England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE, QUEEN'S BENCH DIVISION
ADMINISTRATIVE COURT IN BIRMINGHAM
HHJ David Cooke
[2015] EWHC 314 (Admin)

		Royal Courts of Justice Strand, London, WC2A 2LL
		17/06/2016

Lord Justice Beatson :

I. Overview:

1. This appeal concerns salmon fishing at Lydney in the Severn estuary. Since 2012 the appellant, the Environment Agency ("the Agency"), has imposed an annual limit on the number of salmon caught by using a putcher rank, an ancient method of fishing. The method involves the use of conical baskets to trap adult salmon as they make their way back from the sea to the river of their birth to spawn. The respondent, Mr Nigel Mott, and the interested party, Mr David Merrett, are the leasehold owners of a right to fish for salmon at Lydney using putchers. Before the introduction of the limit they caught about 600 salmon in a year. The conditions imposed on their licences reduced their permissible catch by about 95% to 30 in 2012, 23 in 2013, and 24 in 2014.
2. The conditions limiting the number of salmon caught are referred to as "catch conditions". The Agency imposed them because it considered that it was necessary to do so for the protection of the salmon fisheries in the River Wye, which is a Special Area of Conservation ("SAC") under the Habitats Directive, Council Directive 92/43/EEC. It had assessed that: (a) the Severn estuary fisheries exploited a mixed stock, (b) the salmon caught there included salmon that originated in the River Wye and would otherwise spawn in that river, and (c) the fishery in the Wye is at risk of not achieving its spawning targets and of becoming unsustainable. Lydney is some 15 kilometres upstream from the mouth of the River Wye, and thus further from the open sea than the mouth of the Wye. The Agency proceeded on the basis of its assessment that salmon caught at Lydney which it estimated originated from the Wye would, if they had not been caught, inevitably have returned to the Wye to spawn, even though they had swum some 15 kilometres past its mouth.
3. Mr Mott challenged the conditions limiting his permissible catch in two applications for judicial review. The first, filed on 29 August 2013, challenged the conditions imposed for the 2012 and 2013 seasons. The second, filed on 23 September 2014, challenged the conditions imposed for the 2014 season. Mr Mott was successful. Following a careful and well-structured judgment ([2015] EWHC 314 (Admin)) handed down on 13 February 2015, in an order dated 18 March 2015, HHJ David Cooke, sitting as a judge of the High Court in the Administrative Court in Birmingham, allowed the claims. The order declared that the Agency's decisions to impose these limits were unlawful because they were irrational in the Wednesbury sense and unlawfully interfered with Mr Mott's right to the peaceful enjoyment of his possessions under Article 1 of the First Protocol to the European Convention on Human Rights ("A1P1"). The judge also gave Mr Mott permission to amend his claims to include a claim in damages under section 8 of the Human Rights Act 1998 for the breach of his A1P1 rights. The Agency appeals from that order.
4. It will be necessary to consider the regulatory background, the basis on which the Agency imposed the limits, and the reasons the judge found them to be unlawful in some detail. At this stage, it suffices to say that the Agency's assessment that the putcher ranks in the Severn Estuary were having a material effect on the salmon fishery in the Wye and its consequent decisions were largely based on the "Report to the Environment Agency on the Genetic Assignment of Severn Estuary Adult Atlantic Salmon", April 2012, by King, Counter and Stevens, three researchers at the School of Biosciences in the University of Exeter's College of Life and Environmental Sciences ("the Exeter Report"). The 2014 decision was also in part based on an updated report by the same authors in 2013.
5. The evidence before the judge consisted of the Exeter reports, three statements of Mr Mott, dated 29 August 2013, and 26 March and 8 August 2014, and two statements of Mr Crundwell dated 24 September 2013 and 21 October 2014. Mr Crundwell is a senior technical specialist in fisheries in the Agency's Midland region and has been employed by the Agency or its predecessors for 21 years. The evidence before the judge also included three reports about the Exeter report and the methods it used which were prepared after the 2013 decision as to the catch limit and therefore were not available to the agency at the time of the first two decisions that are challenged. The reports are of Professor Rachel Fewster, a specialist in statistical ecology at the University of Auckland, whom Mr Mott consulted, and Dr Terry Beacham, the Head of Stock Identification at the Molecular Genetics laboratory at the Pacific Biological Station of the Canadian agency Fisheries and Oceans Canada, whom the Agency consulted. Professor Fewster's reports are dated June and November 2014, the latter a response to Dr Beacham's report which is dated 23 October 2014.
6. The judge considered that flaws in the Exeter Report meant it did not provide a rational basis for the Agency's view and its decisions and that they were consequently Wednesbury unreasonable or irrational. On A1P1, he held that the irrationality of the decisions meant there was a disproportionate and therefore unlawful interference with Mr Mott's rights. He also held that, even if the Agency could properly have imposed these limits, the extent of the interference with Mr Mott's rights meant that in the absence of compensation the burden was disproportionate and a breach of A1P1.
7. The two principal questions for decision are the rationality of the Agency's approach to its regulatory powers against the background of its regulatory duties under the Habitats Directive and whether, in the context of A1P1, the Agency's interference with Mr Mott's right to fish by imposing a limit on the number of fish caught each year is a "deprivation" or a "control", and, if the limit is imposed without compensation, it is disproportionate and therefore unlawful. The first question also involves considering the approach of a judicial review court when considering a regulator's decision that is predictive in the sense that it is the result of an evaluation of assessments as to what might happen in the future and that evaluation is wholly or partly made on the basis of scientific material.
8. For the reasons given at [60] - [80] below, I have concluded that the appeal against the declaration that the decisions to impose the catch limits were unlawful because they were irrational in the Wednesbury sense should be allowed. I have also concluded that the appeals against the declaration that the decisions unlawfully interfered with Mr Mott's right to the peaceful enjoyment of his possessions under A1P1, and against the judge's decision to permit an amendment to include a claim in damages, should be dismissed. My reasons are given at [83] - [90] below.

II. The legislative framework

9. Putchers were used for many years before the controls on salmon fishing introduced by the Salmon Fisheries Acts 1861 and 1865. Those with the right to fish using them had the right to continue to do so, notwithstanding those controls under certificates of privilege granted under the 1865 Act. As a result, putchers are referred to as a "Privileged Engine" or "an historic installation". Since 1975 those using putcher ranks have been required to obtain a licence: see section 25 of the Salmon and Freshwater Fisheries Act 1975 ("the 1975 Act"). The power to issue such licences is now vested in the Environment Agency.
10. Until recently the position of those using "an historic installation" differed from those using other licensable means of fishing, such as rod and line or net. In case of a licence issued for use of "an historic installation" there was no power under the 1975 Act to impose conditions such as the "catch limitations" imposed in the decisions challenged in these proceedings. Moreover, the power to limit the number of licences of a particular type issued in a given area applied only to fishing by net: section 26 of the 1975 Act.
11. The position changed at the beginning of 2011, when section 217(7) of the Marine and Coastal Access Act 2009 came into effect. This introduced a new paragraph 14A into Schedule 2 to the 1975 Act. Sub paragraphs (1) and (4) of paragraph 14A empower the relevant regulator to impose conditions on the use of a fishing licence granted in respect of "an historic installation" "where it considers that it is necessary to do so for the protection of any fishery".
12. As to the wider regulatory environment, the relevant EU provision is the Habitats Directive, Council Directive 92/43/EEC. It was transposed into United Kingdom Law by the Conservation (Natural Habitats, &c.) Regulations 1994, SI No 2716 of 1994, which were consolidated by the Conservation of Habitats and Species Regulations 2010, SI No 490 of 2010 ("the 2010 Regulations"). The main aim of the Habitats Directive is to promote the maintenance of biodiversity by requiring Member States to take measures to maintain or restore natural habitats and wild species listed, in particular in sites designated as SACs.
13. Article 6(3) of the Habitats Directive requires the Agency to consider whether any plan or project is likely to have a significant effect on a SAC. If it is likely to have such an effect, the Agency is required to carry out an appropriate assessment of the implications for the site in view of its conservation objectives in consultation with the relevant nature conservation bodies, now Natural England and Natural Resources Wales. The Agency is permitted to agree to the plan or project only after having ascertained that the integrity of the site will not be adversely affected. Article 6(3) has been transposed into the law of England and Wales by regulation 24 of the 2010 Regulations. As the River Wye is an SAC, the Directive and the Regulations require the Agency to consider whether licences to fish for salmon in the Severn estuary are likely to have a significant effect on it and, if so, to undertake an assessment.

III The factual background

14. I have stated that Mr Mott and Mr Merrett have a lease to fish at Lydney. They have had one since 1975. The present lease is for a period of 20 years from 1 April 1998. Under it, the landlords:

"hereby grant and demise unto the Tenants ALL THAT right to fish two stop nets and 650 putchers in the estuary of the River Severn near Lydney Pill granted under Certificate Numbers 9/56 and 10/57 issued by the Special Commissioners for English Fisheries on the fourteenth day of May 1866 under the Salmon Fishery Acts 1861 and 1865 in the approximate position shown on the plan attached to the said Certificates a copy of which is annexed hereto and marked 'A' (which right is hereinafter referred to as 'the Fishery') TOGETHER WITH the right to store equipment ancillary to the Fishery on the land measuring 14 yards wide and 35 yards long adjacent to the Fishery appointed by the Landlords AND TOGETHER WITH a right of way … along the trank shown with a red line on the plan marked 'B'… "

The certificates referred to in the lease declared the putcher rank to be a "Privileged Engine" for the purpose of the 1861 and 1865 Acts.

15. I referred (at [1]) to the evidence that an average of about 600 salmon were caught by Mr Mott's putcher rank. The average value of a single fish was £100 and the gross annual income was therefore approximately £60,000. In his first witness statement dated 29 August 2013 Mr Mott also stated that, since 1979, operating the rank of putchers has been his full-time occupation and the putcher rank has supported his and Mr Merrett's families. It is his case that the conditions reducing the total number of salmon that could be taken to between 23 and 30 fish each year, make the fishery wholly uneconomic and the lease, the rent of which was £180 a year, worthless.
16. In a letter dated 16 April 2012, the Agency informed Mr Mott that the Exeter report "provides clear evidence of the mixed stock nature of the catch" in the Severn estuary, the estimated proportions of salmon originating in other rivers, and of its decision to proceed with its Habitats Regulations Assessment ("HRA") in consultation with Natural England and what was then the Countryside Council for Wales and is now called Natural Resources Wales. The Agency also stated that its conclusion was that, in order to conclude that the estuary fisheries will not adversely affect the integrity of the European sites, it had to set conditions on all licences issued and that it proposed that, for 2012, the condition on putcher fisheries would be set at 30 salmon.
17. The 2012 HRA, dated 24 May 2012, contains the Agency's assessment that "recent and future potential unconstrained catches of salmon in the estuary threaten the integrity" of the River Wye SAC which it judged "to be 'at risk' of failing to achieve its annual conservation limit and its five-year management target until at least 2016" and the "catch condition" of 30 salmon. The HRAs for 2013 and 2014 are dated 21 May 2013 and 2 May 2014. I consider them and the other material as to the basis of the Agency's assessment of the proportion of the salmon caught in the Severn Estuary that originated from the River Wye in section IV below. The decisions challenged in these proceedings were made on 1 June 2012, 30 May 2013, and 30 May 2014. The reports of Professor Fewster were not available at the time of the first two decisions: an interim report by Professor Fewster was exhibited to Mr Mott's statement of 29 August 2013, the date the first judicial review was lodged, and a draft of the final report was made available to the Agency in January 2014. Dr Beacham's report post-dates all three decisions.

IV. The policy and regulatory background and the evidence

18. (i) Introduction: I have referred to the Habitats Directive. The background against which the Agency's decisions were taken also included national policies which were influenced by international organisations and, albeit to a limited extent, two studies apart from the Exeter Report, one by Swain in 1982 and one by Jones in 1994.
19. In his oral submissions, in order to inform the court of the regulatory background and the views of experts on the topic, Mr Gwion Lewis principally relied on a document issued by the Agency in February 2014 entitled "Technical case for two net limitation orders for the Severn Estuary" ("the 'Technical case' document"). The judge was only taken to two pages of the Technical case document and it appears that the HRAs were the focus of Mr Lewis's submissions below. It is understandable that before us he relied on the Technical case document and first took us through it carefully. I consider that it sets out the relevant background material in a more accessible way than in the HRAs which present the material in what Mr Lewis described as a more "compacted" way.
20. I refer to the HRAs below, but my summary of the regulatory background and the assessments made is largely taken from the Technical case document. In relation to material in it on the specific points that fall for decision as opposed to background, I have cross-checked what is stated in it against the HRAs. This is because: (a) the Technical case document is directed to net limitation orders and, (b) it is dated almost two years after the first of the challenged decisions and about six months after the first of the two judicial reviews were filed.
21. (ii) National and international policies: The regulatory background includes government policies from 1996 to phase out fisheries which exploit "predominantly mixed stocks". The judge stated (at [16]) that this was an important aspect of the Agency's thinking. It is stated in paragraph 1.2.3 of the Technical case document that this policy was adopted in 1996 in England and Wales, and that in 2001 the Welsh and English Governments accepted advice of the Salmon and Freshwaters Fisheries Review "that mixed stock fisheries should be phased out wherever possible". The United Kingdom Government had previously identified three such fisheries in England, one of which is the "Severn Estuary nets and fixed engines". The position was "reaffirmed" by the Agency in its 2008 strategy for salmon and sea trout which stated it "will move to close net fisheries that exploit predominantly mixed-stocks where our capacity to manage individual stocks is compromised".
22. The background also involves consistent advice by the North Atlantic Conservation Organisation to national regulators such as the Agency to control the exploitation of salmon where river stocks do not exceed their conservation limits and, in 2013, to prevent exploitation in such cases: see Technical case document, paragraph 1.4. That document stated earlier (paragraph 1.2.1) that the Countryside Council for Wales concluded that the River Wye salmon feature is in "unfavourable status", that while the position of the Severn Estuary European Marine Site "has not been specifically assessed", as its status "draws on that of the rivers contributing salmon to the mixed stock within the estuary, it is also deemed unfavourable".
23. ...iii) The Exeter Reports: It will be necessary to consider the Environment Agency's reasons for imposing these catch conditions and the 2012 and 2013 Exeter Reports when analysing the judge's decision. At this stage it suffices to state that, as the judge stated at [16], "[a] considerable part of the Agency's focus has been on whether the Lydney fishery is a mixed stock fishery". As the criticisms of the methodology used in the Exeter Reports to assess the genetic composition of salmon found in the various tributaries of the Rivers Severn, Wye and Usk are at the core of Mr Mott's challenge, I first summarise those reports. They are summarised more fully in the judgment below at [37] – [49].
24. The 2012 Exeter Report is a genetic analysis of DNA samples collected by the Environment Agency Wales from the fins or scales of fish in various tributaries of the Severn, Wye and Usk and three other rivers. The report concluded that there is comparatively little genetic difference between the salmon population of the various tributaries. It then set out how genetic markers for salmon could be used to "assign" a fish to a particular river based on the similarity of the genetic profile of the fish with those of the whole river. It considered a percentage probability of a match to each river could be assessed using two methods, Geneclass2, a Bayesian "degrees of belief" approach, and ONCOR, a "maximum of likelihood" approach. These methods were used on the genetic markers from 55 adult fish caught in the Severn estuary net fishery which the judge stated at [42] were caught at or near Mr Mott's fishery. The robustness of these methods and the use made of the resulting estimate of the rivers of origin of the fish are at the heart of the dispute between the parties.
25. The percentage of the salmon assigned to the Wye was 36.5% using GeneClass2 and 56.4% using ONCOR. The report noted (at page 10) that with "generalist fish/genetic profiles", "all methods will struggle to be able to [reliably] assign [a fish] to a river of origin with a high degree of certainty"' but concluded that salmon taken from the Severn estuary nets supplied for genetic analysis "definitely constitute a mixed stock fishery" and a "significant number appear to originate from all three key rivers in the region". The 2012 Report concluded "[t]hus it appears highly probable that fish from the Wye, Usk and Severn are being caught in these Severn nets".
26. The March 2013 updated Exeter Report again used two main types of analysis. On this occasion they were ONCOR and cBayes, although the report also refers to GeneClass2. The report stated (page 5) that mixed stock analysis of the known origin samples showed that both ONCOR and cBayes "underestimated the contribution of the Severn and overestimated the contribution of the Usk". It also stated that only a minimal number of the sample of salmon caught in the Severn estuary nets were from the Severn. The judge stated at [46] "this produced the even more striking result" that none were assigned to the Severn, 76% were from the Wye, and the remainder from the Usk.
27. In the section "Do the Severn nets represent a mixed stock fishery?" after answering the question "yes", it is stated that "not all the adult salmon analysed [can be] assigned with the same degree of confidence to a baseline river and, therefore, for these fish it is not possible to definitively determine their river of origin". It, however, continued "[n]onetheless, a significant number of adult salmon analysed have high assignment probabilities/percentages to one particular river using all three assignment methods", and that the report writers had a "high degree of confidence" that these were from the Wye or the Usk. Only 4 fish of the sample of 55 were put into this category, so that it was not possible to definitively determine the river of origin of the remaining 51. The Technical case document described the Exeter Report as using "a new genetic analysis using improved baseline genetic material [that] has now given unequivocal evidence of the mixed stock status and provides our first estimate of river apportionment from the sampled dataset": paragraph 1.2.3.
28. (iv)The Agency's assessments: The 2012 HRA stated that the population of the Severn estuary salmon fisheries is mixed and its mixed-stock nature represented "a significant problem for rational stock management, as the weakest stock(s) cannot be discretely managed". It referred to the United Kingdom government's policy that, in such circumstances, mixed-stock fisheries should be closed and to a proposal to replace voluntary "catch and release" arrangements for the Wye with byelaws requiring this. At the hearing Mr Lewis stated that proposal was a way of addressing the poor status of the Wye stock and that byelaws requiring all salmon caught in the Wye to be returned alive to the river came into effect in July 2012.
29. The 2012 HRA also stated that "an in-combination assessment for the mortality of Wye salmon in the Severn estuary fisheries and in the Wye rod fishery" had been carried out. The assessment "compare[d] the actual catches over the past decade… and then simulate[d] catches and mortality that [would] have occurred under the proposed new regulatory regime of full catch and release rod fishing in the river and conditioned licences for the estuary fishery". On the basis of that and other evidence the Agency considered that the licences for fishing using putcher ranks in the Severn Estuary were likely to have a significant effect on the Wye SAC. This was because the salmon catch there consisted of salmon from each of the three rivers with the proportion determined by what the more recent Technical case document referred to as "probabilistic methods" to be the Severn 5.5%, the Wye 56.4%, and the Usk 38.2%, that is, the figures in the Exeter report using the ONCOR method.
30. In paragraph 3.5.2 of the Technical case document, referring to the Exeter report and the earlier studies by Swain and by Jones, the Agency stated:

"We are content that the three studies provide evidence that the Severn Estuary Fishery exploits salmon from at least the three principal rivers and it therefore a mixed stock. The genetic study [the Exeter report] has quantified a probabilistic 'most likely' assignment of fish to home river. Many of the estuary fishermen accept the mixed stock findings but feel that the proportion of fish from each river remains open to debate, as the consultation demonstrated. The [Agency's] view is that the estuary clearly hosts a mixed stock fishery and the proportions are the best evidence we have currently so should be used."

Paragraph 2.6 stated:

"As with all models there are uncertainties associated with the data used to make our assessment".

The document considered uncertainties from undeclared catches, undeclared effort, exploitation rates, and the possible existence of a run of fish out of season and gave reasons for considering that they were not significant.

31. The HRAs set out two tables. One showed the total numbers of salmon from the Wye killed by the combination of rod fishing in the Wye itself and the catch in the Severn estuary in a ten year period. The other estimated the number of Wye fish that would have been killed in those years if the proposed "catch and release" regulatory regime had been in force. In the 2012 HRA the period was 2000-2009 and it was assumed that 9% of the fish released would die, the residual mortality assumption, was 9%. In the 2013 HRA the period was 2003-2012, and the residual mortality assumption was 20%. The 2013 HRA also stated that the 56.4% figure was an "estimate" rather than, as in the 2012 HRA, a "known" figure. Apart from these differences and the one I refer to at [32] below, the two HRAs contained much the same information.
32. No reason is given in the 2012 HRA for selecting a condition of 30 fish per licence for historic installation fisheries number, a matter noted at [30] by the judge. The 2013 HRA, which proposed a catch limit of 23 for the historic installations, a reduction from 30, referred to restricting the catch to the "approximate long term de minimus (sic) catch" but does not explain what that is. An explanation is, however, provided in Mr Crundwell's first statement and in paragraph 3.6.3 of the Technical case document, a paragraph which was not shown to the judge.
33. In Mr Crundwell's first statement it is stated that that the total allowable catch in the Severn estuary is:

"derived from the ten year average residual mortality in the River Wye rod fishery, assuming that full catch and release fishing had been in place. We then use the best available evidence (the genetic analysis of salmon caught in the estuary fisheries) to extrapolate the residual mortality to set a commensurate [total annual catch] for the net fisheries. This helps us to achieve equity between the rods and nets. The analysis of genetic samples taken in 2010 net fishery indicates the most likely origin of 54% of the adult salmon was the River Wye".

34. The explanation in paragraph 3.6.3 of the Technical case document is that the number was obtained by using the lowest catch sought by any holder of a historic installation licence, "the de minimis HI catch". At the hearing Mr Lewis stated by reference to that document that the figure was not in fact purely based on the lowest catch sought by any holder of a historic installation licence, because that would have resulted in a figure lower than 23. He stated that the Agency also took into account the assumed figure for residual mortality of fish caught in the Wye and returned to that river. That is, however, not explained either in the HRA or in the Technical case document.
35. The conclusions section of the 2012 HRA stated that the new information derived from the Exeter report "confirm the nature of the mixed stock, and for the first time quantify (sic) the contributions from each river". It stated that the data indicate that over 90% of the salmon caught are derived from the two riverine SACs and gave the percentages referred to at [29] above and the Agency's assessment quoted at [17] above. The Agency also stated that, notwithstanding the impact on the Wye, it recognised "the heritage value of the fishing methods used in the estuary, some of which are wholly unique to this site". It accepted "that some residual exploitation of salmon will take place but consider[ed] that the residual catch is sustainable and therefore acceptable…"
36. The Technical case document stated that the Agency's conclusion was that the salmon features of the three European sites are "in unfavourable condition". It considered that managing exploitation through the application of appropriate "catch limits" in addition to limiting the number of licences would allow it to conclude "no adverse effect". It also stated that exploitation control for the River Wye stock, "where there is a substantial shortfall in egg deposition, is urgently required".
37. As to the "catch conditions", the document stated that these were determined on the basis of the Agency's assessment of the most likely assignment of fish to home-river but also had regard to what the Agency described as the heritage value of the fishery. In 2012 it determined the permitted catch for historic installations, i.e. for putcher racks, by reference to the lowest catch by any of the historic installation fisheries had that sought a licence in the preceding ten year period. In 2013, this method would have resulted in a permitted catch of 15 salmon, which the Agency considered would "be unreasonable [and result] in a reduction in licence uptake and failure to maintain possible heritage". The document stated that the controlling factor in the decision was the Agency's assessment of the sustainability of the fisheries, in particular the Wye fishery, absent control, and its determination of the number of salmon caught in the Severn estuary that would have returned to the Wye to spawn.
38. (vi) The evidence of Professor Fewster and Dr Beacham and the 2014 HRA: I have referred to the reports produced by Professor Fewster and Dr Beacham after the 2013 HRA. The judge summarised Professor Fewster's principal points of criticism of the Exeter report at [52], and the Agency accepted that summary. He summarised Dr Beacham's principal conclusions at [53].
39. I can summarise their evidence very briefly indeed in the light of the way the judge approached his task. Their evidence was not available at the time of the 2012 decision. The judge correctly stated (at [59]) that the lawfulness of a decision must be assessed in the light of the evidence and information available to the decision taker at the relevant time and its rationality must be evaluated accordingly. He did not therefore consider their reports when assessing the legality of the 2012 decision. When dealing with the later decisions, he stated (at [70]) that the position was "the same or a fortiori" as for the 2012 decision. He only referred in general terms to the later expert evidence as undermining the Agency's position: see [49] below.
40. In a nutshell, Professor Fewster considered that there were fundamental logical flaws in the analysis used in the Exeter report to assign individual fish to the most likely river of origin because of the low degree of genetic differentiation between the candidate populations. She considered that all the 55 fish caught at or near Lydney could plausibly have originated from the River Severn. Dr Beacham agreed that the technique used in the Exeter report could not provide a reliable estimate of the proportion of fish originating from each of the three rivers but considered nevertheless that there was evidence to show that some of the fish originated from the River Wye and that the salmon caught in the Severn estuary net fishery are from a mixed-stock sample.
41. (vii) The 2014 HRA: This was dated 2 May 2014, about 5 months before Dr Beacham's review of Professor Fewster's evidence was available. Although the Agency received a draft of Professor Fewster's final report in January, the report was not finished until June. The Agency stated that until Dr Beacham's review was complete, the Exeter Report was still the best evidence available. Accordingly, it used the estimated figure of 56.4% as part of its determination of the catch limit set in the decision dated 30 May 2014. Mr Mott challenged that decision in his second application for judicial review.

V Procedural history

42. I have stated that Mr Mott's first application for judicial review challenging the conditions imposed for the 2012 and 2013 seasons was filed on 29 August 2013. I have stated that Professor Fewster's interim report was exhibited to the statement he filed in support of that application. I add that Mr Mott's evidence included a statement by him that the Agency had estimated that the annual salmon run in the Severn is between 10-15,000 fish. Permission was granted to challenge the conditions imposed for the 2013 season on the papers by HHJ McKenna on 3 February 2014. A draft of Professor Fewster's final report was completed in June 2014 and, on 17 July 2014, HHJ Purle QC gave Mr Mott permission to rely on it.
43. Mr Mott's second application for judicial review, filed on 23 September 2014, challenged the conditions imposed for the 2014 season. On 7 October 2014, HHJ McKenna adjourned it to be listed as a "rolled-up" hearing on notice to the Agency at the same time as the first application. An extension of time was needed for the challenge to the conditions imposed for the 2012 season, and that was also considered on a "rolled up" basis at the hearing.
44. In the first application, there were four grounds of challenge:-

Ground 1: The imposition of catch limits in the 2012 and 2013 licences which deprived Mr Mott of his livelihood was irrational;

Ground 2: The level of the limit on catch imposed was wholly disproportionate to the alleged environmental benefit and the system of licence fees levied by the agency, and was therefore unlawful;

Ground 3: The decision effectively deprived Mr Mott of his livelihood and was an unlawful interference with his peaceful enjoyment of his possessions as protected by A1P1;

Ground 4: The Agency's reasons for imposing a limit on the catch in the 2012 and 2013 licences were not adequate or intelligible.

45. In the second application, Mr Mott also submitted that the Agency unlawfully and unreasonably failed to undertake a timely independent review of the scientific evidence in the light of Professor Fewster's report before imposing the limits on the salmon catch in the 2014 licence.
46. On 12 November 2014, very shortly before the hearing which was listed for 20 November, Mr Mott applied to amend his claims to include a claim in damages for the breach of his A1P1 rights.

VI The Judgment

47. The hearing took place on two days, 20 and 28 November 2014. Judgment was handed down on 13 February 2015. I have stated that the judge held that the Agency's decisions to impose the limits in the three decisions were irrational in the Wednesbury sense. He did so because he concluded they were based in part on the Exeter Report. As I have stated, he assessed the lawfulness of the 2012 decision solely in the light of the evidence and information available to the Agency at the time of the decision when the evidence of Professor Fewster and Dr Beacham was not available. As I have also stated, their evidence was also not available before the 2013 decision, and Dr Beacham's evidence was not available before the 2014 decision.
48. The judge stated that (see judgment, [65]) the flaws in the Exeter Report were such that in the case of the 2012 decision "the only rational conclusion was that [its] findings were no basis for drawing any conclusion as to the numbers of salmon caught at Lydney that would otherwise spawn in the Wye" and (see [94]) that "there was no reasonable basis for concluding that the Severn estuary putcher installations had any material effect on the Wye fishery". He accepted the submission (see [55]) that "its flaws and the incredibility of its conclusions were such that no reasonable authority could rely on it to conclude that the conditions imposed on Mr. Mott's licence were necessary to protect any fishery, which is the statutory precondition to exercise of the power to set conditions". He also stated (at [64] - [65]) that this should have been apparent on first receipt of the Exeter report, before the 2012 decision was taken, without any additional expert input because "it [did] not require any knowledge of the technical issues relating to the genetic or statistical analysis". The steps in his reasoning were:

(1) The phrase "predominantly mixed stocks" in the government policy (see [21] above) to phase out over an appropriate timescale fishing in fisheries which can be shown to exploit predominantly mixed stocks referred to a stock which is predominantly not from a single river. See [18].

(2) The Agency was entitled to consider the Exeter report, and was bound to form a considered view as to what conclusions could be properly drawn from it but was required to take account of the criticisms Mr. Mott as a layman had made of it and to consider in the light of those criticisms the report was credible or reasonably supported the conclusions drawn from it for the purposes of the decision. See [58] and [59].

(3) It was likely that, where experts had maintained their view notwithstanding criticisms by a layman who inevitably had less expertise on the technical issues addressed, the decision-maker could rely on the expert's view but this was not inevitable. In principle it was possible that a non-expert view of an expert's opinion raises matters that cast sufficient doubt that the only rational response of a decision taker is that it is not proper to rely on the expert's view, either at all or without seeking further advice. See [59].

(4) Mr. Mott's lay criticism that the Agency's conclusion that 56.4% of fish caught at Lydney could be estimated to have originated from the Wye and that a similar percentage would have otherwise returned to the Wye to spawn was obviously incredible was "well justified". This was because:

"The conclusion that fishing in the upper Severn estuary has a significantly greater effect on the River Wye than the Severn itself should have been immediately startling and have caused the Agency to examine whether it could be correct. As Mr. Mott points out in his evidence, the Agency's own estimate is that the annual salmon run in the Severn is between 10-15,000 fish. It has no estimate for the Wye, but the figure must be smaller bearing in mind that (a) it is a smaller river and (b) the Agency's assessment is that stocks in the Wye were more imperilled than those in the Severn. And yet, as he says in his first witness statement (paras 10-11) if it was the case that of the salmon present at Lydney 19 times as many were destined to return to other rivers, overwhelmingly the Wye and Usk, rather than the Severn, those rivers would be "teeming with salmon" when they plainly are not."

The judge then stated that, on the basis of the figures relied on by the Agency "at least 55,000 fish would be present in the Wye". He considered that "given the Agency's concern about low stocks in the Wye, it plainly and obviously cannot be the case that there are 55,000 salmon returning to it to spawn, let alone more" and that "[t]his impossible contradiction should have been immediately apparent to the Agency if it had taken an objective look at the implications of the conclusions they drew". See [61] [62] and [64].

(5) Mr. Crundwell's suggestion that fish from the Severn would travel straight up the river whereas those from other rivers finding themselves at Lydney might pass up and down that part of the estuary several times (and so be more likely to be caught in the putchers) before going downstream to their native river is "no more than conjecture and rationalisation long after the decisions under challenge were taken". This was not mentioned in any of the HRAs and Mr Crundwell did not set out any basis on which the size of such an effect it could be estimated. It would have to be a very large effect indeed to account for the figures taken from the Exeter report. See [62].

(6) The Agency had imposed the 2012 conditions on the basis that a particular proportion of the fish caught at Lydney would otherwise have returned to the Wye and it was necessary to impose a restriction such that the loss of that proportion of the reduced permitted total catch would have no adverse effect on the Wye fishery. The judge rejected the Agency's argument see [19]) that it was simply the high likelihood of the Severn estuary hosting a mixed stock fishery that led it to impose the 2012 conditions so that it was not necessary to know the precise proportion of the Lydney catch that would otherwise have returned to the Wye so long as some number, however, small, would have done so. See [66] - [67] and [69].

49. The judge considered (at [70]) that "the position [was] the same or a fortiori in relation to the later decisions, since nothing emerged after 2012 that would strengthen the conclusions drawn from the Exeter report, and insofar as additional expert material was later received it undermined those conclusions". He did not, however, accept the submission on behalf of Mr Mott that those reports "completely discredited" the Exeter Report.
50. As to A1P1, the judge considered the authorities, including Back v Finland (2004) 40 EHRR 1184, Tre Tractore AB v Sweden (1989) 13 EHRR 309, Axa General Insurance Ltd v Lord Advocate [2011] UKSC 46, and R (Trailer & Marina (Leven) Ltd) v Secretary of State for the Environment [2004] EWCA Civ 1580 and the distinction between "deprivation" and "control" at [82] – [91]. He stated (at [83] – [84]) that the distinction is not always obvious and that in Back's case (at [58]), where it was not obvious, the Strasbourg court had considered the matter against the general principle enunciated and established by the first sentence of A1P1: "Every natural or legal person is entitled to the peaceful enjoyment of his possessions".
51. The judge stated that there was no evidence that the Agency considered the extent of the effect on Mr Mott and his livelihood in any meaningful way at all and thus no evidence that any balanced consideration took place. The commercial nature of Mr Mott's operation would have been relevant to that consideration because by making them uneconomic to exercise he was being deprived of his livelihood and not merely of a leisure activity or the opportunity to maintain an ancient tradition. The method chosen of levelling all permitted catches down to the previous lowest meant that by far the greatest impact fell on Mr Mott whereas others who may only have used their rights for leisure or hobby purposes would be much less affected. See [97] and [98].
52. The judge concluded (at [94]) that the irrationality of the catch limits set by the Agency meant that they were a disproportionate and therefore unlawful interference with Mr Mott's right to the peaceful enjoyment of his possessions under A1P1. He further concluded (at [96]) that "the extent of the restriction imposed which eliminated at least 95% of the benefit of [Mr Mott's] right", meant that it was "to be considered as closer to deprivation than mere control". He stated that, even if the Agency could properly have imposed the conditions limiting the numbers of fish caught, "the size of that limit and the way it was apportioned would still have meant that [Mr Mott] has been required to shoulder an excessive and disproportionate burden, such that a breach of A1P1 could only be prevented by payment of compensation".
53. In a supplementary judgment on 26 February 2015, the judge allowed Mr Mott to amend his claims to include a claim in damages and made directions in respect of that claim. The judge allowed the amendment because, had an application to amend been made at an earlier stage, issues as to quantification would have been stayed pending the determination of the lawfulness of the challenged decisions, and because he was sceptical about submissions that the level of resources put into the proceedings had been affected by the fact that there was no claim for compensation. The claim had been opposed very vigorously on the issue of lawfulness. He did not therefore accept that the Agency suffered any prejudice.

VII The Agency's grounds of Appeal

54. The Agency appeals on three grounds. They are:-.

(1) The judge erred in concluding that the Exeter report provided no reasonable basis for the "catch limits" in the decisions because he relied not on any expert evidence to the contrary but exclusively on criticisms made by Mr Mott and upon his own observations and assumptions about the migratory behaviour of salmon. Both Mr Mott's criticisms and the judge's observations and assumptions about the migratory behaviour of salmon are misconceived.

(2) The judge's decision that it followed from his conclusion that the limits in the three decisions challenged were not rational meant that there was therefore a disproportionate interference with Mott's under A1P1 was also erroneous. It is submitted on behalf of the Agency that the judge erred in concluding that the conditions were "closer to deprivation that mere control" and thus required compensation to be paid to Mr Mott to ensure compliance with A1P1. On the relevant authorities, the conditions imposed were only a control on his possession and not a deprivation because his leasehold interest remained.

(3) The judge erred in his supplementary judgment in allowing Mr Mott to amend his claims to include a claim in damages for the breach of his A1P1 rights.

VIII The application to adduce further evidence

55. In an application filed on 19 November 2015, the Agency applied to rely on two witness statements "so that the Court of Appeal understands why so many of the observations and assumptions made by the High Court Judge regarding salmon migration were misconceived". The evidence was the third statement of Mr Crundwell dated 19 March 2016 and a statement dated 18 March 2016 of Peter Gough, Senior Technical Specialist in Fisheries in Natural Resources Wales. This application was opposed. At the hearing the Court stated that it had looked at the evidence de bene esse, that is without determining its admissibility. It has not been necessary for me to consider this evidence in reaching the conclusions that I reach in section VII of this judgment and I do not consider that it should be admitted.
56. The proper time for a public body which is a defendant in judicial review proceedings to explain the reasons and justification for a decision is before the hearing of the application for judicial review. It is the duty of such a body to assist the court with full and accurate explanations of all the facts relevant to the issue the court must decide, the legality of the challenged decision: see Sir John Donaldson's "cards upwards on the table" statement in R v Lancashire County Court, ex p. Huddleston [1986] 2 All ER 941 at 945, and Laws LJ in R (Quark) v Secretary of State for Foreign and Commonwealth Affairs [2002] EWCA Civ 1409 at [50]. The 2012 and 2013 HRAs were not disclosed together with the Agency's acknowledgement of service and summary grounds or exhibited to Mr Crundwell's first two statements, but were disclosed only after a specific request by Mr Mott in a letter from the Agency dated 14 November 2013 over a year after the first judicial review was filed, and two months after the second judicial review. These are key documents justifying the decision and the Agency would not have needed a resource-intensive disclosure process to realise that was so.
57. There are, of course, situations in which it is appropriate for a defendant public body to provide a later explanation amplifying the reasons for and justification of a decision previously made: see the cases digested by Fordham, Judicial Review Handbook 6th. ed. 62.4.4. Even where it is appropriate, there may be scepticism about what are essentially ex post facto reasons and justifications: see R v Westminster CC, ex p. Ermakov [1996] 2 All ER 302, 315-316.
58. As to the appellate stage, where what is at issue is the court's understanding of the regulatory environment and the factual underpinning of the public interest factors in play, there may be a case for flexible application of the rule in Ladd v Marshall [1954] 1 WLR 1489. But I consider that admitting such evidence at that stage should be exceptional and only occur where it is necessary to put the public interest factors which are needed for the determination of the appeal before the court. Moreover, in considering whether to admit such evidence, the proportionality of the application must be taken into account. Mr Crundwell's third statement contains 85 paragraphs on 25 pages and is significantly longer than his two previous statements put together. It contains in depth coverage of the topics of salmon biology and population, the Severn estuary fisheries, obligations under the Habitats Directives, principles adopted by the Agency when setting catch conditions and the HRA process for the years between 2011 and 2014. During the hearing the Master of the Rolls indicated that he considered it to be wholly disproportionate. I respectfully agree. The material in paragraphs 24 – 33 substantially explains why the Agency considers the judge's observations and assumptions were misconceived.

IX Wednesbury unreasonableness and rationality

59. The Agency's critique of the judge's conclusion that the Exeter report provided no rational basis for the catch limits in the decisions had two limbs. The first was what Mr Lewis described at the hearing as the judge's "stark contention" that there was no rational basis for concluding that the Severn estuary fisheries were mixed stock fisheries. It is true that, in paragraph 22 of the statement of facts in support of the application, in Mr Mott's first statement, and in paragraph 39 in the section containing the grounds it is contended that there is no valid scientific basis to justify the Agency's assertion that the fish are mixed stock. It was suggested that, at the hearing, the case was not argued on behalf of Mr Mott on that basis. But, whether or not that is so, the judge did not decide that it was irrational to find the stock mixed. He was considering government policies to phase out fisheries shown to exploit "predominantly" mixed stocks over an appropriate timescale. There is nothing in this point.
60. The second and principal limb of this part of the challenge concerned the effect salmon fishing in the Severn estuary had on the Wye fishery. I have stated that the judge was correct (a) to rely only on Mr Mott's criticisms of the Agency's methodology because (see [39] and [47] above) the evidence of Professor Fewster and Dr Beacham was not available at the time of the 2012 and 2013 decisions, and (b) to treat the expert evidence in the way I have described at [49] above. I consider that a very important factor in this part of the Agency's argument. I briefly consider the position had the expert evidence been considered at [81] and [82] below.
61. Mr Lewis submitted that the judge erred in concluding that the Exeter report provided no rational basis for estimating that 56.4% of fish caught in the estuary originated from the Wye and that, absent the fishing in the estuary, those fish would have returned to the Wye to spawn. He emphasised that this was only an estimate and argued that, in the light of the Exeter report and the two earlier reports by Swain and Jones, the judge was not entitled to rely on Mr Mott's lay criticisms of the Exeter report or to dismiss the evidence of Mr Crundwell, the Agency's technical fisheries expert with 21 years' experience in the field, as "conjecture" and ex post facto rationalisation. Mr Lewis also criticised the judge for ignoring (for example, at [65]) the reality of the limited way the figure of 54.6% was used and the fact that it was reasonably based on the ONCOR maximum likelihood model of estimating mixture proportions, and for assuming that because the Wye was a smaller river than the Severn, its spawning area was smaller. He submitted that it was reasonable to take the higher percentage in the ONCOR model in the light of the "precautionary" principle". Finally, he submitted that the justification for imposing restrictions is not the 56.4% but the requirements of the Habitats Directive and the risk assessment in respect of the Wye fishery.
62. On behalf of Mr Mott, Mr Hart QC and Mr Beard submitted that the judge was entitled to rely on the incredibility of the Agency's conclusion based on the Exeter report that 56.4% of the salmon taken in the Severn estuary were destined to return to the rive Wye to spawn. The evidence did not support the conclusion that the Severn estuary was a "predominantly" mixed stock fishery. The Agency did not adduce evidence to counter Mr Mott's evidence that the Agency estimated that the annual salmon run in the Severn is between 10-15,000 fish and did not make a distinction between the size of a river and the size of the spawning area which was the basis for one of Mr Lewis's criticism of [61] of the judgment. In those circumstances, the judge was entitled to proceed on the basis that the figures did not add up and, and to reach his conclusion by an educated guess. They also submitted that the issue turned on the Exeter report because the Agency had commissioned the study which led to it because the earlier studies were insufficient to justify the imposition of these "catch conditions".
63. Underlying Mr Lewis's submissions was the contention that the judge did not accord sufficient weight to the nature of the exercise or the expertise of the Agency. Before turning to those submissions, I observe that, while a regulatory body such as the Agency is clearly entitled to deploy its experience, technical expertise and statutory mandate in support of its decisions, and to expect a court considering a challenge by judicial review to have regard to that expertise, it is also necessary for the decision-maker, when faced with such a challenge, to provide a clear explanation to the court. If it does not, criticism such as that levelled against the judge in this case on behalf of the Agency (which has been expressed in strong terms) will carry less weight.
64. I referred at [56] above to the duty on a public body which is a defendant in judicial review proceedings to assist the court with full and accurate explanations of all the facts relevant to the issue the court must decide. There is a loose parallel with that duty in this context. In my judgment, the need for a defendant to have its "cards upwards on the table" is particularly important where the context is a technical or scientific one in which the defendant expects the courts to tread warily and accord a wide margin of appreciation to the decision-maker. A reviewing court needs to be given a sufficient explanation by a regulator operating in a technical or scientific area of how the science relates to its decision so that the court can consider whether it embodies an abuse of discretion or an error of law. I note that, despite the very different regulatory framework and approach to the scope of judicial review in the United States, in Kennecott Copper Corp v EPA 462 F.2d 846, 849 (1972) Judge Leventhal in the DC Circuit demanded a more enlightening explanation from agencies for their contested technical choices. He stated that "[i]nherent in the responsibility entrusted to this court is a requirement that we be given sufficient [explanation from the EPA] … so that we may consider whether it embodies an abuse of discretion or error of law". In that case, the matter was remanded to the EPA because it had failed to provide any explanation for how the science related to its decision.
65. How does the Agency's approach in the present case measure up to the "enlightening explanation" approach? The Agency had not adduced evidence to counter Mr Mott's evidence that the Agency had estimated that the annual salmon run in the Severn is between 10-15,000 fish but the judge is criticised for relying on that figure. The judge had the HRAs a week before the hearing but was only provided with the Technical case document on the second day of the hearing, some five days after the first day when the submissions on behalf of Mr Mott were made. Moreover, he was only taken to two pages of that document. He was not taken to the sections dealing with inter alia the national and EU regulatory framework those on socio-economic and cultural impacts, mixed stock issues, and changes between the 2012 and 2013 methodology. He was taken through the HRAs, but even if, as Mr Lewis submitted, they set out the essential policy prescriptions and methods sufficiently adequately, they set them out in a less accessible way because they are what Mr Lewis described as "compacted". The judge was criticised for commenting at [77] that the evidence before him "smack[ed] of a fixed determination by the Agency" to achieve the closure of Mr Mott's putcher fishery. That comment, which was not in the section of the judgment dealing with the rationality of the decisions, might possibly reflect the way the national regulatory and policy framework had been put before him.
66. I also observe that the judge referred to the fact that some evidential material had not been put before him: see [16] and [22] about the source of the policy to close mixed stock fisheries, and the Agency's strategy for sea trout and salmon fisheries. He also stated (see [14]) that no scientific evidence had been put before him to support the Agency's assumption that all or almost all adult salmon return to their native river. Mr Crundwell's second witness statement does refer (at paragraph 21) to "considerable peer-reviewed evidence" that shows that the vast majority of salmon return to the river in which they originated to reproduce (at paragraph 21) but does not identify it.
67. In all these circumstances, it is perhaps understandable that the judge sought to make an educated guess on a number of matters. I have, however, concluded that in doing so he fell into error and strayed beyond what is proper for a reviewing judge dealing with complex scientific material. Before giving my reasons, I observe that this may have been because of the matters I have referred to in the last two paragraphs or because, as Mr Lewis stated, the Agency had focussed on what it believed to be the crux of the case – a contention that there was no evidence that the Severn estuary was a mixed fishery. By the end of the first day of the hearing, however, it was clear that an important aspect of the challenge concerned whether the estimates as to the percentage of salmon that were Wye salmon and would return were robust enough to provide a rational basis for the Agency's decisions.
68. My starting point in considering the contention (put at times in very strong terms) that the judge in this case had not accorded sufficient weight to the nature of the exercise or the expertise of the Agency is the approach a judicial review court should take when considering a challenge to the decision of the designated statutory regulator that is the result of an evaluation of assessments made using scientific material as to what might happen in the future, and is in that sense predictive. Neither party drew the attention of this court to any guidance in the authorities as to how to approach such a question. There is also none in the skeleton arguments for the hearing before the judge.
69. After the hearing, the court invited the parties to make submissions on this and drew their attention to three decisions, R (British Union for the Abolition of Vivisection) v Secretary of State for the Home Department [2008] EWCA Civ 417, (2008) 105(18) LSG 24, Secretary of State for Environment, Food and Rural Affairs v Downs [2009] EWCA Civ 664, [2009] Eu LR 799, and R v Director General of Telecommunications, ex p. Cellcom [1999] ECC 314. The very helpful submissions from both parties showed that it was common ground that in principle the court should afford a decision-maker an enhanced margin of appreciation in cases, such as the present, involving scientific, technical and predictive assessments. The difference between them was that, whereas Mr Lewis's submission was that the approach in those cases applied to the Agency's decisions in this case, Mr Hart QC and Mr Beard submitted that it did not. They argued that the decisions taken by the Agency did not fall within the area to which particular deference should be paid because, as the judge had stated at [64], the identification of the contradiction between the inference drawn by the Agency from the ONCOR data and Mr Crundwell's estimate of the River Severn fish run did not require knowledge of the technical issues relating to the genetic or statistical analysis. They also submitted that, even if the decisions did fall within the area to which particular deference should be paid, in the particular circumstances of this case they could properly be subject to the sort of review carried out by the judge.
70. The judge correctly recognised (at [52]) that it was not the function of the court to form its own view as between the views of different experts in a technical area, and that he was concerned with the rationality of the decisions. He was also clearly aware (see, for example, [59]) that he was dealing with a technical area in which the professionals would have more expertise. He did not, however, address the particular issues that arise when dealing with a decision based in part on scientific assessments which involve a predictive element.
71. R (British Union for the Abolition of Vivisection) v Secretary of State for the Home Department [2008] EWCA Civ 417, (2008) 105(18) LSG 24 and Secretary of State for Environment, Food and Rural Affairs v Downs [2009] EWCA Civ 664, [2009] Eu LR 799 are decisions of this court in which the challenges had succeeded at first instance but appeals by the defendants were successful in broad terms because this court considered that the judge had substituted his own inexpert view of the science for the view of the decision-maker which was based on a tenable expert opinion.
72. The Abolition of Vivisection case concerned a challenge to a decision of the Chief Inspector of Animals that the adverse effects experienced by marmosets as a result of tests forming part of a university's research into the functioning of the human brain were "moderate" and not, as the claimant maintained, "substantial". The judgment was given by May LJ, with whom Dyson and Moses LJJ agreed. May LJ stated at [1] that a challenge to "a composite scientific judgment based upon an expert analysis of scientific material" is a type of decision that is intrinsically less amenable to a successful judicial review application than others and at [54], in the context of perversity arguments, that "an analysis of apparently competent expert scientific opinion [is] not …. a proper subject of judicial review proceedings".
73. In Downs' case, Ms Downs, an experienced campaigner with expertise and knowledge of pesticides, challenged the United Kingdom's regulatory regime for pesticides on the ground that it did not comply with the provisions of Directive 91/414 EEC because it did not properly protect residents in rural areas who were exposed to the effects of crop-spraying. The evidence included criticism by the Royal Commission on Environment Pollution in its 2005 report of a model used by the Advisory Committee on Pesticides on which the Secretary of State had relied. The judgment was given by Sullivan LJ, with whom Keene and Arden LJJ agreed. Sullivan LJ stated at [76] that "whilst the [Secretary of State's] decisions in this respect are not immune from judicial review, the hurdle of "manifest error" in such a highly technical field is a formidable one …. [Ms Downs] is not able to surmount that hurdle".
74. I have also been assisted by two other cases. In R v Director General of Telecommunications, ex p. Cellcom [1999] ECC 314 the challenged decision was based on the regulator's assessment that releasing two new network services providers in the mobile telephone market from restrictions requiring them to provide airtime on a wholesale basis to independent service providers for resale to their customers would be more likely to promote effective competition between new and well established undertakings than maintaining the restrictions. Lightman J stated (at [26]) that "[t]he court must be astute to avoid the danger of substituting its views for the decision-maker and of contradicting (as in this case) a conscientious decision-maker acting in good faith with knowledge of all the facts" and (see [78] below) this was particularly so in the case of the decision-maker's educated predictions for the future. The second case is R (Levy) v Environment Agency [2002] EWHC 1663 (Admin), which Mr Lewis relied on in his further submission. It was a challenge to a decision permitting the use of scrap tyres as a substitute fuel in kilns at a cement works. Silber J stated at [75] – [76] that he accorded an enhanced margin of appreciation to the Environment Agency, an environmental regulator making a specialist judgment, applying very sophisticated specialised scientific and environmental knowledge and expertise.
75. The contexts of these cases and the evidence before the bodies whose decisions were challenged are different from the position in the present case. In Downs' case, there were differences between different experts, and in the Animal Experiments case, the view of the decision-maker was supported by other experts. As well as those factors, the scope of judicial review is acutely sensitive to the regulatory context and, in particular, decisions involving what Professor Lon Fuller called "polycentric" questions pose particular challenges to a judicial review court. Notwithstanding the differences and recognising the importance of sensitivity to context and flexibility, I consider that these cases provide general assistance in considering the approach to the decisions of the Agency.
76. The judge was very conscious of the fact that Mr Mott's critique of the Exeter report was that of a layperson (see [61] summarised at [48(3)] above) and that it was likely that a decision-maker could rely on the views of experts who maintained their view despite lay criticism. But he then entered into an analysis of the reliability of the scientific evidence and the models used and undertook calculations of his own. He did so because (see [64]) he considered the identification of the contradiction between low stocks in the Wye and his assessment that on the material relied on by the Agency there would be 55,000 salmon returning to it to spawn, does not require any knowledge of the technical issues relating to the genetic or statistical analysis. That was, in my judgment, inappropriate. In the Downs case [2009] EWCA Civ 664 Sullivan LJ (at [46]) stated that the Royal Commission's critique of the model used by the Secretary of State was the high watermark of Ms Downs's case. Although he recognised Ms Downs was an experienced campaigner with expertise and knowledge of pesticides, she had no scientific or medical qualifications and he stated that there was no possible basis for accepting criticisms by her that went further than those of the Royal Commission.
77. More broadly, in the Abolition of Vivisection case [2008] EWCA Civ 417 May LJ stated at [1] that scientific analysis "is not immune from lawyers' analysis" but a reviewing court must be "careful not to substitute its own inexpert view of the science for a tenable expert opinion". A reviewing court should be very slow to conclude that the expert and experienced decision-maker assigned the task by statute has reached a perverse scientific conclusion. May LJ also stated at [15] that the assessment of the effect on animals of tests which were part of research into the functioning of the human brain made when a project is licensed is a predictive assessment. The dividing lines between the different categories of effect "are more a matter of scientific judgment than legal analysis". So too, in my judgment is the adequacy of a model used to estimate percentages of fish originating from a given river or whether a fishery exploits "predominantly mixed stocks".
78. In the present case the decisions were based on three principal factors. First, the Agency's assessment on the basis of the shortfall in egg deposition that the salmon fishery in the Wye is at risk of becoming unsustainable, which was not challenged. Secondly, the views of the researchers and the Agency, reflecting a broad scientific consensus, that salmon return to their rivers of origin to spawn. Thirdly, the genetic data gathered from the 55 fish taken from the estuary and the ONCOR, GeneClass2 and cBayes models used in the Exeter report to estimate their rivers of origin. The decisions were thus made against an unchallenged assessment as to the risk to the Wye and a background assumption on which there is scientific consensus that salmon return to their river of origin to spawn. The decisions were then the result of an amalgam of assessments which are in part factual and in part predictive in nature. They also involved consideration of other factors, such as how to balance the interests of those primarily affected with the wider public interest, and how factors such as the "heritage installation" aspect should be factored into the decision and are in this sense "polycentric". I respectfully agree with the statement of Lightman J in ex p. Cellcom [1999] ECC 314 at [26] that "if … the court should be very slow to impugn decisions of fact made by an expert and experienced decision-maker, it must surely be even slower to impugn his educated prophecies and predictions for the future".
79. The judge took no account of the fact that the Agency's approach, including that the Severn estuary is a mixed fishery and its use of the de minimis catch to determine the catch limit had been accepted by Natural England and by the Countryside Council for Wales, the bodies it consulted when making the HRAs. I also consider that the judge took too sceptical an approach to Mr Crundwell's evidence that Wye fish in the estuary at or near Lydney would go back down the water and into the Wye.
80. The judge's detailed critique of the models also proceeded on the basis of some errors, for example in [61] that the comparative sizes of the rivers Severn and Wye were relevant when what was relevant in estimating the annual salmon run of a river was the comparative size of the spawning areas. These errors may be the result of the Agency's failure to adduce evidence to counter Mr Mott's evidence. The judge's use of what he said was the Agency's estimate that the annual salmon run in the Severn is between 10 - 15,000 fish in his assessment that the Agency's approach, if correct, would mean that at least 55,000 fish would be present in the Wye is also an error. It is also an example of why a judge considering a judicial review of a scientific topic to that effect should not engage in a detailed examination of the merits of an approach and the accuracy of calculations based on models. The second of these figures became a major factor in the judge's conclusion that the decisions were Wednesbury unreasonable and irrational.
81. Before I leave this part of the case, I make an observation about the position had it been proper to take account of the critique of the Exeter report by Professor Fewster and of Dr Beacham's partial agreement with her. Their evidence put into question the Exeter report's figure of 56.4% of the sample taken as originating in the Wye although, in the light of Dr Beacham's report, not necessarily its view that the salmon caught in the Severn estuary net fishery are from a mixed-stock sample. The guidance obtained from Downs' case, where there were differences between different experts suggests that the judge would not necessarily have been entitled to find that the decisions were irrational in the Wednesbury sense because of their criticisms of the Exeter report. The position is that there is more than one scientific view as to the percentage, not that the Exeter report was totally discredited.
82. In Downs' case Sullivan LJ stated at [91] that where there is no scientific consensus and there are differences of view, a judge is "not entitled to substitute his own view for that of the [decision-maker] and would have been bound to conclude that there was no", in EU terms, 'manifest error' or, in common law terms, Wednesbury unreasonableness in the decision-maker's approach: see also [43]. Downs' case may differ because in the present case it is Dr Beacham, one of the experts instructed by the Agency, who disagreed with the 56.4% figure. However, in relation to Dr Beacham's suggestions for improving the confidence in assigning samples to individual rivers, Mr Crundwell in his second statement stated (at paragraphs 13-14) that most of the suggestions require significant additional baseline genetic material at considerable and disproportionate additional expense, and that the Agency is obliged to have regard to costs and benefits in exercising its functions. Sullivan LJ also stated at [48] that the fact that a model has been criticised by an expert body does not necessarily mean that it is not a suitable model: "suitability is a relative concept and a decision as to whether a model is suitable must take into account what is available in the 'real world'".

X Article 1 of Protocol 1

83. If the catch limits set by the Agency were not Wednesbury unreasonable and irrational, the determination of whether there was a breach of A1P1 is not straightforward. I proceed on the assumption that the Agency could properly have imposed the total catch limits that it did.
84. It is well-known that, in the context of A1P1, where a measure pursues a legitimate aim the question whether an interference with the right is proportionate and thus lawful is one in which the state is accorded a wide margin of appreciation. Mr Lewis submitted that the judge erred in concluding that the extent of the restriction meant that it was closer to deprivation than control, so that absent compensation it was disproportionate. He argued that Mr Mott had not been deprived of his property because he still had the right to fish. He argued that, unless the underlying right is obliterated, an interference cannot be deprivation but only the control of the exercise of the right. In this case, although the extent of the right had been controlled significantly and its value substantially diminished, if the position of the Wye fisheries changed there was a prospect that he would be able to use it more extensively in the medium to long term.
85. Mr Lewis relied in particular on the statement of Neuberger LJ (as he then was), delivering the judgment of this court in R (Trailer & Marina (Leven) Ltd) v Secretary of State for the Environment [2004] EWCA Civ 1580. The court held that there was no general doctrine that measures amounting to control of property, as distinct from deprivation, generally required compensation. This, Mr Lewis argued, was particularly so in a case of environmental control. In such a case, the court should accept that the public interest did not require that individuals should receive public funds in return for being prevented from pursuing activities that damage the environment.
86. Although it is true as Clayton and Tomlinson, The Law of Human Rights 2^nd ed., 18.104 state, that the primary criterion for establishing a deprivation of property is the extinction of all the legal rights of the owner, Mr Lewis's submissions take no account of the concept of a de facto expropriation which is recognised in the Strasbourg jurisprudence. The Strasbourg Court stated in Sporrong and Lönnroth v Sweden (1983) 5 EHRR 35 at [63] that where there is no formal expropriation:

"the court considers that it must look behind the appearances and investigate the realities of the situation complained of … Since the Convention is intended to guarantee rights that are 'practical and effective' … it has to be ascertained whether that situation amounted to a de facto expropriation …".

It has also been stated (Fredin v Sweden (No 1) (1991) 13 EHRR 784 at [41] ff) that there will be a deprivation if the owner of property is deprived of all meaningful use of it, although not if the owner remains free to sell it.

87. In this case, the restriction imposed eliminated at least 95% of the benefit of Mr Mott's right to fish using his putchers. In my judgment, although he remains able to sell his rights under the lease, the extent of the restriction imposed in this case means that the reality is, as the judge stated, that the interference is closer to deprivation than mere control. His lease is for a period of 20 years from 1 April 1998. It will thus expire some six years after the first of the challenged decisions and less than two years from now. Even if the interference is only a "control", it does not follow that because it is made on environmental grounds that any restriction can be made without compensation. In R (Trailer & Marina (Leven) Ltd) v Secretary of State for the Environment [2004] EWCA Civ 1580 the court stated that the principle of "fair balance" applies. That case concerned a challenge by the owner of a canal designated as a Site of Special Scientific Interest to sections 28 – 28Q of the Wildlife and Countryside Act 1981 as amended by the Countryside and Rights of Way Act 2000 which curtailed operations on the canal. Neuberger LJ stated at [58]:

"The right analysis seems to us to be that provided the state could properly take the view that the benefit to the community outweighs the detriment to the individual, a fair balance will be struck, without any requirement to compensate the individual. Should this not be the case, compensation in some appropriate form may serve to redress the balance, so that no breach of Article 1P1 occurs."

He, however, stated (at [65]) that, "given the purpose and genesis of the [Wildlife and Countryside Act 1981 as amended], and the jurisprudence of the ECtHR", the absence of compensation "cannot of itself justify an argument that there has been an infringement of the Article 1P1 rights of the owner of an SSSI whose value has been substantially diminished as a result of the amendments effected by the 2000 Act".

88. I agree with the judge (at [93]) that the Trailer & Marine case does not go as far as Mr. Lewis submitted. The court assumed without deciding that the uses for which the canal could be put were substantially curtailed for the foreseeable future and this had a substantial effect on its market value. The challenge was to the legislation as a whole, and not a particular executive decision made under it. Accordingly, the question of infringement of A1P1 was considered in principle. As the judge stated (at [93]), "issues as to the margin of appreciation related to the general judgment of the legislature and not the specific circumstances of the claimant". Although in that case the court was referred to the possibility that a measure could amount to a de facto deprivation of property, there was no suggestion that the Wildlife and Countryside Act 1981 in fact had that effect, either on the claimant or generally.
89. Moreover, there is no evidence that the Agency considered the extent of the effect of the condition on Mr Mott and his livelihood. As the judge stated (at [98]), in his case the method chosen of levelling all permitted catches down to the previous lowest meant that by far the greatest impact fell on him. Others, whose rights were less extensive and who may have used their putchers only for leisure or hobby purposes, would be much less affected. In those circumstances, even if the Agency could properly have imposed the total catch limit that it did, I agree with the judge that the extent of the limit and the way it was apportioned meant that, in the case of Mr Mott, a breach of A1P1 could only be prevented by payment of compensation.

XI The late amendment to the claim

90. Finally, I turn to the submission that the judge erred in allowing Mr Mott to include a claim in damages for the breach of his A1P1 rights. I reject this submission. The Agency submitted that it had suffered detriment because, had it appreciated that what was in issue was a claim for damages, it might well have devoted a different level of resource to the case. Mr Lewis stated that it might have decided to instruct leading counsel and prepared the case differently. The Agency is a public body charged with the protection of the environment. The challenge to the legality of its regulatory assessment and decisions was a major one. The judge, in what was essentially a case management decision, was entitled to conclude that, had the claim for damages been made when proceedings were instituted, it would have been dealt with after a decision had been made on the legality of the decisions.

XII Conclusion

91. If my Lords agree, the appeal on ground 1, that the judge erred in concluding that the decisions to impose the catch limits were unlawful because they were irrational in the Wednesbury sense, will be allowed but the appeals on grounds 2 and 3, that he erred in relation to Article 1 of the First Protocol of the European Convention on Human Rights and by permitting a late application to include a claim in damages, will be dismissed.

Lord Justice McFarlane:

92. I agree.

The Master of the Rolls:

93. I also agree.