rakhmanin, The Safe Use Of Expert Evidence, [2023] Irish Judicial Studies Journal Vol 7(1) . 1991 Duffy v McGee [2022] IECA 254 – Collins J §19. 1992 Op cit. supra.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

speculative (as opposed to, I infer, based in the evidence and demonstrated deployment of recognisable expertise). I have found these observations useful in considering the evidence in this case.

Order for Cross-Examination & Cross-Examination of Dr Bass and Dr O’Toole #21993

1304. By the judgment of 16 March 20231994 I granted leave to SWI to cross-examine Dr O’Toole and Dr Bass on the issues:1995

• whether it was correct to use “Typical” DIN levels i. in a worst-case scenario analysis ii. modelled as generated by the salmon farm, in computing whether a breach of the 170 µgDIN/l limit1996 will occur.

iii. specifically those modelled as generated by the Salmon Farm and chosen by RPS as Typical, in computing whether a breach of the 170 µgDIN/l limit will occur.

• whether data which informed the “Maximum Plume Envelope” should have been used instead of the “Typical” DIN levels.

1305. I think it fair to say that while Dr O’Toole became ALAB’s full time technical advisor in October 2020 prior to its Impugned Decision and in succession to Dr Saunders, she had no relevant interaction with Dr Saunders. She was reliant for her understanding of the WFD analysis as to DIN on the documents before the court.

1306. In short, neither Dr Bass nor Dr O’Toole were able in cross-examination to discern from the RPS Report or explain to the Court the basis and rationale on which

• RPS selected their typical plumes (other than that as a matter of fact they were mid-tide as opposed to slack water1997 – though it was not apparent why mid-tide values were chosen as opposed to slack water ).

• RPS’s averages – which were means – could be mobilised in assessing compliance with an EPA monitoring system based on medians.

• RPS or Dr Saunders1998 considered that those typical plumes could properly contribute to a worst-case scenario.

• in their own view, those typical plumes could properly contribute to a worst-case scenario.

1993 On Day 5 of the trial. 1994 Salmon Watch v ALAB [2023] IEHC 129. 1995 I have amended for clarity the wording I used in that judgment without changing substance. 1996 My judgment says “EQS” but it seems to me that the term “limit” is preferable for present purposes. 1997 See below.

1998 As I have said, it seems to me that properly understood RPS had not made that assertion – in my view it originated from Dr Saunders’ interim report.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1307. Dr Bass described the “typical” plumes as “snapshots” of a point in time.1999 He agreed with a leading

question from MOWI’s counsel suggesting they were “representative of what would occur once the site was operational”. That his answer held no weight is apparent from the following passage of evidence:

“MR. JUSTICE HOLLAND: They chose a snapshot.

A. Yes.

MR. JUSTICE HOLLAND: Why choose that snapshot?

A. Well, as I explained, I leave that to the expertise of RPS who are very expert in my opinion.

And -- MR. JUSTICE HOLLAND: Well, I mean if you can't interrogate that how can ALAB? We're at the mercy of whatever decision RPS make and how can they interrogate it?

A. Well, as I said earlier, as far as their interrogation is concerned they are quite dependent on the maximum plume plots and the average plume plots to reach that snapshot point. I think really you need to see the way in which they're looking at this thing. It's like a movie, and I have seen some of these, incidentally, I haven't been involved in any selections of snapshots.

MR. JUSTICE HOLLAND: I have seen them run through, I have seen versions.

A. Yes, and the whole thing just runs. What you will see, since you may know, you literally see the tide going in and out, you see these values for whatever it is going up and down during that period and it is a question of choice.

MR. JUSTICE HOLLAND: But everything depends on when you choose to press the stop button?

A. Pretty well, yes.

MR. JUSTICE HOLLAND: So you have to have standards or rationales for deciding when to press the stop button, when to choose a particular snapshot?

A. Yes, I would agree with that. There are certain expertise in this courthouse that I know absolutely nothing about. In my field there is another completely different range of expertises which I think, with respect, should be left to the scientists to judge.

MR. JUSTICE HOLLAND: Sorry, which should?

A. Should be left to the scientists to involve the experts that are involved with that technology to judge when that snapshot should be taken.”

While, as so often when oral testimony is reduced to writing – it is no criticism of Dr Bass – his last two replies look a little unclear, I am entirely satisfied that, in referring to the scientists to whom the choice of snapshots as typical should be “left to judge”, he was referring to RPS and their report of 2015.

1308. Dr Bass said that “What RPS did in selecting typical, and this is a selection process, it's nothing more than that, guided by the maximums and the averages, they selected by going through the entirety of the

1999 Transcript Day 5 p124.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

output, as it were, as a moving sequence, they selected typical values themselves”.2000 When asked by what standard RPS had selected typical values, Dr Bass replied, “By their expertise.” When asked how could those in court, or ALAB, interrogate that that selection by RPS of what they decided was typical – because everything turns on it – Dr Bass strikingly replied “Well, I think you're asking the wrong person. I don't have that information, I am afraid.”

1309. When asked “Well, if you can't interrogate it how could ALAB interrogate it?” Dr Bass replied “Well, I think, on the basis of my expertise and what I saw from the RPS report, and presumably on the basis of Dr Saunders’ expertise and the expertise within ALAB, they took what they read from the RPS report on face value.” “

1310. “Face value” is not a term of art. Generally, and in its ordinary meaning, it refers to the apparent value of a proposition or assertion before that value has been tested in any way. I confess that I have great difficulty understanding how Dr Bass’s assertion that ALAB took the RPS report at face value conforms to the required deployment of ALAB’s expertise and scrutiny in reaching their conclusion as the law requires. In fairness, ALAB retained Dr Saunders to advise – but that does not assist if he took the RPS Report’s choice of typical values at face value also. However the conclusion is what matters – at least in demonstrating that RPS’s reasoning was opaque to Dr Bass.

1311. This evidence of Dr Bass seems to me a version of the sort of expert evidence which Collins J had in mind when he said in Duffy v McGee:2001

“To properly perform its function, the court must be able to understand and engage with the evidence, which in turn requires that experts should sufficiently explain their opinions and the basis for them. Their entitlement to express such opinions “is predicated upon also informing the court of the factors which make up their opinion and supplying to the court the elements of knowledge which their long study and experience has furnished to them whereby they have formed that opinion so that, in those circumstances, the court may be enabled to take a different view”:…. It follows that the expert witness must “provide material on which a court can form its own conclusions on relevant issues” … Mere assertion or “bare ipse dixit” on the part of the expert witness is, accordingly, “worthless””2002

Indeed, in the present case the “bare ipse dixit”2003 was not even that of the witness under crossexamination: Dr Bass uncritically adopted and tendered to the court the “ipse dixit” of another expert who did not testify and could not be cross-examined – RPS.

2000 Transcript Day 5 p117 et seq. 2001 [2022] IECA 254, §19.

2002 Citing Flynn v Bus Eireann [2012] IEHC 398, per Charleton J §9 and Pora v The Queen [2016] 1 Cr App R 3, §24.

2003 The phrase “ipse dixit” refers to an assertion which the court is asked to accept based solely on the maker’s authority.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1312. This is clearly revelatory of the basis on which Dr Bass, in his authoring the 2018 sEIS, had adopted the typical plumes as properly contributing to a worst-case scenario and, indeed in his co-authoring with RPS themselves, their September 2017 presentation to the Oral Hearing. While, no doubt, it would be notable in any event, it is especially notable given the advance notice of the subject matter of inquiry given to these witnesses by the order for cross-examination. I do not interpret this as a failure on their part to consider the matter or prepare for cross-examination – rather it reinforces the impression that the basis and rationale for choosing the typical plumes and for deploying them counterintuitively in a worst-case scenario was simply

not discernible to Dr Bass or Dr O’Toole – or, for that matter, to ALAB – from the RPS Report. Of course, I do

not rule out that it may have been discernible to RPS – but they did not adopt or state such a basis and rationale in their 2015 Report and were not witnesses in the trial.

1313. When it was directly put to her in cross-examination that these typical plumes were selected

(deliberately) to reassure the reader, Dr O’Toole replied: “I don’t know”.2004 I asked Dr O’Toole “how can

something that is described as "typical" form part of a worst-case assumption when chosen in preference to the maxima in Figure 5.1?” She replied, referring to RPS, that “ ….the maximum is worst case maximum, the average is worst case maximum. I know they called it "typical", I don't know why they particularly picked typical, it doesn't have a statistical definition. … It means they have gone through the data and picked a – I am going to say a point in time, it's not a real point in time, obviously, it's a point in time during the simulation where they feel this particular figure best represents what is most likely to happen under worstcase scenarios.” I asked Dr O’Toole whether she knew “what criteria they used in deciding what was typical?” She answered “No. That's why you need to think about it in terms of the maximum average and in terms of the hydrological model itself. So the first phase of the model.”

1314. I confess that my only clear impressions from this evidence, and that which preceded and succeeded it, which I have not set out, is that

• “typical", doesn't have a statistical definition,

• Dr O’Toole didn’t know why RPS particularly picked typical data and

• Dr O’Toole didn’t know what criteria RPS used in deciding what was typical.

1315. Indeed, and it seems to me noteworthy, Dr O’Toole accepted that the RPS typical scenarios were at the fastest-flowing points in the tidal cycle – mid-ebb and mid-flood – at which point DIN dispersal would have been greatest and best not least and worst.2005 Dispersion reduces concentration. In contrast, at slack water DIN dispersal would have been at its least and one might imagine, exceedance of the DIN limit more likely to occur. Dr O’Toole said she would have been concerned at the absence of a slack water presentation

2004 Transcript Day 5 p102. 2005 Transcript Day 5 p101.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

but for the fact that the tidal current in Bantry Bay is very slow indeed – about 10 cm/s. But she agreed that, while it did not cause her concern, this did not explain the choice of mid-tide figures as opposed to slack water figures. She could not explain why slack water had not been chosen as a worst-case. I returned to the RPS Report in light of this evidence. It records the generally unusual tidal conditions in Bantry Bay to which

Dr O’Toole referred – slow tidal currents.2006 But RPS take some care to distinguish dispersal potential at Shot

Head from the general conditions – not unexpectedly perhaps where, at least generally, such potential is reassuring as to dilution of various forms of pollution. RPS state:

“ ……. particularly dispersion potential in an area may be assessed by the examination of residual currents .…. residual currents are relatively low in the main body of Bantry Bay. However, around most islands and promontories … much higher residual currents are generated, … In regions such as this, residual currents provide enhanced mixing and dispersion. Good residual currents are evident around the salmon farm sites in Bantry Bay such as at .. Shot Head. Thus, although relatively low tidal currents in these areas may suggest reduced dispersion, the dispersion models in this report show that the presence of good residual currents … enhance dispersal … at the sites. This encourages … the carriage of wastes out of the bay and into the Atlantic circulation.”2007

1316. So, whatever Dr O’Toole’s view of RPS choice of mid-tide as opposed to slack water for their Typical figures, it does not seem that RPS’s choice was based on a want of dispersal capacity due to tidal conditions

– slow tidal currents – in the Bay. Ultimately, I confess that I did not find Dr O’Toole’s view in this regard

reassuring as to this element of RPS’s choice of Typical figures. In any event, Dr O’Toole was unable to assist on the more important general question of why in September 20172008 (they had not done so in 2015) RPS chose typical figures of any kind, as opposed to maxima, as part of a worst case scenario and why it deemed typical the particular data it did chose.

1317. Of course all this may be somewhat unfair to RPS. As I read their 2015 report, RPS did not deploy the typical values in their worst-case analysis – they deployed the maxima. It was Dr Saunders who, in his interim report, first deployed the typical values in his worst-case analysis. But Dr Bass and RPS in their presentation to the oral hearing and Dr Bass in his sEIS followed suit. It is in those circumstances disconcerting that Dr Bass cannot explain why he did so.

1318. In short, Dr O’Toole and Dr Bass were at a loss to provide a rationale for: o the RPS’s choice of Typical data for DIN likely to be generated by the Shot Head Salmon farm. o the counterintuitive deployment of those Typical data as part of a worst case scenario analysis of the risk of breach of the limit of < 170 µgDIN/l.

2006 RPS Report 2015: p11 – “The general magnitude of ebb end current speeds are generally less than 0.1m/s as the convergence of tides in the outer

domain limit the prevailing currents.” 2007 RPS Report 2015: pp13 & 14. 2008 i.e. in the Bass/RPS presentation to the oral hearing.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1319. Of course, the counterintuitive is not ruled out in principle – especially as to matter scientific or calling for expertise. But where the counterintuitive is important to the reasoning process and YY2009 identifies common sense as the starting point in assessing the adequacy of reasoning, it follows that the counterintuitive usually requires explanation. This view is consistent with the view stated in Connolly that required reasoning may be “complicated or scientific if the issues which arose in the context of the grant or refusal of permission required engagement with such issues.”2010

1320. To apply YY,2011 I am at a loss to “genuinely understand the reasoning process” – even on a “broad and common sense” basis – as to the choice and deployment of Typical data which were central to the WFD analysis as to DIN. Strikingly, it became apparent on examination that the experts tendered by ALAB and MOWI – Dr O’Toole and Dr Bass – were equally at that loss.

WFD – DIN Issue – Decision

1321. Both the inherent complexity of the WFD and the interposition of the considerable developments wrought by the caselaw – most notably the Weser case – while the licensing process was in train, render it impossible not to have appreciable sympathy with MOWI’s advisors and ALAB and its advisor’s as to their treatment of the WFD. However that sympathy cannot decide the issue of the legality of the Impugned

Decision. I confess to having had real difficulty in discerning the relationship in the Impugned Decision between the WFD regime as recorded in the various relevant EU Directives and the Irish implementing regulations and the applicable case law described above, and the role of the EPA (including its primary reliance on medians) on the one hand and, on the other hand, ALAB’s assessment of the issue of DIN discharges by the proposed salmon farm.

1322. I bear in mind that I cannot review the decision in this regard as to its merits. However I hope the analysis above may assist ALAB’s approach to such matters.

1323. It seems to me that I should consider ALAB’s decision by reference to the approach overtly taken by RPS, Dr Bass and Dr Saunders, on whom ALAB relied. They clearly approached the matter on the basis of a “worst case” analysis the rationale of which is that the <170 µgDIN/l will not be breached at and within 1 km of the Shot Head Farm. It may be that the rationale was that the <170 µgDIN/l will not be breached at and within 1 km of the Shot Head Farm it will not be breached in the bay more generally. Either way, the proposition is that there will be no deterioration of ecological status class either locally or in the bay generally. However either conclusion is clearly dependent on calculating the possibility of breach of the <170

2009 YY v Minister for Justice [2017] IESC 61. 2010 §74. 2011 §80.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

µgDIN/l limit using RPS’s Typical data – the basis for choosing which is nowhere explained. That this is so is amplified by the inability of Dr Bass and Dr O’Toole to identify that basis in evidence – all they could do was defer to the expertise of RPS as to its unexplained choice of the Typical data. And Dr Bass in the sEIS and Dr Saunders in both his reports in put that Typical data to a use to which RPS itself did not put it in its 2015 Report and to a use which RPS did not explain in its 2017 presentation to the oral hearing.

1324. Phrases such as “peak elevated ambient”2012 are at best confusing when applied to what are clearly means not maxima – though, as on a careful reading it is clear that the reference was to a highest mean ambient DIN level, I do not decide the case on that basis. However, the phrase “Typical worst case” used in the Bass/RPS presentation to the Oral Hearing seems, prima facie, oxymoronic. In my view, the inclusion of the Typical data in a “worst case” analysis, as to which RPS “emphasised that, at every point where a choice was available, the worst case input option was selected for simulation”,2013 required explanation and was not explained. Even in cross-examination and by their own admission, Dr Bass and Dr O’Toole were unable to ascribe clear meaning or rationale for RPS’s choice of “typical” data and could not explain RPS’s choice of mid-tide as typical when at that point in the tide flow, and hence dispersal of DIN, are at their height, as compared, for example, to slack water. However, this observation is only a particular aspect of a general absence of explained rationale for the choice of data deemed typical.

1325. Clear, cogent and meaningful reasons must disclose the essential reasoning process underlying an impugned decision. As has been observed, SWI plead2014 inadequacy of reasons in the Impugned Decision in that ALAB failed to explain how it concluded that there would be no breach of the “EQS” of 170 µgDIN/l. I have commented on the question whether the concept of an “EQS” for DIN is apt. But if it is inapt it is an error common to all parties and there is no doubt that the standard of <170 µgDIN/l is a limit imposed for WFD purposes by the Surface Waters Regulations. The plea is adequately clear and it would be unjust to shut SWI out on this account – nor did any party suggest that it should be shut out on this account.

1326. I confess to discerning a general poverty of engagement with and analysis of WFD requirements which ALAB will wish to consider in any remitted process. For example, I am unclear how effect on a status determined by the EPA by reference to median values at all, some or one of particular monitoring points over particular periods can be discerned without reference to that method of determination and/or based on RPS data which do not include medians and include only means. In any event, MOWI’s and ALAB’s

mobilisation of RPS’s “typical” data was essential to its asserted “worst-case” analysis. To my mind, that

position was highly counter-intuitive and so required explanation. I allow for the possibility that there may in fact have been unarticulated good reason underlying the mobilisation of the Typical data in ALAB’s analysis and so I confine myself to quashing the Aquaculture Licence decision in this respect for inadequacy of

2012 sEIS, April 2018, pp7 & 98. 2013 RPS Report pp1, 27, 28, 30. Emphases added. 2014 Grounds §E1.16.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

reasons by way of explaining the criteria for deeming the typical DIN data typical and for mobilising typical DIN data in the worst case analysis of whether the limit of <170 µgDIN/l will be breached.

1327. I am content to rest my decision to quash the Aquaculture Licence by reference to its WFD analysis on inadequacy of reasons. But it is of some interest to note that, in England and Wales the decision of Saini J in Wells2015 has been influential2016 in its pointing up two distinct strands of Wednesbury irrationality. The

first is “an unexplained evidential gap” is in what I may call the “evidential space”, as is O’Keeffe2017 – though

O’Keeffe is less demanding of decision-makers. The second is an unexplained leap in reasoning which fails to justify the conclusion reached by the public law decision-maker. Indeed it seems to me that the concept of reasoning deployed in Irish law as to adequacy of reasons is a concept not greatly distinct from that strand of

the concept of rationality identified in Wells. Arguably, in both its choice and identification of the “typical”

data and its deployment, the Impugned Decision fell short as to the second type of irrationality identified by Saini J.

1328. In light of my decision above to quash the aquaculture licence on grounds of inadequacy of reasons as they relate to the DIN issue under the WFD, I need not consider other WFD grounds – for example as to Adrigole Harbour or protected areas within the meaning of the Surface Water Regulations 2009. However, I will consider further the WFD issues as they relate to EmBZ and Benthos.

WFD – EmBZ

1329. EmBz is a dietary treatment for sea lice, typically used over 7 consecutive days. It can protect for up to 120 days – whereas waterborne treatments act only at administration. The greater the biomass of salmon to be treated, the greater the dosage of EmBz required at standard concentrations and vice versa. EmBz is discharged to the water column via waste feed and faecal material.

1330. SWI’s submissions as to EmBz were less extensive than the case pleaded.

2015 R (Wells) v Parole Board [2019] EWHC 2710 (Admin) §33.

2016 For example, recently in Friends Of The Earth v Secretary Of State For Energy Security And Net Zero [2024] EWHC 995 (Admin) §127.

2017 O’Keeffe v An Bord Pleanála [1993] 1 I.R. 39, [1992] I.L.R.M. 237.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Promulgation and binding force of the EQS for EmBZ

1331. SWI pleads that the EQS for EMBZ2018 set by the Minister is invalid in effecting requirements of the WFD and the EQS Directive 20082019 as set without giving it unquestionable binding force contrary to Article 4(3) TEU2020 and Article 288 TFEU.2021 It says the EQS is unenforceable/invalid as not properly published in accordance with the Statutory Instruments Act 1947. The State pleads Article 19 of the Control of Dangerous Substances in Aquaculture Regulations2022 in reply – to the effect that Article 19(3) requires compliance with a “water quality standard” established by the Minister for Agriculture in accordance with Article 19(1), such that water quality standards are of unquestionable binding force.

1332. The oddity of the position in this regard is that all were agreed that the EQS for EmBz, whether valid or not, is 0.22 ng/l measured in the water column 24 hours after use and 100m from the site. However, noone exhibited the formal (or, as SWI would have it, inadequately formal) legal instrument adopting that EQS and no-one addressed the question whether an “environmental quality standard” falls within the meaning of a “water quality standard” (a term not defined in the Regulation or in Directives 2006/11/EC2023 and 2000/60/EC (the WFD) which it carries into effect) as the latter term appears in Article 19. The RPS report2024 says that the Dangerous Substances Regulations 2008 “states” the EQS for EmBz at 0.22 ng/l measured in the water column 24 hours after use and 100m from the site. Dr Saunders cites RPS for that EQS.2025 In fact,

those regulations do not set an EQS for EmBz – Article 6(2) envisages “publication” of EQSs by the Minister in

accordance with the WFD, taking into account, in particular, the toxicity, persistence and bioaccumulation of the substance concerned in the environment into which it is discharged.2026 No instrument, headed, signed or published by the Minister setting such an EQS for EmBz has been put in evidence, as one would have expected.

1333. However, this issue, though pleaded, was not pursued in submissions and, in the absence of the instrument setting the EQS for EmBz, it does not appear to me that SWI has discharged its onus of proof of its invalidity.

2018 Emamectin benzoate is, as used in aquaculture, a chemical pesticide used to kill sea lice. It is commonly known by its trade name “Slice”.

2019 Directive 2008/105/EC on environmental quality standards in the field of water policy.

2020 The Treaty on European Union – Article 4(3) cites the principle of sincere cooperation and requires that Member States take any appropriate measure,

general or particular, to ensure fulfilment of the obligations arising out of the Treaties or resulting from the acts of the institutions of the Union.

2021 The Treaty on the Functioning of the European Union – Article 288 states inter alia that a directive shall be binding, as to the result to be achieved,

upon each Member State to which it is addressed, but shall leave to the national authorities the choice of form and methods.

2022 European Communities (Control Of Dangerous Substances In Aquaculture) Regulations 2008. 2023 On pollution caused by certain dangerous substances discharged into the aquatic environment. 2024 RPS Water Quality Modelling Report November 2015, p60. 2025 Dr Saunders’ Final report 8 December 2020 §6.6, p62. 2026 European Communities (Control Of Dangerous Substances In Aquaculture) Regulations 2008. Art 6(2).

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Breach and Adequacy of the EQS for EmBz

1334. SWI’s submissions record that Dr Saunders cites RPS to the effect that, due to low residual currents at Shot Head, the EQS for EmBz may be breached at a post-year 1 stock biomass. Essentially, the EQS of 0.22 ng/l will be met in 36 hours not 24.2027 The RPS report2028 makes clear that:

• This forecast of breach is modelled at the EmBz dosage for a biomass of 1,300 tonnes of fish.

• The conclusion is that “EmBZ cannot be used at this dose rate at the Shot Head site”.

• Further calculation and modelling discloses that the maximum stock that could be treated (without breaching the EQS) is 440 tonnes, at month 7 post-smolt transfer, in May of the first growth year which would protect the stock to about the following September. SWI says that, accordingly, ALAB restricted EmBZ use to the first 7 months of the growing cycle and required compliance with the EQS.

1335. However, the net complaint by SWI in its written submissions2029 is not of anticipated breach of the EQS but is that application of the EQS is itself insufficient to the high quality water status of Bantry Bay.

1336. SWI invokes Annex V §1.2.4 WFD which provides normative definitions for, inter alia, high ecological

status coastal waters and, as to “Specific synthetic pollutants” requires concentrations “close to zero and at

least below the limits of detection of the most advanced analytical techniques in general use.” This normative definition may be contrasted with the normative definition for specific synthetic pollutants in good ecological status coastal waters as “Concentrations not in excess of the standards set in accordance with the procedure detailed in section 1.2.62030 without prejudice to Directive 91/414/EC2031 and Directive 98/8/EC.2032 (< EQS)”. This definition, when contrasted with that for high ecological status coastal waters may mean that compliance with EQSs merely achieves good ecological status for coastal waters. In this regard it bears recollection that, non-deterioration apart, the general aim of the WFD for surface waters is good status, not high status, so setting EQSs to that end could make sense.2033 The WFD does not define “synthetic” for this purpose – ordinarily it means “man-made”. However the WFD does identify “specific” pollutants2034 as “priority substances”2035 and “other substances identified as being discharged in significant quantities into the body of water”. Priority substances are listed in Annex X WFD. My perusal of Annex X has failed to ascertain and, more importantly given the impenetrability of that list to the layperson there is no evidence,

2027 Dr Saunders’ Final report 8 December 2020 §6.6, p54 2028 RPS Water Quality Modelling Report November 2015 §5.6.1 p60. 2029 §3.17.

2030 Annex V §1.2.6 WFD relates to setting EQSs for 9 pollutants listed in Annex VIII WFD which provides an “Indicative List of Main Pollutants”. That list

includes “Biocides” – which are defined by the Biocides Regulation 528/2012 as substances used “with the intention of destroying, deterring, rendering

harmless, preventing the action of, or otherwise exerting a controlling effect on, any harmful organism by any means other than mere physical or

mechanical action”. While that definition may well encompass EmBz, the Biocides regulation is highly complex – not least in its exclusion of application

to substances regulated by other more specific legislation. 2031 Plant Protection Products Directive 91/414/EEC– replaced by Plant Protection Products Regulation 1107/2009.

2032 Biocides Directive 98/8/EC replaced by Regulation 528/2012 – concerning the making available on the market and use of biocidal products.

2033 WFD recitals 4, 19, 24, 25, 26, 31, 32, 33 & Article 1. 2034 WFD Annex V §1.1 Quality elements for the classification of ecological status. 2035 Article 2 §30 WFD – substances identified in accordance with Article 16(2) and listed in Annex X.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

that EmBz is a priority substance. The phrase “other substances identified as being discharged in significant quantities into the body of water” is not elaborated in the WFD and seems to allow appreciable discretion to Member States, though significance must presumably be considered in terms of environmental significance for the status of the body of water in question having regard to the normative definitions set in Annex V. It will be appreciated from the foregoing brief, and obiter, account that the issue raised in submissions by SWI is not simple and could have widespread ramifications.

1337. ALAB, MOWI and the State objected to this complaint as not pleaded. SWI relied on §E2 – 9.10 &

9.11 of its Grounds. §9.10 is entirely narrative. §9.11 is entirely general in pleading breach of legislation identified only by title. Neither comes close to pleading the essential point made in submissions – that mere compliance with an EQS for EmBz is insufficient to meet the normative definition for high ecological quality coastal water status and avoid deterioration thereof.

1338. Accordingly, I must uphold the pleading objections and reject the allegation of breach of WFD requirements as they relate to EmBz.

WFD – Benthos

1339. The benthos, or benthic community, is the community of organisms living on, in or near the bottom of a water body. Many such organisms are invertebrates.

1340. As has been seen, SWI CG9a pleads ALAB’s breach of Article 4 WFD – which prohibits deterioration of surface water bodies – by allowing discharges to waters that would eliminate the benthic community beneath the fish farm. As to fact, this is not disputed in any significant degree. SWI pleads2036 that Dr Saunders in 2020 advised of,2037 and ALAB found2038 a highly2039 localised adverse impact of the fish farm by settleable solids – organic waste – on the unremarkable and locally common benthic community – but confined to the area directly beneath the cages. This would “significantly impact the benthic habitats in that area. Most of the sessile epifauna2040 is expected to be lost. The infaunal2041 communities will be substantially degraded and are likely to be dominated by more pollution tolerant species.”2042 This effect would be similar to that of other finfish farms of equal density and involved no concerns for rare or vulnerable species. In

2036 Grounds 20/12/22 §E2 9.8 & 9.9. 2037 Dr Saunders’ Final Report 8 December 2020 §6.5.4, p51-52. 2038 ALAB Determination §6.5.1, p23-24. 2039 Dr Saunders’ Final Report 8 December 2020 §9.8.3 Benthic impacts. 2040 Immobile fauna living on the sea bed. 2041 Living in, as opposed to on, the sea bed. 2042 Dr Saunders’ Final Report 8 December 2020 §6.5.4 Benthos.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

similar vein, SWI’s submissions also assert that Dr O’Toole2043 cites ALAB’s former Technical Advisor2044 to the

effect that fish farm discharges are “expected to degrade the seabed communities immediately beneath the cages, but given their unremarkable nature, this will represent a small loss that will not have a significant impact on the benthic ecology of Bantry Bay as a whole.” ALAB’s former Technical Advisor also observed in this passage that the benthos is “typical for the north-east Atlantic and … widespread in Bantry Bay”.

1341. SWI plead that ALAB noted that the licence would prescribe compliance with Monitoring Protocol No. 1 for Offshore Finfish Farms – Benthic Monitoring (the “Benthic Monitoring Protocol”). I note that it requires and sets the protocol for an annual benthic survey and annual report to CZMD2045 which inform the establishment of an Allowable Zone of Effect and any need for a Benthic Amelioration Plan. However it is not apparent to me that this plea informs any claim for relief.

1342. As to legal error, SWI plead2046 that ALAB “failed to recognise”:

• the normative definition set by WFD Annex V §1.2.4 for “Benthic invertebrate fauna” in high ecological status coastal waters that,

“The level of diversity and abundance of invertebrate taxa is within the range normally associated with undisturbed conditions. All the disturbance-sensitive taxa associated with undisturbed conditions are present.”

• Article 37(1) of the Surface Waters Regulations – to the effect, inter alia, that a surface water body shall be

classified as high ecological status if, the calculated Ecological Quality Ratio (“EQR”) values for the biological quality elements2047 show no or only very minor evidence of distortion from undisturbed or reference conditions. The plea is not explicitly applied to benthos and no quantified EQR is invoked as applicable thereto – though I note that biological quality elements include composition and abundance of benthic invertebrate fauna.

1343. As to this somewhat oblique plea of failure to “recognise”, SWI’s submissions state that the “point here” is that the described effect on the benthos will be a permanent significant effect not “within the range normally associated with undisturbed conditions” which effect is “not in accordance with” the WF. There is no attempt to any more elaborate jurisprudential basis for the plea.

2043 Affidavit sworn 23 January 2023 §34. 2044 Dr Saunders’ Final Report 8 December 2020 §6.1, p47. 2045 Coastal Zone Management Division of the Department of Agriculture Fisheries and Food. 2046 Grounds 20/12/22 §E2 9.8 & 9.9 & 9.12.3. 2047 Composition and abundance of benthic invertebrate fauna.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1344. Strangely, ALAB, MOWI and the State make no plea in direct response to the issue of effect on the benthos beneath the cages.

1345. MOWI’s submissions take a pleading point2048 that there is no plea relating the agreed factual description of the effect on benthos to any allegation of illegality – but they do not engage with the substance of the issue. They also refer2049 to Annex II §1.3(i) and Annex V §1.4.1 WFD as they require Member States to establish biological quality reference conditions/values for high ecological status coastal waters and ecological quality ratios (“EQR”), inter alia for Benthic Invertebrate Fauna in high ecological status coastal waters. The EQR value boundaries between high, good and moderate status are set by an intricate “intercalibration” as between member states. However MOWI’s submissions take this issue no further than stating the obligation.

1346. ALAB’s submissions also make what I interpret as pleading points: that

• SWI’s assertion of non-compliance with the WFD “is made baldly and without explanation”.

• SWI’s submission that the agreed factual description of the effect on benthos is not “within the range normally associated with undisturbed conditions” as required by Annex V, Table 1.2.4 WFD is not pleaded.

• that requirement is pleaded only in a general recital – not by way of the precise statement of each ground required by O.84 r.20(3) RSC.

• SWI has not identified what the “range” is that is “normally associated with undisturbed conditions” and

how this range would be breached.

1347. Though its plea is oblique, I think it was adequately clear that SWI’s plea was, simply enough, that if as is agreed the fish farm, as to the area directly beneath the cages and even somewhat more widely, would significantly impact the benthic habitats such that most of the sessile epifauna will be lost and the infaunal communities will be substantially degraded and are likely to be dominated by more pollution-tolerant species, it must follow – must be self-evident – that the resultant Benthic Invertebrate Fauna in that area is not within the range normally associated with undisturbed conditions as is required of high quality coastal waters by Annex V §1.2.4 WFD. The plea is adequately clear – as an assertion in its own terms as conveying what the substance of the assertion is.

1348. However, SWI’s problem is that its conclusion is not legally self-evident. It is clear on a consideration of the WFD that this plea falls far short of stateably asserting a legal defect by reference to WFD requirements. The plea does not engage at all with the highly complex system of the WFD – a brief description of part of which I attempt with some diffidence.

2048 MOWI Submissions 3/4/23 §56.

2049 MOWI Submissions 3/4/23 §30 and 31. In fact the submissions are framed more generally – I have reframed them as relevant to Bantry Bay.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1349. The general system of the WFD is to place a numerical value, in the form of an EQR, on the comparison, as to a particular biological element (here Benthic Invertebrate Fauna), between the normative description of “the range normally associated with undisturbed conditions”2050 and actual conditions in the waterbody in question. From the WFD it can be discerned that this involves, first, identifying the relevant “reference conditions” (“undisturbed conditions”) for High Quality Coastal waters specific to the geographical area in question. Then it is necessary to adopt a mathematical method for comparing those reference conditions with actual conditions so as to produce a numerical ratio – the EQR. Next it is necessary to identify the EQR ranges considered to represent respectively High, Good and Moderate water quality. These methods vary between Member States. So intercalibration exercises for EQRs for each biological element (here Benthic Invertebrate Fauna) in each water category (here North East Atlantic coastal waters) were facilitated (over a considerable period of time) by the Commission and expressed in successive Commission Decisions2051 to ensure that the ranges (boundaries between classes high, good and moderate) are both comparable between Member States and consistent with the normative definitions in Annex V §1.2.2052 This description alone suffices to demonstrate a complex process such that a simplistic comparison of the kind pleaded is legally inadequate as a basis for quashing a presumptively valid impugned decision. Such a simplistic comparison also fails to engage with how localised conditions – such as those under a fish farm in a bay having a nominal sea area of about 23,000 ha – are considered to bear on EQR outcomes. Perhaps they do, perhaps they don’t and perhaps there are criteria for deciding. But SWI’s pleadings engage with none of these issues.

1350. In fact, to 2019 the Surface Waters Regulations2053 did not specify an EQR for Benthic Invertebrate Fauna in high ecological status coastal waters2054 and from 2019 specified2055 an EQR of 0.75 or greater – as

calculated using a classification system called the “Infaunal Quality Index”.2056 What this demonstrates is that

• the concept of “the range normally associated with undisturbed conditions” does not imply perfect coincidence with reference conditions – which would be represented by an EQR of 1.

• there is a protocol – the “Infaunal Quality Index” (“IQI”) – which is to be applied in calculating the EQR. I

have not seen it but no doubt it stipulates applicable methods and standards and calculations informing the relevant EQR. It seems reasonable to infer that it is a protocol of some complexity.

1351. As an aside, I observe that in his 2020 final report Dr Saunders appears2057 to have applied a different EQR assessment “standard methodology” – the Infaunal Trophic Index (ITI) apparently used by SEPA – to the

2050 E.g. as to Ecological Status of Coastal Waters, Annex V §1.2.4 WFD.

2051 See e.g. Commission Decision (EU) 2018/229 establishing, pursuant to (the WFD), the values of the Member State monitoring system classifications

as a result of the intercalibration exercise. The recitals to this decision usefully describe the progress of the intercalibration exercises over time. Recital

10 reads: “Member States should apply the results of the intercalibration exercise to their national classification systems in order to set the boundaries

between high and good status and between good and moderate status for all their national types”. 2052 Annex v §1.4.1 WFD. 2053 Schedule 5 Table 8.

2054 Presumably as the intercalibration in that regard had not been completed – see recitals to Intercalibration Decision (EU) 2018/229.

2055 Table 8 – Biological quality elements, Coastal Waters. 2056 Thereby implementing in this regard Intercalibration Decision (EU) 2018/229. 2057 Saunders Final Report p82.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Infaunal Quality Index (“IQI”) stipulated in the Surface Waters Regulations 2019. Dr Saunders does not mention the IQI. However, no issue in this regard was canvassed in the proceedings and an expert view on any significance would be required. I can take none. Whether an issue might arise in this regard would be a matter for ALAB to consider in any remitted decision.

1352. SWI does plead2058 “failure to recognise” Art. 37(1) of the Surface Waters Regulations, which provides for high ecological status EQRs, but takes the plea no further as to Benthos and falls far short of the required precision of pleading. SWI pleads a catch-all as to the WFD issues2059 – that ALAB misdirected itself in law, applied the wrong legal test, failed to consider relevant issues, acted without evidence, failed to give any or any adequate reasons for its decision, and acted unreasonably, and accordingly its decision is invalid. But catch-all pleas do not avail – Halpin2060 – and precision is required in judicial review pleadings. SWI, in pleading that the effects of the fish farm on the benthos will result in a departure from the range “normally associated with undisturbed conditions”, has not engaged at all with the question whether those effects will bring the EQR below 0.75 for the geographical area of the water body relevant for EQR calculation. Indeed SWI never referred, in pleadings or argument to the EQR boundary of 0.75 above which, and as to specifically benthos, coastal waters can be regarded as high quality. My reference to the relevant geographical area is necessarily vague as SWI did not engage with this concept either.

1353. In light of the foregoing, it is clear that, whether one consider them as pleading objections or in terms of stateability, the objections made by ALAB and MOWI are well-founded and the challenge in this regard must be dismissed. Whether, had it been properly pleaded, a case as to effect on Benthos might have been stateable is something ALAB may wish to consider for future reference.

1354. In that context, and obiter, I note that Dr O’Toole considered that:

• the increase in cage numbers and of their footprint from 5.88 hectares to 8.82 hectares, while maintaining the MAB at 2,800 tonnes was, overall, an ecological benefit.

• while discharges would spread over a wider area of the benthic community, as their absolute quantum would stay the same, they would be more dilute and so have reduced impact. While I am not qualified to contradict Dr O’Toole and do not do so, I observe that her view would seem to require further elucidation. Logically, it does not necessarily follow that a more dilute discharge would have a lesser effect on the benthos: the question would seem to be as to whether, even though diluted, the discharge would nonetheless remain above the threshold concentration which would substantially or completely degrade the benthos below the larger area.

2058 Grounds 20/12/22 §9.112.21. 2059 Grounds 20/12/22 §9.16. 2060 Halpin v An Bord Pleanála [2019] IEHC 352, Simons J.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

ALAB – FAIR PROCEDURES

1355. SWI pleads2061 that ALAB’s procedures were not fair and equitable contrary to Article 9(4) of the Aarhus Convention. IFI make similar complaints. It is not apparent to me that the convention alters or amplifies the obligations of fair procedures imposed by Irish law and I will analyse this ground by reference to the Irish law of fair procedures. I have deferred addressing this issue to this point of the judgement as I think it best considered in light of a reasonably complete understanding of the procedures adopted by ALAB and the resultant sequence of events in the process and in light also my decisions on the specific substantive issues alleged to have arisen therefrom. There is some overlap here with the case made in bias, as allegations of bias rejected as internal to the process may fall to be reassessed by reference to the law of fair procedures and whether ALAB, as is alleged, bent over backwards to grant the licence. I have explained why I reject this last complaint. That leaves the question whether the objectors were heard – audi alteram partem. In my view, the foregoing content of this judgment and a consideration of the chronology of events allows ready disposal of that question. The Applicants assert that certain documents were not published until after ALAB made its decision. These were largely internal reports by its expert advisors to ALAB. It is true that, of the protagonists, MOWI got, as it were, the last word. But somebody had to – as authorities cited earlier – Haverty, Wexele and Southwood establish. Viewed through the prism of Haverty and Genport, I am happy on “the application of broad principles of commonsense and fair play” that the objectors were given ample opportunity to address the issues requiring ALAB’s consideration and “a reasonable general opportunity of making representations to it and of replying to submissions or representations by any other party”. Accordingly I reject the allegations of unfair procedures.

DECISION AS TO THE IMPUGNED AQUACULTURE LICENCE

1356. I will quash the impugned Aquaculture Licence for inadequate:

i. AA Screening of the risk of effects of seal scarers on seals of the SAC.

ii. EIA as to the risks of escape of salmon from the fish farm. This finding relates to necessity of reconsideration of bespeaking the DAFM reports on the 2014 farmed salmon escape in Bantry Bay, and to comprehensiveness of the EIA as it related to the specification and structural integrity of the cage installation.

iii. reasons for the conclusion that the proposed fish farm will not lead to a breach of WFD limits as to Dissolved Inorganic Nitrogen – specifically, reasons for reliance on RPS’s “typical” data in reaching that conclusion.

1357. In addition I will declare that ALAB delayed unreasonably as to AA Screening from the making of the Appeals in October 2015 to embarking on AA Screening after the Oral Hearing Report of November 2017. However I do not see that any further relief is required in that respect.

2061 SWI CG4.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1358. In this judgment and apart from the foregoing bases on which I will quash the impugned Aquaculture Licence, I have expressed various views intended to assist ALAB in any remitted decision on the Aquaculture Licence application in which its obligation will be to decide the application in accordance with the law as declared in this judgment as underlying certiorari but also in accordance with the law generally. For example, the question of EIA of salmon escape due specifically to seal predation may be considered to arise – though it is not a ground of certiorari as it was not pleaded. Indeed, as I have already observed, it would seem generally prudent that ALAB consider, on remittal of the matter to it, issues which were ventilated at trial but were not pleaded – as it seems unlikely that omission to plead them would be repeated in any judicial review of ALAB’s decision on remittal. Also, and as has been seen and even in respects in which I have not quashed it I have been critical of the form, layout and content of the Aquaculture Licence. There is every reason of good administration why ALAB would seek to rationalise the Licence with coherence in mind.

1359. I should mention specifically one ground I have rejected – that of objective bias. Notwithstanding my quashing their decision, it would be remiss not express my appreciation of and sympathy for the strenuous and determinedly professional efforts of ALAB in their consideration of this appeal. As the length of time it took to dispose of the Appeal, and indeed the length of the judgment it has taken to dispose of these proceedings, may illustrate, this was, on any view, a highly complex appeal involving myriad issues. Not merely that, it was highly controversial as part of a public controversy going back decades – to the point even of unusual opposition to ALAB’s decision, by way of judicial review, by another state agency, IFI. IFI’s was an opposition all the more unusual for its allegation of objective bias even to the extent of structural bias (though that is not, per se, a criticism of IFI). I confess to wondering whether it was ever practical – or fair to ALAB’s members – to expect that part-time ALAB members meeting once a month, attended by an apparently very small secretariat and only belatedly supplied with in-house expertise, could devote the focus and continuity of attention required to decide these appeals expeditiously. This Appeal would have taxed the full-time resources and full-time members of An Bord Pleanála had it fallen within its jurisdiction. Nonetheless, it is clear that the members of ALAB took their duties – including their duties of investigation – very seriously, to the point of prompting accusations of objective bias by excessive investigation – which accusations I have rejected.

1360. Though the vital duty bears re-emphasis that ALAB members are not delegates mandated to advance the interests of the constituencies from which they are drawn, there is no evidence that any of them failed to bring the required high standards of independence, integrity, neutrality and objectivity to bear on their decision-making.

1361. Given I will quash the Aquaculture Licence and remit it to ALAB for re-decision, I need not decide the Sweetman Applicants application (PS CG2) to quash the Minister’s decision of 18 September 2015 to grant an aquaculture licence.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

FORESHORE LICENCE

Foreshore Licence – Introduction & Co-Dependency of Foreshore & Aquaculture Licences

1362. A Foreshore Licence, in the case of aquaculture, is a licence only to use and occupy the site for the purpose of aquaculture. S.82 of the 1997 Act2062 requires the Minister, in considering a Foreshore Licence application in connection with the carrying on of aquaculture pursuant to an Aquaculture Licence, “to have regard to any decision of the licensing authority in relation to the Aquaculture Licence”.

1363. In June 2011, MOWI made a single application to the Minister on a single standard form provided by the Minister, in which it sought both an Aquaculture Licence and a Foreshore Licence. The Minister himself, having determined the Aquaculture Licence application and exercising a quasi-judicial function, determined the Foreshore Licence application in September 2015 – approving a draft Foreshore Licence which left blank the details of the Aquaculture Licence, presumably in anticipation of the possibility of appeal to ALAB.

1364. On its face and explicitly, the Foreshore Licence issued in 2022 is “for the purpose of the cultivation set out in Aquaculture Licence Number 1/2021 and shall remain in force for a maximum period of 10 (or) for so long as the Aquaculture Licence Number 1/2021 granted on 26 January 2022 is in force”. Yet the Foreshore Licence issued in 2022 explicitly records that it issued on foot of the Minister’s determination of September 2015 and it records no other Ministerial determination – nor has one been suggested. There is no evidence of any quasi-judicial consideration or determination in the Foreshore Licence application after that of 2015. The Minister never revisited the issue whether or in what terms he should grant the Foreshore Licence. In the events which occurred, the Minister’s civil servants simply deferred to 2022 sealing and issuing the Foreshore Licence pending the outcome of the Aquaculture Licence Appeal to ALAB.

1365. On ALAB’s determination of the Appeal in favour of granting the Aquaculture Licence, the Minister’s civil servants in 2022 sealed and issued the Foreshore Licence on foot explicitly of the Minister’s Foreshore Licence determination of 2015. It is clear that the Foreshore Licence issued in 2022 by way, not of any quasijudicial determination, but by way of administrative acts of civil servants effecting the Minister’s determination of 2015 by issuing the Foreshore Licence under his seal. Those civil servants amended the Foreshore Licence as issued, as compared to that drafted in 2015, by inserting details of the Aquaculture Licence of 2022 issued on foot of ALAB’s determination. But that was, as a matter of fact, a clerical act and involved no substantive reconsideration of the Foreshore Licence application. Nor has the Minister deposed otherwise.

2062 “The Minister, in considering an application for a lease or a licence under the Foreshore Acts, 1933 and 1992, which is sought in connection with

the carrying on of aquaculture pursuant to an aquaculture licence, shall have regard to any decision of the licensing authority in relation to the

aquaculture licence.” SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1366. IFI and the Sweetman Applicants seek to quash the Foreshore Licence on various grounds briefly described earlier in this judgment – notably and for example, a want of EIA. However, given I have decided to quash the Aquaculture Licence, it is of particular note that IFI, by CG7.1, seeks to quash the Foreshore Licence as contingent on the Aquaculture Licence. It also pleads that the Minister erred in granting the Foreshore Licence in 2022 having regard to his decision of 2015 rather than to ALAB’s impugned Aquaculture Licence decision of 29th June 2021. IFI’s written submissions make the simple case that that ALAB acted ultra vires its powers in making the Aquaculture licence Determination and if it is quashed, then Orders quashing the Aquaculture Licence and the Foreshore Licence should also issue. I agree that if the premise is correct the conclusion follows.

1367. The Minister’s challenge to the standing of IFI to pursue these proceedings was withdrawn.2063

1368. As to the relation of the Foreshore Licence and the Aquaculture Licence to each other, the following seem to me of note:

• the sealing of the Foreshore Licence in 2022 created a Foreshore Licence the continuation in force of which was explicitly contingent on the continuation in force of the 2022 Aquaculture Licence. Obviously in 2015 when deciding to issue the Foreshore Licence, the Minister had not had regard to ALAB’s 2022

decision to issue the 2022 Aquaculture Licence. But those who sealed the Foreshore Licence in 2022 as a merely clerical act clearly had at least some regard ALAB’s 2022 decision to issue the 2022 Aquaculture Licence as it prompted their action.

• the Foreshore Licence sealed in 2022 is explicitly conditional on compliance with the 2022 Aquaculture Licence – the obligations imposed by which differ from those of the 2015 draft Aquaculture Licence, to which the Minister had had regard in 2015.

• had ALAB refused to grant an Aquaculture Licence, a decision not to seal and issue the Foreshore Licence, revoking the Minister’s decision of 2015 to issue the Foreshore Licence, would have been required. This illustrates the obligation to have regard to ALAB’s 2022 decision.

• the civil servants’ recommendation2064 to the Minister to grant the Aquaculture Licence states, in my view correctly, “The continuing validity of each licence is contingent on the other licence remaining in force."

2063 Transcript Day 3 p55. 2064 Hodnett to Quinlan 31 July 2015.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1369. Notably, by its express terms, the Foreshore Licence is to “use and occupy” the Shot Head Salmon Farm Site “for the purpose of the cultivation set out in Aquaculture Licence Number 1/2021”. As to duration, the Foreshore Licence is for a maximum of 10 years from 26 January 2022, “.. provided for so long as” the Aquaculture Licence Number 1/2021 granted by ALAB “in respect of the same site for the purpose referred to in said Aquaculture Licence is in force.” The grammar is off but the meaning is clear: if the Aquaculture Licence ends, so too does the Foreshore Licence. The conditions to the Foreshore Licence are consistent with that view. They include the following:

“2. The Licensee shall use that part of the foreshore, the subject matter of this Licence, for the cultivation set out in Aquaculture Licence Number 1/2021 only, and for no other purpose whatsoever.

3. The Licensee shall comply fully with all terms and conditions of Aquaculture Licence Number 1/2021.

9. In the event of the breach, non-performance or non-observance by the Licensee of any of the conditions herein contained, the Minister may forthwith terminate this Licence without prior conditions herein contained, the Minister may forthwith terminate this Licence without prior notice to the Licensee.”

1370. The correlation, co-dependency and co-contingency of the Foreshore Licence and the Aquaculture Licence is echoed (even to the grammatical infelicity) in the latter. It provides that:

“This Aquaculture Licence shall remain in force for a maximum period of ten (10) years …provided for so long as the Foreshore Licence …… in respect of the same site for the purpose referred to is in force.”

Condition 10.1. provides:

“This Licence shall remain in force until 26 January 2032 and as long as the accompanying Foreshore Licence remains in force.”

1371. Which of the two licenses one should describe as substratum and which should be described as superstructure is perhaps a matter of taste: but the legal symbiosis essential to the continued existence of both is clear.

Foreshore Licence – Regard to Aquaculture Licence Decision

1372. No doubt there are good reasons why the Oireachtas decided that aquaculture activities should require both an Aquaculture Licence and a Foreshore Licence rather than a single licence encompassing the SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

concerns of both regimes. That decision having been made, it is easy to see why, as the Minister put it in submissions, his decision on each licence application was made “having regard to one-another”. But there is equally no doubt but that this dual licencing system, combined with the non-availability of appeal of the Foreshore Licence decision and the availability of de novo appeal of the Aquaculture Licence decision to ALAB, has created, at least in practice as followed in this case, an anachronistic dissonance as to such regard.

1373. Statutory obligations to “have regard” to matters are generally considered to rest lightly on the shoulders of decision-makers. However, as in all statutory interpretation, context is critical - Dunnes2065 and Heather Hill.2066 Obligations to “have regard” may impose greater or lesser burdens depending on context – Shadowmill2067 and, recently, Delaney v PIAB.2068 The context here includes the facts that,

• the very – the only – purpose of the Foreshore Licence is to facilitate the operation of the Aquaculture Licence.

• the link between the licences is emphasised in that, by s.1B of the Foreshore Act 1933, the Minister who decides foreshore licence applications as to aquaculture activities is not, as is the case as to foreshore licences generally, the Minister for the Environment, Community and Local Government – it is the Minister for Agriculture, Fisheries and Food, who also decides the corelative Aquaculture Licence application.

• the Minister asserts, though strictly I need not decide on the view I take, that whereas EIA is commonly done in foreshore licence applications generally, in foreshore licence applications as to aquaculture EIA is unnecessary as EIA is done in the corelative aquaculture licence. The suggested inference seems to be that in foreshore licence applications as to aquaculture the “regard” stipulated by s.82 of the 1997 Act includes regard for the EIA done by the aquaculture licensing authority.

In that context it seems to me to follow that the degree of “regard” required of the Minister by s.82 to the correlative Aquaculture Licence decision is appreciably greater than the light burden generally imposed by a “have regard” to obligation.

1374. But however heavy or light the obligation to “have regard” imposed by s.82, I can’t see it as anything but a mandatory obligation on the Minister – correctly, no argument was made to the contrary. Ceteris paribus in judicial review, non-regard to a matter to which regard is mandatory is ordinarily fatal to an impugned decision.

2065 Dunnes Stores v Revenue Commissioners [2019] IESC 50 §63.

2066 Heather Hill Management Company v An Bord Pleanála and Burkeway Homes [2022] IESC 43 ( [2022] 2 I.L.R.M. 313) §107.

2067 Shadowmill v ABP & Lilacstone [2023] IEHC 157, §343.

2068 Delaney v The Personal Injuries Assessment Board, The Judicial Council, Ireland and the Attorney General [2024] IESC 10, Collins J, §114.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1375. In 2015, the Minister fulfilled this obligation by having regard, in his capacity as decisionmaker as to the Foreshore Licence, to the decision he had made in his capacity as licensing authority as to the Aquaculture Licence. On the Minister’s argument, he had regard in 2015 to the EIA done therein and that sufficed as to his 2015 decision to issue the Foreshore Licence and that sufficiency persisted, despite the intervening Aquaculture Licence Appeal and the EIA done therein, to underpin the Foreshore Licence issued in 2022. For obvious purposive and logical reasons, that is a distinctly unattractive proposition. It would allow the Minister to ignore:

• the terms on which ALAB had issued the Aquaculture Licence – which may have differed appreciably from the Aquaculture Licence terms on which he had been willing to grant a foreshore licence in 2015.

• the most up to date EIA.

• the fact that the 2015 EIA having regard to which he had been willing to grant a foreshore licence in 2015 had not merely been superseded by a more recent EIA but had been deemed by the competent statutory authority, ALAB, to have been inadequate,

• a “decision of the licensing authority in relation to the Aquaculture Licence” within the meaning of s.82 of

the 1997 Act.

1376. As to the last point, counsel for the State, and in my view properly, agreed that the word “any” in

s.82 of the 1997 Act must be understood in the sense of “all or “every” decision of the licensing authority. By

s.3 of the 1997 Act, ALAB is an aquaculture “licensing authority”. It follows that its determination of the Appeal was a “decision of the licensing authority in relation to the aquaculture licence” within the meaning of s.82. It further follows, as a matter of the literal, plain, ordinary and natural meaning of the text of s.82 (these descriptors being synonyms identifying the first and most important recourse in statutory interpretation – Heather Hill2069), that ALAB’s decision is one to which the Minister must have regard in making his decision as to the Foreshore Licence. From that starting point, the question must then be asked whether interpretation of the text of s.82 in context requires a different interpretation and “the onus is on those contending that a statutory provision does not have the effect suggested by the plain meaning of the words chosen by the legislature to establish this”.2070

1377. Given the Minister’s submission that he had regard in 2015 to “the activity which was ultimately licensed” and leaving aside the differences between the Minister’s draft Aquaculture Licence and that issued on foot of ALAB’s determination of the Appeals, it bears some emphasis that, by s.82 of the 1997 Act, his obligation of regard was not to the activity licensed but was to “any decision of the licensing authority” – which phrase includes ALAB’s determination.

1378. The Minister’s s.82 obligation to have regard to ALAB’s decision is to have it “in considering” the Foreshore Licence application. That must require Ministerial regard to ALAB’s decision, including its EIA,

2069 Heather Hill §§106, 124, 214. 2070 Heather Hill §§214 – a case in which that onus was not met.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

before deciding the Foreshore Licence application. As it is impossible that in making his determination of September 2015 the Minister could have had regard to the ALAB’s decision in 2021 on the Appeal to issue the 2022 Aquaculture Licence and, on its face, the Foreshore Licence issued for the purposes of that 2022 Aquaculture Licence, it necessarily follows that, the Minister did not have regard to the decision of ALAB as licensing authority in relation to that very Aquaculture Licence.

1379. Indeed, by the time the Foreshore Licence issued, ALAB’s determination of the Appeal had, in June 2021 and by s.40(6) of the 1997 Act, immediately, annulled the Minister’s own 2015 decision as to the Aquaculture Licence to which he had had regard in making his Foreshore Licence decision in 2015. So, at the issue of the Foreshore Licence in 2022, ALAB’s decision on the Appeal was not merely a decision as to the Aquaculture Licence, it was the only decision as to the Aquaculture Licence and it was a decision to which the Minister had not had regard. And the 2015 Aquaculture Licence decision to which the Minister had had regard and on the basis of which the Foreshore Licence issued was a nullity by the time the Foreshore Licence issued.

1380. As I have found, the Minister did not “consider” or decide the Foreshore Licence application after 2015 – the Foreshore Licence issued in 2022 explicitly on foot of the 2015 determination to grant it and by way only of a non-quasi-judicial decision to issue it. One could argue therefore, as a matter of statutory interpretation that, as the Minister did not “consider” or decide the Foreshore Licence application after 2015 and as his obligation to have regard to a decision as to the Aquaculture Licence was only to do so when “considering” the Foreshore Licence application, no obligation on the Minister to have regard to ALAB’s

decision of 2021 as to the Aquaculture Licence arose.

1381. This observation prompts the following, admittedly somewhat overlapping, further observations:

i. First, the observation sits ill with the s.3 definition of “licensing authority” as including ALAB and the resultant implication that its decisions on appeals are amongst the decisions on Aquaculture Licences to which the Minister must have regard in considering correlative Foreshore Licences.

ii. Second, the observation sits ill with the possibility that the outcome of ALAB’s de novo determination of an Aquaculture Licence Appeal will be in terms of a grant appreciably different from that on which the Minister had decided and different in respects to which the Minister might sensibly have regard in deciding the Foreshore Licence application. In that respect, the observation does not enure to good administration.

While this is a general point, it is arguably illustrated in the present case by the increase of the cage numbers and the total cage footprint by 50% each as between the Minister’s draft Aquaculture Licence and ALAB’s determination of the Aquaculture Licence Appeal. But, as the point is a general one, I need not decide if the increase in this case is such that regard by the Minister to that difference is likely to have mattered in practice. Indeed, that seems a question of merits on which I SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

should not opine. What matters at this point is the vindication of the legal obligation to have that regard.

This general point is also illustrated by the conditions added by ALAB to the Aquaculture Licence and by the considerable differences between the Minister’s EIA of 2015 and ALAB’s of 2021. Again, I do not decide the validity of the Minister’s EIA of 2015. I merely note the differences. I note also the requirements that EIA be “comprehensive”, “as complete an assessment as possible” – Namur-Est2071 and Commission v Ireland2072 and the requirement of Article 8a(6) of the EIA Directive that the reasoned conclusion for a development consent be “still up to date” when making a decision as illustrative of the purposive good sense of requiring the Minister in deciding a Foreshore Licence Application to have regard to ALAB’s Aquaculture Licence decision.

iii. Third, the observation sits ill with the general scheme and purpose of the 1997 Act, which must have been that the most important “decision of the licensing authority in relation to the aquaculture licence” to which the Minister must have regard is the decision ultimately effective to grant the Aquaculture Licence, including regard to the terms in which the Aquaculture Licence is thereby granted. That most important decision is, in the event of an appeal, ALAB’s determination of that appeal. Regard by the Minister to a licence which had been superseded on appeal would not conform to a purposive interpretation of the relevant statutory provisions, inasmuch as that purpose clearly is that both the decision to grant and the terms of the Foreshore Licence will have been informed by regard to the decision to grant and the terms of Aquaculture Licence as actually effective.

In a general sense, the present case illustrates this point (though the point would be valid even absent such illustration given ALAB’s obligation to decide appeals de novo and hence, at least possibly, in terms differing significantly from those on which the Minister had decided.):

▪ ALAB’s decision is clearly informed by, records and considers in considerable detail, a licensing process, oral hearing, AA, EIA and expert reports, all of appreciable substantive content, and much of which postdated the Minister’s decision.

▪ The Aquaculture Licence on which ALAB decided is in terms appreciably different to those of the draft Aquaculture Licence on which the Minister decided. For example, the number of cages and their footprint has increased. Not least, Schedule 5 imposes new obligations2073 as to ▪ monitoring and recording bird populations ▪ underwater archaeology ▪ single bay management ▪ compliance with the Structural Design protocol ▪ limiting and applying EQSs to the use of Emamectin Benzoate ▪ well boat discharges

2071 Case C-463/20, Namur-Est Environnement ASBL v Région wallonne, Opinion of Kokott AG of 21 October 2021. 2072 Case C-50/09 Commission v Ireland, Judgment 3 March 2011.

2073 Unfortunately ALAB’s Aquaculture Licence is not paginated and Schedule 5 uses only bullet points, so it is a little inconvenient to identify the new

content. But it is on the 5th page of Schedule 5. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

iv. Fourth, it would be absurd, in the context of the clear symbiosis of the Foreshore Licence and the Aquaculture Licence, that in fulfilment of his duty under s.82 to have regard to any decision of the aquaculture licencing authority, the Minister would be obliged to limit that regard to his own earlier, draft and potentially very different, Aquaculture Licence rather than that which would ultimately take effect after a de novo appeal to ALAB – and for the very purpose of facilitation of which Aquaculture Licence, as issued by ALAB, the Foreshore Licence was granted after ALAB’s decision, its issue having been deferred to that very end. In my view, such an approach fails to have regard “for the proposition that statutory provisions should be read, where possible, so as to produce a workable and coherent interpretation, thereby avoiding interpretations which were … incongruous ….” – Waltham Abbey.2074

v. Fifth, the observation and those set out above raise the question whether the Minister could in this case, and if so should, having already decided the Foreshore Licence application in 2014, have reopened and reconsidered it having regard to ALAB’s determination of June 2021 and decided the Foreshore Licence application a second time or whether the Minister was functus officio as to the Foreshore Licence application from 2015. I consider that the statutory scheme should have more carefully and explicitly coordinated the consideration of foreshore licence applications for aquaculture with the existence of the system for appeals of aquaculture licence decision to ALAB.

That said, as to whether a decision-making authority such as the Minister in this case as to the

decision to issue a foreshore licence is functus officio is a matter of statutory interpretation and “When considering such a question, the focus of the courts tends to be on the statutory context of the particular power or duty in question.”2075 Also, by s.22 of the Interpretation Act 2005, “A power conferred by an enactment may be exercised from time to time as occasion requires.” In my view the Minister cannot have become functus officio in 2015 as there remained a duty on him, imposed by s.82 of the 1997 Act, to have regard to ALAB’s decision once made.

vi. In my view, assuming his decision on the Foreshore Licence application immediately having made his

decision on the Aquaculture Licence application, that he should have reconsidered the Foreshore Licence application having regard to ALAB’s decision is the clear logic of the statutory scheme.

vii. Sixth and especially given the fifth, the observation suggests that the better course for the Minister – to enable the required regard to any ALAB decision – is to defer determination of any Foreshore Licence application for an aquaculture project pending expiry of the time limit for appealing his Aquaculture Licence decision and thereafter pending ALAB’s decision of any such appeal, so that he can have the statutorily necessary regard to ALAB’s decision when considering the Foreshore Licence application. Whatever the practice may have been generally and in this case, I do not see that the Minister is obliged to decide the Foreshore Licence application immediately after deciding the

2074 Waltham Abbey Residents Association & Pembroke Road Association v An Bord Pleanála and Derryroe [2022] IESC 30; [2022] 2 I.L.R.M. 417 §43.

2075 Auburn et al, Judicial Review, Oxford 2013, §12-13.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Aquaculture Licence application or, in the event of appeal to ALAB, at any time before ALAB’ decision on the Aquaculture Licence application.

The Minister’s practice cannot determine statutory interpretation and cannot override his duty to have regard to ALAB’s decision in making his Foreshore Licence decision. Indeed, on reflection, it seems to me that the anachronism to which I earlier referred may be one not generated by the statutory scheme. It is merely an artifact of the practice adopted by the Minister – had he awaited ALAB’s decision before deciding the Foreshore Licence application no such difficulty would have arisen. On such a view, the Minister’s decision on the Foreshore licence application was premature.

It can hardly be argued that a decision-maker can evade a statutory obligation to have regard to a matter (here, ALAB’s decision) by making its decision before that matter has come into being – at least where the prospect of its coming into being is in view and the decision-maker is not bound by any time-limit within which to make its decision. Here the prospect of an appeal to ALAB was clearly in view when the Minister decided the Foreshore Licence application – the Minister’s presumed awareness of the scheme of the 1997 Act sufficed to bring it into view. That the prospect of an appeal to ALAB was in fact in view is apparent from the express terms of the notice published in Iris Oifigiúil in September 2015 and from the deferral of the sealing and issuing of the Foreshore Licence.

viii. Seventh, the Minister’s argument that his having regard to ALAB’s decision would “lead to uncertainty for putative foreshore licensees who would be subject to having their positive licence determinations unilaterally called into question by the Minister at any time prior to their grant” is illusory when one considers that in practice, and as the licenses are legally co-dependent for their existence, the posited substantive uncertainty necessarily subsists pending ALAB’s decision in any event.

That is graphically illustrated by the facts of this case, in which the Minister made his Foreshore Licence decision in 2015 but deferred actually issuing the Foreshore Licence until after ALAB’s

decision of 2021 and indeed until the resultant issuing of the Aquaculture Licence.

And, in any event, the detriment of any such uncertainty could not outweigh a statutory obligation to have regard in making his Foreshore Licence decision to relevant matter – in this case ALAB’s

decision.

1382. Whether one views the matter on the basis that the Minister should have deferred his Foreshore Licence decision until after ALAB’s decision or, having made his Foreshore Licence decision was not functus officio and was entitled – indeed obliged – to revisit it after ALAB’s decision, the practical outcome is the same. The Minister should have decided the Foreshore Licence application, whether for the first or the second time, after, and having regard to, ALAB’s 2021 decision.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Foreshore Licence – Decision

1383. On a very mechanistic and technical view and by the terms of the Foreshore Licence itself and regardless of any remedy in judicial review, certiorari of the Aquaculture Licence, without more, terminates the Foreshore Licence. By its express terms, the Foreshore Licence subsists only “so long as” the Aquaculture Licence “is in force.” On this view and as, on certiorari, the Aquaculture Licence is annulled, the Foreshore Licence is no longer in force and to order quashing it is required for that effect to occur. On this mechanistic and technical view, and as to the Foreshore Licence, MOWI would have to apply again from scratch.

1384. But that view is unnecessarily mechanistic and technical – to no purposive end. It ignores that while certiorari is a “remedy of annulment” (Waltham Abbey2076) such that the quashed decision no longer exists,

• remedies in judicial review are flexible with a view to achieving their overriding aim to do no more and no less than undoing as clinically as possible the consequences of a fault in the impugned decision by returning the situation to where it would have been but for the faulty decision – Kelly.2077

• certiorari is generally accompanied in modern practice, where possible, by remittal to re-decision. In that event, ALAB’s decision on remittal would o be a decision of the licensing authority to which, by s.82 of the 1997 Act, the Minister must have regard in making any associated Foreshore Licence decision. o enable the Minister to reconsider on remittal a quashed Foreshore Licence having regard to ALAB’s

decision on remittal.

1385. The Minister’s own logic in deferring issuing the Foreshore Licence on foot of his decision in 2015 awaiting, and until after, ALAB’s decision of 2021 is a logic which presumes the necessity of co-existence – the co-contingency – of the two licenses. Indeed, as I have observed, it is a logic which suggests he should have deferred not merely issuing the Foreshore Licence but the decision to issue the Foreshore Licence. It is a logic which suggests that certiorari of the Aquaculture Licence should, as it were, “wind the clock back” by reinstatement of that deferral.

1386. I agree with IFI (CG7.1) that the proper course is to quash the Foreshore Licence as contingent on the Aquaculture Licence and also as the Minister erred in granting the Foreshore Licence in 2022 having regard to his decision of 2015 rather than to ALAB’s impugned Aquaculture Licence decision of 29th June 2021.

2076 Waltham Abbey Residents Association v An Bord Pleanála [2022] IESC 30 ([2022] 2 I.L.R.M. 417), §53. 2077 Kelly v Minister for Agriculture [2023] 1 IR 38 citing Tristor v Minister for the Environment [2010] IEHC 454.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1387. I reject the Minister’s alternative argument that the terms of the 2022 Foreshore Licence, in referring to the 2022 Aquaculture Licence, disclose regard to it for the purposes of s.82 of the 1997 Act. However light the burden of regard it is not satisfied by the merely clerical act of inserting the relevant references in the 2022 Foreshore Licence – a comparison of this clerical act to the record of the Minister’s

decision-making process in 2015 makes that quite clear.

1388. Further, the Minister’s logic to which I have referred informs the view I take of the Minister’s objection that IFI is out of time to impugn the Ministerial Foreshore Licence decision of 2015. In one sense, this is an extraordinary case: ordinarily in licensing processes the quasi-judicial decision to grant a licence is followed in short order – perhaps a day or two or following a brief time lapse pending expiry of a time limit for appeal – by the administrative act of issuing the licence. Here the time lapse was of almost 6.5 years. Importantly, that time lapse reflected the entirely contingent nature of the decision to grant the Foreshore Licence – the contingency of the Foreshore Licence’s very existence being the grant of the Aquaculture Licence. Viewing foreshore licencing and aquaculture licensing as distinct statutory processes, it seems in a sense incorrect to regard the Ministerial Foreshore Licence decision of 2015 as an intermediate decision which need not be challenged pending the final decision by way of either the Aquaculture Licence or the issuing of the Foreshore Licence. Nonetheless, even assuming that they are distinct statutory processes, given their co-contingency, the analogy with an intermediate decision seems apt and the logic applicable seems to me the same. In any event, it seems to me that the peculiar statutory scheme, peculiar to aquaculture licences and foreshore licences for aquaculture activities, should be interpreted as one. Put another way, there is a real sense in which judicial review of the Foreshore Licence decision of 2015 would have been premature pending ALAB’s decision – just as the Minister clearly, and in my view correctly, regarded the issuing of the Foreshore Licence as premature pending ALAB’s decision.

1389. It must be said that,

• the Minister’s 2015 notice in Iris Oifigiúil of his decision to grant both licences advised both of the remedy

of appeal of the Aquaculture Licence decision and, as required by s.21A of the Foreshore Act 1933, of the possibility of judicial review of the Foreshore Licence decision.

• ordinarily in licensing, and like matters in which the decision is followed by a formal document of grant or the like, and at least where the ground for judicial review relates to the decision-making process as it was effected prior to and in the decision, time for making an application for leave to seek judicial review runs from the decision or its publication, not the grant. For a recent example, see Gardiner2078 as to judicial review of planning decisions.

2078 Gardiner v Mayo County Council [2024] IEHC 5.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1390. However, an important feature of Gardiner, is missing here. In Gardiner, Simons J said that the subsequent step of making the “grant” of planning permission is purely mechanical and perfunctory. He said that the “planning authority does not enjoy any discretion: it cannot refuse to make the grant, nor can it change any aspect of the underlying decision to grant. Rather, the planning authority is required to faithfully transpose the terms and conditions of the decision previously made into a formal grant of planning permission.” That is not entirely so here – notably and though I have described the issuing of the Foreshore Licence as an administrative as opposed to a quasi-judicial act, whether the grant of the Foreshore Licence was made at all was contingent on ALAB’s decision. The Foreshore Licence would not have issued if ALAB had refused to grant an aquaculture licence. And the substantive content of the Foreshore Licence condition as to compliance with the Aquaculture Licence is contingent on the terms of ALAB’s decision and the terms of the Aquaculture Licence issued on foot of it: which terms may well, and here did, differ from those contemplated in the Minister’s Foreshore Licence decision.

1391. Indeed, it is notable that Simons J contemplated that a ground on which to challenge the grant of permission, as opposed to the decision to grant it, “might be where a planning authority purported to attach an additional condition to the planning permission, over and above those notified as part of the decision to grant. In such a scenario, the grant could successfully be challenged by way of judicial review.” The substantive addition of conditions, via the 2022 Aquaculture Licence is in reality what has happened here. – The statutory scheme here is different from that as to planning permission and a different approach seems to me to be required to the running of time for judicial review.

1392. Also, and unlike, for example, s.50(6) and (8) PDA 2000,2079 the Foreshore Act 1933 contains no statutory or truncated time limits within which leave to seek judicial review must be sought.

1393. In the particular circumstances of the interaction of the 1933 and 1997 Acts as to aquaculture, it would, at least ordinarily, have been wasteful litigation to challenge the Foreshore Licence decision while awaiting the outcome of the alternative statutory remedy by way of appeal to ALAB which, by any refusal of an Aquaculture Licence, would have doomed the Foreshore Licence. Further, it would have been wasteful even in the event of ALAB’s grant of an Aquaculture Licence given, as I find, the Minister was obliged by s.82 of the 1997 Act2080 to have regard to ALAB’s decision before issuing a Foreshore Licence. If it comes to a choice between applying the law as to when the Minister becomes functus officio (for example on publication of his decision in Iris Oifigiúil in 2015) and ensuring that the Minister has regard to that to which he must, by statute, have regard in making his decision, I would choose the latter. However it appears to me that, in truth, no such choice arises. As I have said, statute prevails over any general doctrine of functus officio and in my view the relevant provisions of the 1933 and 1997 Acts must, given the co-dependency of

2079 Planning and Development Act 2000.

2080 “The Minister, in considering an application for a lease or a licence under the Foreshore Acts, 1933 and 1992, which is sought in connection with

the carrying on of aquaculture pursuant to an aquaculture licence, shall have regard to any decision of the licensing authority in relation to the

aquaculture licence.” SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

an Aquaculture Licence and its accompanying Foreshore Licence, be read together as requiring that the Minister in considering and making his Foreshore Licence decision, have regard to ALAB’s aquaculture licence decision.

1394. In the circumstance of the peculiar statutory scheme in question, I hold that time ran, for seeking leave to seek judicial review of the Foreshore Licence, from its issue in 2022.

1395. In my view,

• in light of the co-contingency of the continuance in force of the Aquaculture Licence and the Foreshore Licence and in light of the quashing of the Aquaculture Licence decision, Ministerial regard to which was a statutory requirement of granting the Foreshore Licence, the Foreshore Licence must be quashed on ground IFI CG7.1 as not in accordance with law.

• the Foreshore Licence application must be quashed also on ground PS CG9F for the Minister’s failure, in breach of s.82 of the 1997 Act, to have regard to ALAB’s determination of the Aquaculture Licence appeal.

1396. As to the latter, the impossibility of such regard was brought about not by the Minister’s compliance with any statutory requirement to make his Foreshore Licence decision in 2015 – he could have waited until after ALAB’s decision to do so and doing so would not have appreciably delayed matters. This reality is illustrated by the logic and the fact of the deferral of the sealing and issuing of the Foreshore Licence pending ALAB’s decision as to the Aquaculture Licence. In any event, I take the view that the Minister was not functus officio on foot of his 2015 Foreshore Licence decision and was obliged to revisit it after ALAB’s

decision.

1397. Accordingly I need not decide any other grounds of challenge to the Minister’s decision on the Foreshore licence. No doubt the Minister will consider those grounds in making any further decision he may be called upon to make.

1398. I am of the provisional view that the Foreshore Licence application should be remitted to the re-

decision of the Minister – which decision should await and be made having regard to such decision as ALAB may make on the remitted Aquaculture Licence application. I will hear the parties as to the form of order required in this regard.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

Is EIA of Foreshore Licence applications for Aquaculture Required?

1399. Given I have decided to quash the Foreshore Licence, as contingent on the quashed Aquaculture Licence it is now not is necessary to decide any other challenge to the validity of the Foreshore Licence – including whether EIA was required distinctly in the Foreshore Licence Application as well as in the Aquaculture Licence application. The Minister says not – though even were that so, his regard for ALAB’s

decision would have to have included regard for the EIA determination recorded thereon. But as the question will arise on remittal and is raised in these proceedings, I think I should decide it.

Introduction & Purposive Analysis

1400. The Minister’s 2015 EIA was explicitly performed in accordance with the Aquaculture EIA Regulations 20122081 and cited the Aquaculture Licence Regulations 1998 for the obligation to do EIA.2082 Though the EIA contextually referred to the Foreshore Licence process it was done in the Aquaculture Licence process.

1401. The Minister says that EIA of Foreshore Licence applications for salmon farms is not required by law. He says that, as EIA of such foreshore licence applications is legally unnecessary, it does not lie for the Applicants to impugn the grant of the Foreshore Licence in 2022 on the basis of a failure in the Foreshore Licence Application to conduct EIA of the larger iteration of the Salmon Farm permitted by the Aquaculture Licence.

1402. Foreshore licenses come in many sorts and licence many different things. By s.3 of the 1933 Act they can authorise a person “to place any material or to place or erect any articles, things, structures, or works in or on such foreshore, to remove any beach material from, or disturb any beach material in, such foreshore, to get and take any minerals in such foreshore and not more than thirty feet below the surface thereof, or to use or occupy such foreshore for any purpose”. A given foreshore licence may authorise only some or one of those possibilities.

1403. It is thus clear that at least some foreshore licences will constitute a development consent2083 for a project of a class within the relevant EIA legislation such that EIA, or at least EIA screening will be required. S.13A of the 1933 Act provides accordingly.

2081 S.I. No. 410 of 2012 – European Union (Environmental Impact Assessment) (Aquaculture) Regulations 2012. 2082 S.I. No. 236 of 1998 – Aquaculture (Licence Application) Regulations, 1998, Art 4A. 2083 Within the meaning of the EIA Directive.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1404. At least in broad terms, s.3 of the 1933 Act might be considered to allow Foreshore Licences to issue for aquaculture without any need for another form of licence. But that is not the statutory architecture adopted. The 1997 Act prescribes Aquaculture Licences and a detailed statutory and regulatory scheme for their grant or refusal. It is common case – and in my view clear, that EIA is required for all aquaculture licences for salmon farms: such licences are clearly development consents.2084 They permit and regulate development – what the aquaculture licensee may do by way of aquaculture and do by way of development of the site to enable aquaculture.

1405. The Foreshore Licence, in the case of aquaculture, is a licence to use and occupy the site for the purpose only of licenced aquaculture. Such a foreshore licence is analogous to a developer’s purchase of land and the aquaculture is analogous to the grant of planning permission which enables a developer to develop such land. Leases and licences of land very often include covenants or conditions prohibiting or prescribing particular uses of the land in question. It is, rightly, never suggested that a lease or licence by the State of lands for development is a development consent requiring EIA. Nor, it seems to me, is there any good reason why such a suggestion should be upheld as to licenses to use foreshore for purposes licensed by an aquaculture licence.

1406. The State say and I accept that nothing by way of development or which is likely to have a significant effect on the environment is even arguably permitted by the Foreshore Licence which is not in any event permitted by the Aquaculture Licence and was subjected to EIA in the aquaculture licence process. Thereby the objective of the EIA Directive, as expressed in its Article 2(1), to ensure that “before development consent is given, projects likely to have significant effects on the environment by virtue, inter alia, of their nature, size or location are made subject to a requirement for development consent and an assessment with regard to their effects on the environment” is fully achieved.

1407. Accordingly, EIA in the Foreshore Licence would not advance or enhance the attainment of that objective and would, by reference to that objective, be in substance unnecessary, duplicatory and wasteful. It would be an exercise in mere formalism of the kind deprecated by Sharpston AG in Boxus2085 (“The EIA Directive is not about formalism. It is concerned with providing effective EIAs for all major projects”), by Kokott AG in Nomarchiaki2086 and by McGovern J in Ó Gríanna #2.2087 McGovern J said: “The principle of effectiveness is not a mandate for construing the Directive in the most onerous manner possible. This involves the courts being astute to ensure the objectives of the Directive are met but not in an overly pedantic way.”

2084 Within the meaning of the EIA Directive.

2085 Joined Cases C-128/09, C-129/09, C-130/09, C-131/09, C-134/09 and C-135/09 Boxus et al v Région Wallonne, Opinion of Sharpston AG, 19 May 2011.

2086 Case C-43/10 Nomarchiaki Aftodioikisi Aitoloakarnanias – Opinion of Advocate General Kokott 13 October 2011.

2087 Ó Gríanna v An Bord Pleanála (No. 2) [2017] IEHC 7. Also Kelly, Eoin v An Bord Pleanála [2019] IEHC 84 and Fitzpatrick v An Bord Pleanála [2017] IEHC

585. SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1408. The foregoing analyses the question whether the Foreshore Licence required EIA from a purposive EU law point of view in light of the purpose of the EIA Directive. From that point of view, EIA of a foreshore licence application is not required where EIA is done in the associated aquaculture licence application as long as what is permitted by the former does not differ significantly from what is permitted by the latter. Nonetheless, it remains to consider the domestic statutory position.

S.13A(6) of the 1933 Act

1409. S.13A was inserted in the 1933 Act in 19902088 explicitly to transpose, as appliable to foreshore

licences, the State’s obligations under the EIA Directive. It requires EIA of “relevant applications” – which

S.13A(5) defines, exhaustively, as being authorisations issued under the 1933 Act.2089 Accordingly no question arises, on that definition, of aquaculture licences under the 1997 Act being “relevant applications” for the purpose of S.13A of the 1933 Act. In that context S.13A(6) is surprising:

“(6) In this section ‘relevant application’ does not include an application for an aquaculture licence (within the meaning of the Fisheries (Amendment) Act 1997) that is accompanied by an environmental impact assessment report.” 2090

1410. The default rule of statutory interpretation – the “first and most important port of call”2091 – is that

words are given their literal meaning in the context in which they appear – or, is the same thing, their natural and ordinary meaning. Indeed, “context is critical” – Heather Hill.2092 Taken literally, s.13A(6) is pointless and has no effect: it excludes a possibility that never arises. This put the State in the odd position of arguing that S.13A(6) is, in its literal terms absurd. Odd or not, and while I bear in mind the reluctance to

infer absurdity in statute (AC2093) the State’s position is correct. One may add that taken literally, s.13A(6) is

in breach of the presumption in statutory interpretation against wasted or vain words, or surplusage.2094

1411. The State argues that, to rescue s.13A(6) from absurdity, additional words must be read into it. While they did not suggest the precise form of words set out below, it expresses their argument:

2088 Inserted (1.02.1990) by European Communities (Environmental Impact Assessment) Regulations 1989 (S.I. No. 349 of 1989), reg. 13(c), in effect as

per reg. 2(1). 2089

(5) …. “relevant application” means, as the case may be— an application (a) …. for a lease under section 2 of this Act, (b) … for a licence under section

3 of this Act, (c) … for approval under section 10 of this Act for maps, plans, and specifications for erection of structures on the foreshore, (d) … for …

consent under section 13 of this Act for the deposit of material on the foreshore.” 2090 It formerly read “environmental impact statement”. 2091 People (DPP) v Brown [2018] IESC 67.

2092 Heather Hill Management Company CLG v An Bord Pleanála and Burkeway Homes Ltd [2022] 2 I.L.R.M. 313 (Supreme Court).

2093 The People (Director of Public Prosecutions) v AC [2022] 2 IR 49. 2094 E.g. KN v Minister for Justice [2020] IESC 32.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

“(6) In this section ‘relevant application’ does not include an application of a type described in subsection (5) made in connection with an application for an aquaculture licence (within the meaning of the Fisheries (Amendment) Act 1997) that is accompanied by an environmental impact assessment report.” 2095

Principles of Statutory Interpretation

1412. Traditionally it has been considered that the purpose – the focus – of all statutory interpretation is to discern the intent of the legislature as expressed by the words used in the statute. Such intent has been

described as “what the legislature meant: or as it is put, what is the will of Parliament” and “the intention of

the law maker”. This characterisation was unanimously approved by the Supreme Court in 2019 in Dunnes and 2020 in Bookfinders2096 as having “been said time and time again”. Two years later, in Heather Hill, the Supreme Court unanimously approved four basic propositions relevant specifically to the need to avoid correct purposive and contextual interpretation of statutory text slipping into impermissible judicial mobilisation of their “instinct as to what the legislators would have said had they considered the problem at hand”. I read this as an insistence on the objective nature of statutory interpretation. The first such principle is that, “ “legislative intent” as used to describe the object of this interpretative exercise is a misnomer: a court cannot peer into minds of parliamentarians when they enacted legislation and as the decision of this court in Crilly v Farrington [2001] 3 I.R. 251; [2002] 1 I.L.R.M. 161 emphatically declares, their subjective intent is not relevant to construction. Even if that subjective intent could be ascertained and admitted, the purpose of individual parliamentarians can never be reliably attributed to a collective assembly whose members may act with differing intentions and objects.”

1413. However, the third principle is that the words of a statute are given primacy – primacy as the best guide to “the result the Oireachtas wanted to bring about. The importance of this proposition and the reason

for it, cannot be overstated.” At first blush it seems difficult to distinguish a search for the “intent” of the

Oireachtas from a vital search to identify what it “wanted to bring about”. But the answer is clear from the first principle. What the Supreme Court identified as a misnomer was the concept of subjective intent – the actual mental intent of the lawmakers in person and taken as a whole - as opposed to the intent objectively discernible in the words they used in the statute. The former is unlikely to even exist as parliamentarians may differ inter se and in any event, if the two don’t coincide - if the lawmakers fail to accurately say what they subjectively meant - the intent objectively discernible in the words of the statute prevails. And, as the word misnomer suggests, it seems to me that the issue is ultimately one of nomenclature rather than substance. Thus understood, it seems to me that Heather Hill does not represent a departure from Dunnes and Bookfinders as to the specific use of the phrase legislative intent. Also, of course, there is the fact that

2095 Supplied words underlined.

2096 Dunnes Stores v Revenue Commissioners [2019] IESC 50 §62; Bookfinders v Revenue Commissioners [2020] IESC 60 – see generally, including §53.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

the Interpretation Act 2005 repeatedly deploys the concept of intent.2097 Notably, s.5 of that Act, where it applies, requires “a construction that reflects the plain intention of the Oireachtas or parliament concerned, as the case may be, where that intention can be ascertained from the Act as a whole.”

1414. It seems to me therefore that the misnomer identified in Heather Hill relates to subjective intent and it may remain permissible to refer to legislative intent as long as it is clearly understood that:

• by that phrase is meant the intent objectively discernible from the words of the statute understood in their context. That intent is attributed to the legislature regardless whether it accords with a view of what the actual subjective intent of the legislature was.

• context, for this purpose is identified in the cases – here it suffices to note that it includes the whole statute in which the provision being interpreted appears and any other statutory provision identified as to be construed with it. More pithily put, the context includes the statutory scheme.

• that objectively discernible intent is discerned by application of the transparent, coherent and objectively ascertainable principles, canons, maxims, rules, presumptions and techniques2098 which the law allows.

• As relevant in this case, amongst those presumptions are that: o the legislature did not intend absurdity – “a meaning so plainly absurd that it could not have been intended”2099 (Note again here the language of intention and the high bar for a conclusion of absurdity). o legislation is presumed to have some object in view which it is sought to achieve.2100

1415. The following appears in the recent case of Shadowmill:2101

“219. …. principles of statutory interpretation are not strict rules but rather are tools to aid discerning the true intent of the Oireachtas – “the focus of all interpretive exercises is to find out what the legislature meant”. In Crilly v Farrington2102 Murray J referred to the “difficulties inherent in statutory construction” and to canons, principles and presumptions of construction as “efficient and neutral aids to the interpretation of statutes and .. not some sort of standard formulae automatically shaping the result of an interpretative issue ….. The use of canons or principles of construction, or any one or combination of them in a given case depends on a variety of factors and their interplay - the complexity or clarity of the text in issue, whether applicable precedents exist, whether there are fundamental principles in issue or constitutional considerations - one could go on. The point of

2097 Ss. 3, 4, 5, 20.

2098 Which concepts overlap considerably: for example little usually turns on whether a proposition is a maxim or a presumption.

2099 Heather Hill §111. 2100 Bookfinders §56. 2101 Shadowmill Limited v An Bord Pleanála [2023] IEHC 157. 2102 Crilly v T. & J. Farrington Ltd. [2001] 3 IR 251, §287.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

departure for the court is always the actual text of the statute to be interpreted and it is a matter of judicial appreciation, in the light of submissions from counsel, which canons or method of interpretation are appropriate to the nature of the problem which presents itself in the particular case.”

220. Indeed, it has been authoritatively said by the Supreme Court that “It can be very difficult to try and identify a common thread which can both coherently and intelligibly explain why, in any given case one particular rule rather than another has been applied, and why in a similar case the opposite has also occurred.”2103

1416. So, while the second principle stated in Heather Hill

• identifies the objective approach to discernment of statutory intent as the price of avoiding judicial construction of statutes “by reference to their individual assessments of what they think parliament ought sensibly to have wished to achieve”2104 and,

• vitally, effects that approach by “the application of transparent, coherent and objectively ascertainable principles”, nonetheless, it is recognised that, as between different instances of statutory interpretation, the application of those principles in the real world of actual cases may

• require some “judicial appreciation, (as to) …. which canons or method of interpretation are appropriate to the nature of the problem”.

• on occasion necessarily fall short of rigorous coherence.

1417. In Grange2105 Haughton J cited Dodd2106 to the effect that “An interpretation which renders other statutory phrases or provisions meaningless or surplusage may indicate that the interpretation is not the intended one.” In O'Sullivan2107 the Supreme Court cited the presumption that words are not used in a statute without a meaning and, accordingly, effect must be given, if possible, to all the words used. Presumptions applicable to allegedly surplus words a fortiori apply to allegedly surplus entire subsections. So, s.13A(6) should be given meaning and effect if possible. I agree that s.13A(6) is, in literal terms absurd as conveying a meaning which can have no effect and thus it fails to express the objectively discerned intent of the legislature.

1418. S.5 of the Interpretation Act 2005 applies in interpreting any non-penal statutory provision that is

“obscure or ambiguous” or that “on a literal interpretation would be absurd or would fail to reflect the plain

intention of …. the Oireachtas”. Such a provision “shall be given a construction that reflects the plain

2103 Dunnes Stores v Revenue Commissioners [2019] IESC 50; Bookfinders v Revenue Commissioners [2020] IESC 60. 2104 Citing Finlay C.J. in McGrath v McDermott [1988] I.R. 258 at 276. 2105 Grange v The Information Commissioner [2022] IECA 153. 2106 Statutory Interpretation in Ireland (2008 Edition) §5.28.

2107 O'Sullivan v Ireland [2019] IESC 33 citing Goulding Chemicals Limited v Bolger [1977] I.R. 211 per O’Higgins C.J., §226.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

intention of the Oireachtas …... where that intention can be ascertained from the Act as a whole.” Given the absurdity of s.13A(6), s.5 of the Interpretation Act 2005 requires that, if possible, it be given such an interpretation. It is important to caution however that s.5 requires that the intention of the Oireachtas be ascertained, that it be “plain” and that it be ascertained from the Act as a whole. O’Donnell J in Bookfinders2108 cautioned against an over-liberal understanding of s.5 and emphasised that its technique, while purposive, is rooted in the statutory text.

1419. Dodd2109 under the heading “Reading In Words And Phrases” states a general rule that it is not permissible to read words into an enactment and that “For hard-line literalists, adding words to a text may always be unacceptable and amount to amendment.” He cites Thompson2110 for Lord Mersey’s warning that ‘[i]t is a strong thing to read into an Act of Parliament words which are not there, and in the absence of clear necessity it is a wrong thing to do’. However, Dodd also asserts that “Reading words into the text can be interpretation and not amendment” and identifies exceptions to the rule. He says that “Permissible reading is, in most cases, no more than giving effect to the intention of the legislature evident by implication.”2111 Relevant here is Dodd’s view that “expressions and even whole provisions may in effect be read in, in order to render an enactment effective” or avoid a legislative scheme failing – citing Finlay CJ in McGrath2112 to the effect that ‘[i]n rare or limited circumstances words or phrases may be implied into statutory provisions solely for the purpose of making them effective to achieve their expressly avowed objective’.

1420. This statement is illuminated by its counterpoint in the immediately preceding statement of Finlay CJ: “The courts have not got a function to add to or delete from express statutory provisions so as to achieve objectives which to the courts appear desirable.” Clearly, Finlay CJ was guarding against the courts’ advancing objectives they might consider desirable but which are not apparent in the statute. It therefore seems to me that the word “expressly”, as used by Finlay CJ, was not intended to exclude the application of the technique in cases where the intention of the Oireachtas is clear but implicit rather than express. What is necessary is that the

• “objective” of the statutory provision be reliably discernible.

• additional word or phrase be necessary to effecting that objective.

• additional word or phrase be itself reliably discernible.

• foregoing three criteria be applied with great caution.

1421. It is also of note that, in McGrath, the Revenue had set themselves the Sisyphean task of implying a new subclause or sub-section into a tax act - in effect introducing a new doctrine of fiscal nullity into tax

2108 Bookfinders Ltd v Revenue Commissioners [2020] IESC 60 (Supreme Court, O'Donnell J, 29 September 2020). 2109 Statutory Interpretation in Ireland [12.14] et seq.

2110 Thompson v Goold and Co [1910] AC 409 at 420 – cited often and as recently in England as a “celebrated dictum” in R(Akinsanya) v Secretary Of State

For The Home Department [2021] 1 WLR 5454.

2111 Citing Spigelman CJ, in R v Young (1999) 46 NSWLR 681: ‘The reference in the authorities to courts “supplying omitted words” should be understood

as a means of expressing the court’s conclusion with clarity, rather than as a description of the actual reasoning process which the court has conducted.’

2112 McGrath v McDermott [1988] IR 258 at §276.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

avoidance law. In argument, the Revenue appear to have accepted that such a course was not necessary but merely desirable.2113 So, McGrath seems not to have been a case which mapped the limits of the power of interpolation.

1422. Another exception identified by Dodd permits interpolation where absolutely necessary to avoid an absurd, unworkable, unintelligible, totally unreasonable or unconstitutional result – he cites H v H2114 and Re the Employment Equality Bill 19962115 for this necessity exception. The Supreme Court held in the latter case:

“That where the intention of the legislature was plain on the construction of the statute as a whole, the Court was entitled to transpose, interpolate or otherwise alter words, in order to give effect to the statute and avoid manifest absurdity or injustice.”

In the former, the Supreme Court upheld “one of the fundamental rules of interpretation i.e., that words may not be interpolated into a statute unless it is absolutely necessary to do so in order to render it intelligible or to prevent it having an absurd or wholly unreasonable meaning or effect.”

1423. The generally authoritative Bennion2116 asserts a presumption that the legislature intends the court to apply a construction which rectifies any error in the drafting of the enactment, where required in order to give effect to the legislative intent – there are occasions when the language of the legislature must be modified, in order to avoid inconsistency with its manifest intent.The power is confined to plain cases of drafting mistakes. The courts are ever mindful that their constitutional role in this field is interpretative only. They must abstain from any course which might have the appearance of judicial legislation. I observe that, though the constitutional contexts are very different, in this aspect the anxiety is common to both jurisdictions.

1424. In Inco Europe,2117 a provision intended to effect merely ‘consequential’ amendments in fact on a literal reading made a significant amendment, excluding certain rights of appeal. The drafter had ‘slipped up’ and Lord Nicholls read words into the section to avoid that unintended result. Lord Nicholls, for a unanimous House of Lords, said “The court must be able to correct obvious drafting errors. In suitable cases, in discharging its interpretative function the court will add words, or omit words or substitute words.” As to what might be a suitable case, he said: “A statute is expressed in language approved and enacted by the

2113 Finlay CJ said: “In the course of the submissions such a necessity was denied but instead it was contended that the real, as distinct from what is

described as the artificial, nature of the transactions should be looked at by the court, and that if they were, the section could not apply to them.”

2114 H v H [1978] IR 138 at 146 – albeit a case in which the exception was not applied. 2115 Re the Employment Equality Bill 1996 [1997] 2 IR 321.

2116 Bennion, Bailey and Norbury on Statutory Interpretation §15.1.et seq citing R v Moore [1995] 4 All ER 843, §850; R (on the application of Zenovics) v

Secretary of State for the Home Department [2002] EWCA Civ 273 at [1]; OTC International AG v Perfect Recovery Ltd [2009] 3 HKLRD 13 at [48], Miller

v Salomons (1852) 7 Exch 475 at 553 and Inco Europe Ltd v First Choice Distribution [2000] 1 WLR 586.

2117 Inco Europe Ltd v First Choice Distribution [2000] 1 WLR 586. The State properly draws to my attention a recent application of Inco Europe in

Northern Ireland – FN.2117 SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

legislature. So the courts exercise considerable caution before adding or omitting or substituting words.” Bennion cites Lord Nicholls for the relevant test, which, Bennion states “has been widely applied in the common law world”. To construe an Act so as to correct a drafting error, the court must be “abundantly sure” of the following three matters:

“(1) the intended purpose of the provision in question;

(2) that the drafter and the legislature inadvertently failed to give effect to that purpose in that provision;

(3) the substance of the provision the legislature would have made (though not necessarily the precise words it would have used) had the error in the Bill been noticed.”

1425. Lord Nicholls said “The third of these conditions is of crucial importance. Otherwise any attempt to determine the meaning of the enactment would cross the boundary between construction and legislation …”. He also said that “This power is confined to plain cases of drafting mistakes. The courts are ever mindful that their constitutional role in this field is interpretative. They must abstain from any course which might have the appearance of judicial legislation.” In citing Wrotham Park,2118 Lord Nicholls invoked Lord Diplock who had set three similar tests and had said:

“Unless this third condition is fulfilled any attempt by a court of justice to repair the omission in the Act cannot be justified as an exercise of its jurisdiction to determine what is the meaning of a written law which Parliament has passed. Such an attempt crosses the boundary between construction and legislation. It becomes a usurpation of a function which under the constitution of this country is vested in the legislature to the exclusion of the courts”.

Lord Diplock declined to fill the gap as to do so would be “a matter of pure speculation”. He did not “find it

possible to state with any certainty what words would have been inserted in the Act to fill the gap ….”

1426. Lord Nicholls went further in curtailing the technique:

“Sometimes, even when these conditions are met, the court may find itself inhibited from interpreting the statutory provision in accordance with what it is satisfied was the underlying intention of Parliament. The alteration in language may be too far-reaching. ……. the insertion must not be too big, or too much at variance with the language used by the legislature. Or the subject matter may call for a strict interpretation of the statutory language, as in penal legislation.”2119

2118 Jones v Wrotham Park Settled Estates [1980] AC 74, 105-106. 2119 Citing Western Bank Ltd. v Schindler [1977] Ch. 1, 18, Scarman L.J.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1427. It is clear that Lord Nicholls and Lord Diplock were very concerned not to trespass the boundary between interpretation and legislation and expressed that concern in terms which do not seem to me to differentiate our respective constitutional circumstances.

1428. Dodd, under the heading “Rectifying Errors”, says that in the case of an obvious drafting slip and even where the true intention of the legislature is clear, whether the courts can rectify such errors turns largely on whether rectification is to be characterised as legislative or not. He cites Bula2120 and Rollinson2121 for the view that “the approach in Ireland has generally been against the rectification of clear drafting slips, at least prior to the enactment of s 5 of the 2005 Act. This is particularly so in respect of penalising enactments. Irish courts have generally preferred to render provisions meaningless rather than run the risk of usurping the role of the Oireachtas.” He describes Ireland as “relatively singular” in that approach as compared to other common law countries. However his account includes two significant elements: his references to penal statutes and to the approach preceding the 2005 Act. Here we are not concerned with a penal statute and, of course, the 2005 Act applies.

1429. Dodd also observes that “As such slips can lead to absurdity or render an enactment meaningless thereby frustrating the intention of an enactment, the strict approach may be difficult to reconcile with some of the views expressed in such cases as Re the Employment Equality Bill 1996 and H v H”. I confess to agreeing with this latter observation – where the true intention of the legislature is clear – discerning that intention objectively from the words used taken in context being the fundamental purpose of all statutory interpretation.

1430. Perhaps the primary reluctance may be to correcting positive drafting errors, as opposed to supplying missing words – perhaps all the more so in a penal statute. In Rollinson, the Supreme Court refused to remedy an obvious drafting error by reading ‘number 27’ as meaning ‘number 26’ in a penal regulation even though that 26 was intended was clear. I confess that I am not entirely clear what part the observation that “The Court has no function in remedying error in circumstances where legislation, though clear in its terms, is found to be defective.” played in the outcome of Bula – though the proponent of that proposition in Bula lost the case and it was not a case in which it was argued that missing words should be supplied.

1431. Dodd cites Lord Nicholl’s approach in Inco Europe as accepted in varying degrees in other common law countries. But, he suggests, it is unlikely to find favour in Ireland. He does so first as much of Lord Nicholls’ approach may rest on the admission of evidence of parliamentary history as indicating legislative

2120 Bula Ltd (In Receivership) v Crowley [2002] IEHC 4. 2121 The State (Rollinson) v Kelly [1984] IR 248 at 255 – Henchy and Griffin JJ, dissenting.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

intent2122 – and those materials are excluded in Ireland. But it does not seem to me necessary to adopt the admission of parliamentary history in order to apply Lord Nicholls’ three tests. Irish courts are well used to discerning legislative intent by the transparent, coherent and objectively ascertainable principles, rules, presumptions and techniques it does allow and s.5 of the 2005 explicitly envisages exercises in discerning legislative intent.

1432. Dodd’s other reason for suggesting that Inco Europe is unlikely to find favour in Ireland is that Step 3 was arguably a form of inquiry into the subjective intent of the legislature which has been rejected here –

citing Howard2123 to the effect that it is inappropriate for the court to speculate as to the legislature’s intent.

Certainly, Howard forbids such speculation, nor do later cases permit it - as Heather Hill makes clear. Denham J in Howard cited Derwent2124 to the effect that the court “cannot add words to a statute or read words into it which are not there”. But Derwent related to a penal statute and Re the Employment Equality Bill 19962125 can be cited against Denham J’s having stated a universally applicable proposition. Denham J cites Cox v Hakes2126 for the proposition that if language of the legislature, interpreted according to the recognised canons of construction, involves a result even of great magnitude2127 the court must yield to it, even satisfied that it was not in the contemplation of the legislature. Denham J considered that if a plain intention is expressed by the words of a statute, the court should not speculate but rather construe the Act as enacted. The plain language of the Act must not be extended beyond its natural meaning so as to supply omissions or remedy defects.

1433. To these observations some answers are possible:

• While s.5 of the Interpretation Act should not be “over-read”,2128 Howard preceded it.

• Dodd acknowledges2129 that, subject to any constitutional constraints, s.5 is arguably intended to permit rectification of obvious drafting errors which render a provision absurd or result in its failing to reflect the plain intent of the legislature.

• As Dodd himself observes, Re the Employment Equality Bill 1996 and H v H can be cited in reply.

• Speculation is forbidden. But it is avoided by reliance only on the recognised means of discerning legislative intent (primarily by the words the legislature used given their natural and ordinary meaning in their context) and recognising that it may not be possible in a given case to determine that intent by such means adequately to permit rectification. In that case, the attempt at rectification will fail and the absurdity will survive.

2122 The Law Reform Commission interpreted the judgment similarly in its Report On Statutory Drafting And Interpretation: Plain Language And The Law

(LRC 61 - 2000) §2.26: “This conclusion was mainly drawn from the history of legislation in this field and the fact that there was no evidence of any

intention to change the status quo in this regard.” 2123 Howard v Commissioners of Public Works [1994] 1 IR 101. 2124 R. v Wimbledon Justices, ex parte Derwent [1953] 1 Q.B. 380. 2125 Re the Employment Equality Bill 1996 [1997] 2 IR 321. 2126 [1890] 15 A.C. 506

2127 The Crown argued that the Judicature Act had allowed appeal of an order of discharge of a prisoner on habeas corpus. The argument failed.

2128 Bookfinders, supra. 2129 §12.49.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1434. In Howard, Finlay CJ agreed with Blayney J as to the principles of statutory interpretation. Blayney J expressed what seems to me a nuanced view. He cited Craies2130 as to statutory interpretation by implication:

"If the meaning of a statute is not plain, it is permissible in certain cases to have recourse to a construction by implication, and to draw inferences or supply obvious omissions. But the general rule is 'not to import into statutes words which are not to be found there,' and there are particular purposes for which express language is absolutely indispensable.2131 Words plainly should not be added by implication into the language of a statute unless it is necessary to do so to give the paragraph sense and meaning in its context."

Blayney J added that the first pre-condition to recourse to construction by implication is that the meaning of the statute should not be plain. I interpret this as a condition that the strictly literal meaning should not be plain. That appears to me to be the case as to s.13A(6).

1435. In my view, the Inco Europe test for interpolation of words represents a demanding test. In substance it reflects both Lord Mersey’s warning and, in this State, the proper constitutional restrictions of the judicial role. Its application does not require or imply recourse to parliamentary history – even though Lord Nicholls had such recourse on the facts before him.

Application of Principles of Statutory Interpretation

1436. It appears to me that while the test for interpolation of words into a statutory provision is very demanding, it is satisfied in the present case. First, the condition that the strictly literal meaning should not be plain is satisfied as to s.13A.

1437. But the general purpose of s.13A is entirely obvious. It was inserted in the 1933 Act by the European Communities (Environmental Impact Assessment) Regulations 19892132 for the entirely obvious purpose of introducing, to an Act which long-predated EIA, the requirements of EIA law given that the various authorisations contemplated by the Foreshore Act 1933 – notably foreshore leases and foreshore licences –

2130

Craies on Statute Law (1971) (7th ed.) citing Patterson J. in King v Burrell (1840) 12 Ad. & El. 460, and Evershed M.R. in Tinkham v Perry [1951] 1 T.L.R.

91. 2131 For example, I assume, penal statutes.

2132 S.13A has since been amended by, inter alia, the European Communities (Environmental Impact Assessment) (Amendment)Regulations 1998 (S.I. No.

351 of 1998), European Communities (Environmental Impact Assessment) (Amendment) Regulations 1999 (S.I. No. 93 of 1999), the European Union

(Environmental Impact Assessment) (Foreshore) Regulations 2012 (S.I. No. 433 of 2012), the European Union (Environmental Impact Assessment and

Appropriate Assessment) (Foreshore) Regulations 2014 (S.I. No. 544 of 2014), the European Union (Foreshore Act 1933) (Environmental Impact

Assessment) (Amendment) Regulations 2021 (S.I. No. 145 of 2021). SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

had the capacity, in individual cases and depending on circumstance, to constitute development consents within the meaning of the EIA Directive. While it is not legally impossible that the State could have sought to exceed EIA Directive requirements by requiring two separate but identical EIAs of a project it is, at very least, highly unlikely.

1438. It is implied from the situation of s.13A in the Foreshore Act 1933 and in any event explicit from s.13A(5), that the development consents to which s.13A is to apply are, and are only, the various authorisations contemplated by the Foreshore Act 1933 – notably foreshore leases and foreshore licences. Nowhere in s.13A or from the scheme of the 1933 Act is it apparent that it is a purpose of s.13A of the 1933 Act to require EIA in aquaculture licence applications. That is no surprise as EIA in aquaculture licence applications is required by the 1997 Act.

1439. It is apparent from the scheme of the 1933 Act that the legislature intended special provision for foreshore licences for aquaculture. Notably, and by way of the only such exception, s.1B of the 1933 Act assigned the function of grant or refusal of foreshore licences2133 for aquaculture, sea fishing and the like to the Minister. The grant or refusal of all other foreshore licences2134 is by the Minister for the Environment, Community and Local Government. While the 1933 and 1997 Acts are not generally to be construed together, s.82 of the 1997 Act is an exception2135 in requiring that the Minister, in considering an application for a foreshore licence sought in connection with the carrying on of aquaculture pursuant to an aquaculture licence, shall have regard to any decision of the licensing authority in relation to the aquaculture licence. It is clearly no coincidence that the grant or refusal of foreshore licences for aquaculture is assigned, as an exception, to the Minister responsible for the grant or refusal of aquaculture licences. The clear legislative intent is to facilitate their co-ordinated decision. In my view, it is inevitable that such co-ordination is intended to avoid unnecessary and wasteful duplication of process and the 1933 and 1997 Act should not be construed so as to require duplication of EIAs unless such a construction is unavoidable.

1440. When trying to make sense of s.13A(6) as an explicit exclusion from the general scope of s.13A, it is legitimate to ask the question – what general type of permit2136 could one expect to see excluded by s.13A(6)? To that question the definition of “relevant application” in s.13A(5) admits of only one answer – a permit – lease, licence approval or consent2137 – under the Foreshore Acts. Yet literally, what s.13A(6) purports to exclude is aquaculture licence applications, not Foreshore Acts applications. That makes no

2133 And any other Foreshore Licence authorisations. 2134 And any other Foreshore Licence authorisations.

2135 s.67 of the 1997 Act and S.12 of the 1933 Act as to removal of unlawful aquaculture structures and equipment are another exception.

2136 To use a neutral word.

2137 “5) In this section and in sections 13B, 19A, 19C, 21A and 21B “relevant application” means, as the case may be—

(a) an application to the appropriate Minister for a lease under section 2 of this Act,

(b) an application to the appropriate Minister for a licence under section 3 of this Act,

(c) an application to the appropriate Minister for his approval under section 10 of this Act for maps, plans, and specifications for erection of structures

on the foreshore,

(d) an application to the Minister for the Environment, Heritage and Local Government for his consent under section 13 of this Act for the deposit of

material on the foreshore.” SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

sense. As I have said, construed literally, s.13A(6) is absurd as excluding from the scope of s.13A, aquaculture licence applications which could never have fallen within it in the first place. So, construed literally, it has no effect – and the legislature is presumed not to legislate in vain.

1441. However, it is notable in s.13A(6) that the aquaculture license applications to which it is addressed are specifically and only those accompanied by an EIS/EIAR.2138 The clear implication is that whatever s.13A(6) seeks to exclude will be subjected to EIA in the aquaculture licensing process and that thereby the purpose of the EIA Directive, which s.13A was itself inserted to achieve, will be achieved.

1442. It appears to me to follow from the foregoing that the necessarily true and obvious intent of s.13A(6) was to exclude certain Foreshore Acts applications from the definition of “Relevant Application” in s.13A(5) and hence from the effect of s.13A generally, and thereby from the general obligation of EIA of foreshore licence applications imposed by s.13A. That is rendered comprehensible from a purposive point of view when one notes that, whatever the relevance of Aquaculture Licences in s.13A(6), only aquaculture licence applications which are to be subjected to EIA are within s.13A(6).

1443. It appears to me necessarily to follow that this is one of the relatively rare cases in which, both on the Inco Europe test for interpolation and on the authority of Re the Employment Equality Bill 1996 and H v H, it is necessary to interpolate words in s.13A(6) to reflect the clear intent of the Oireachtas as discernible from the statutory scheme – failing which s.13A(6) would be absurd and meaningless. And it is perfectly plain what those words ought, in substance, to be.

1444. However, one fly remains in that ointment. S.13A(1)(a) & (b)(i)(II) of the 1933 Act require EIA of Foreshore Licence applications for developments specified in Schedule 5, Part 2, PDR 2001. Schedule 5, Part 2, §1(f) specifies “Seawater fish breeding installations with an output which would exceed 100 tonnes per annum.” It follows on a literal reading of s.13A(1) that Foreshore Licences for such installations – including the subject Foreshore Licence must be informed by EIA. This directly conflicts with the interpretation of s.13A(6) which I have suggested above. But from a literal point of view it is difficult to reconcile with any understanding of s.13A(6) – absurd or sensible. It is unpalatable to infer a second drafting error in the section but I think the explanation is that Schedule 5, Part 2, PDR 2001 lists approaching 100 types of project.2139 It seems to me likely that §1(f) was overlooked. If I am wrong in that, I prefer the specific provision of s.13A(6) as I have interpreted it to the general provision of s.13A(1).

2138 2011 EIA Directive/2014 EIA Directive. 2139 The figure is approximate and turns in part on how sub-types are considered.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

1445. Accordingly, with no little hesitation but confident of the matter from a purposive point of view, I take the view that the effect of s.13A(6) is that no EIA was required in the Foreshore Licence Application in this case and that the relevant EIA requirement arose and arose only in the Aquaculture Licence application and Appeal.

1446. However, the Minister in deciding the Foreshore Licence Application, in fulfilment of his duty to have regard to ALAB’s decision in the Aquaculture Licence Appeal, will have regard to ALAB’s EIA done in that appeal.

CONCLUSION

1447. For the reasons set out above, the Aquaculture Licence and the Foreshore Licence will be quashed. Briefly put, the Aquaculture Licence will be quashed for inadequate:

i. AA Screening of the risk of effects of seal scarers on seals of the SAC.

ii. EIA as to the risks of escape of salmon from the fish farm. This finding relates to ▪ necessity of re-consideration of bespeaking the DAFM reports on the 2014 farmed salmon escape in Bantry Bay, and ▪ comprehensiveness of the EIA as it related to the specification and structural integrity of the cage installation.

iii. reasons for the conclusion that the proposed fish farm will not lead to a breach of WFD limits as to Dissolved Inorganic Nitrogen – specifically, reasons for reliance on RPS’s “typical” data in reaching that conclusion.

In addition, I will declare that ALAB delayed unreasonably as to AA Screening from the making of the Appeals in October 2015 to embarking on AA Screening after the Oral Hearing Report of November 2017. I will grant no further relief on that account.

1448. The Foreshore Licence will be quashed as

• contingent on the quashed Aquaculture Licence, Ministerial regard to which was a statutory requirement of granting the Foreshore Licence.

SWI, IFI, Sweetman & Ors v ALAB et al [2024] IEHC 421 Unapproved

• the Minister erred, in breach of s.82 of the 1997 Act, in granting the Foreshore Licence in 2022, in having regard to his Aquaculture Licence decision of 2015 rather than to ALAB’s impugned Aquaculture Licence determination of 29th June 2021.

1449. I have also held that the effect of s.13A(6) of the Foreshore Act 1933 is that no EIA was required in the Foreshore Licence Application in this case and that the relevant EIA requirement arose and arose only in the Aquaculture Licence application and Appeal. However that EIA in the Aquaculture Licence application will have regard to ALAB’s EIA done in the Aquaculture Licence Appeal.

1450. In the interests of good public administration and also as to issues ventilated but not pleaded, have made certain remarks above as to matters which do not ground relief herein but which the respective

decision-makers might usefully bear in mind in making their remitted decisions. I should emphasise that I express no view on the possibility of the application of the rule in Henderson,2140 as articulated by the Court of Appeal in Clonres,2141 to any judicial review of any decision by ALAB and/or the Minister made on remittal. Any such possibility cannot be considered in the abstract or in prospect. My observations are, rather, directed to what will undoubtedly be their anxiety to make any such decisions in accordance with law.

1451. I provisionally consider that both the Aquaculture Licence and the Foreshore Licence should be remitted for re-decision. I will hear the parties as to form of order and as to costs if they can’t be agreed. I encourage the parties to actively seek to agree these matters in whole or in part. I will list the matter for final substantive orders on 30 July 2024 and for further directions if a hearing as to costs is required.

DAVID HOLLAND 12/07/2024

2140 Henderson v. Henderson (1843) 3 Hare 100. 2141 Clonres CLG v Minister for Arts et al [2022] IECA 172.