The Queen (on the application of Ronald Wyatt, Chairperson of BARAD, acting in a representative capacity) v Fareham Borough Council & Ors.

There are two basic questions in this case. First, was the duty to make an “appropriate assessment” under regulation 63 of the Conservation of Habitats and Species Regulations 2017 (“the Habitats Regulations”) lawfully performed by a local planning authority when it granted planning permission for housing development on land near a European protected site in the Solent? Second, did the authority comply with its duty under section 38(6) of the Planning and Compulsory Purchase Act 2004 to determine the application in accordance with the development plan unless material considerations indicated otherwise? Neither question involves any novel issue of law. The relevant legal principles are well established and clear.

With permission granted by Lord Justice William Davis, the appellant, Ronald Wyatt, as Chairperson of Brook Avenue Residents Against Development (“BARAD”), appeals against the order of Mr Justice Jay dated 28 May 2021 dismissing his claim for judicial review of the decision of the respondent, Fareham Borough Council on 1 October 2020 to grant outline planning permission for a development of eight detached houses on land at Egmont Nurseries, Brook Avenue, Warsash. The council is the local planning authority, and the “competent authority” under regulation 7 of the Habitats Regulations. It has filed a respondent’s notice. The fourth interested party is Natural England, the “appropriate nature conservation body” under regulation 5. It too has filed a respondent’s notice. The first, second and third interested parties – Lorraine, Michael and Thomas Hanslip – are the landowners. They filed detailed grounds of resistance opposing the claim but have played no part in the appeal.

The judge rejected Mr Wyatt’s challenge on all eight grounds. Permission to appeal was granted on four of the five grounds in the appellant’s notice (grounds 1, 2, 4 and 5). The issue arising from grounds 1, 2 and 4 and the council’s respondent’s notice is whether the council failed to make a lawful “appropriate assessment” of the proposed development under regulation 63 of the Habitats Regulations, in part because it relied on the technical guidance note published by Natural England, entitled “Advice on Achieving Nutrient Neutrality for New Development in the Solent Region (Version 5 – June 2020)”, which Mr Wyatt contends is legally flawed. The issue arising from ground 5 is whether the council failed lawfully to perform its duty under section 38(6) of the 2004 Act. These two main issues are distinct and can be dealt with separately.

The site of the proposed development lies a little to the east of the mouth of the River Hamble and about 5.5km from the Solent and Southampton Water Special Protection Area (“the SPA”), which is a European protected site. Aquatic habitats for many species of plants and birds within the protected site, including the Brent Goose, are vulnerable to the excess deposition of nutrients – in particular nitrogen compounds in wastewater, which cause algal growth. New housing development can thus harm the integrity of the protected site if suitable mitigation measures are not put in place.

The application for outline planning permission was submitted in June 2018. The proposed development was the “[demolition] of existing buildings[, the construction] of eight detached houses [and the creation] of [a] paddock”. The existing use was described as “[redundant] glasshouses and nursery buildings”. The application form indicated that each dwelling would have four or more bedrooms. When the council’s Planning Committee considered the proposal in December 2018, it resolved that planning permission should be granted. Before the required section 106 agreement had been entered into and a decision notice issued, Natural England published its technical guidance note. The application came back to the committee on 19 August 2020. By then it had been amended to include mitigation measures, and Natural England had approved the nitrogen budget.

As competent authority, the council was required by regulation 63 of the Habitats Regulations to undertake an “appropriate assessment” to ensure that the development would not adversely affect the integrity of the protected site. In undertaking the “appropriate assessment” it had regard to Natural England’s advice about “nutrient neutrality” in its technical guidance note. It used average land use figures in calculating the baseline nitrogen deposition from the site, based its calculation of how much nitrogen the proposed development would produce on a national average occupancy rate for new dwellings of 2.4 persons per dwelling, and applied a 20% “precautionary buffer”.

Article 6(3) of Council Directive 92/43/EEC on the Conservation of Natural Habitats and of Wild Flora and Fauna (“the Habitats Directive”) states:

“3.

Any plan or project not directly connected with or necessary to the management of the site but likely to have a significant effect thereon, either individually or in combination with other plans or projects, shall be subject to appropriate assessment of its implications for the site in view of the site’s conservation objectives. In the light of the conclusions of the assessment of the implications for the site and subject to the provisions of paragraph 4, the competent national authorities shall agree to the plan or project only after having ascertained that it will not adversely affect the integrity of the site concerned and, if appropriate, after having obtained the opinion of the general public.”

That provision was transposed into domestic law by regulation 63 of the Habitats Regulations, “Assessment of implications for European sites and European offshore marine sites”, which states:

“(1)

A competent authority, before deciding to undertake, or give any consent, permission or other authorisation for, a plan or project which –

(a)

is likely to have a significant effect on a European site or a European offshore marine site (either alone or in combination with other plans or projects), and

(b)

is not directly connected with or necessary to the management of that site,

must make an appropriate assessment of the implications of the plan or project for that site in view of that site’s conservation objectives.

…

(3)

The competent authority must for the purposes of the assessment consult the appropriate nature conservation body and have regard to any representations made by that body within such reasonable time as the authority specifies.

…

(5)

In the light of the conclusions of the assessment, and subject to regulation 64, the competent authority may agree to the plan or project only after having ascertained that it will not adversely affect the integrity of the European site or the European offshore marine site (as the case may be).

(6)

In considering whether a plan or project will adversely affect the integrity of the site, the competent authority must have regard to the manner in which it is proposed to be carried out or to any conditions or restrictions subject to which it proposes that the consent, permission or other authorisation should be given.”

An exception to the obligation in paragraph (5) arises under regulation 64, where the authority is satisfied that there are “no alternative solutions” and that there are “imperative reasons of overriding public interest” for the project to be carried out.

There is a wealth of case law relevant to article 6(3) and regulation 63, both in the Court of Justice of the European Union (“the CJEU”) and in the domestic courts. Some basic points emerge:

Natural England’s technical guidance note was issued under section 4 of the Natural Environment and Rural Communities Act 2006, which provides, in subsection (4), that “Natural England may give advice to any person on any matter relating to its general purpose … (b) if [it] thinks it appropriate to do so, on its own initiative”.

The technical guidance note advocated the calculation of a “nutrient budget” for a proposed development. If this showed that the development was likely to generate greater levels of nitrogen than would the existing lawful use of the site, the thrust of the advice given was that the local planning authority, when granting planning permission, would have to secure appropriate mitigation measures to avoid any residual increase in nutrient levels in the Solent.

In its opening paragraph the technical guidance note recognised that the water environment of the Solent is highly protected for its habitats and species of international importance. It acknowledged that the high levels of nitrogen input to this water environment were causing excessive plant growth – “eutrophication” – in the designated sites, and that the resulting mats of green algae and other impacts on the marine ecology were affecting protected habitats and bird species (paragraph 1.1). It referred to the “potential for future housing developments across the Solent region to exacerbate these impacts [,which] creates a risk to their potential future conservation status”. It introduced “nutrient neutrality” as “a means of ensuring that development does not add to existing nutrient burdens”, adding that “this provides certainty that the whole of the scheme is deliverable in line with the requirements of [the Habitats Regulations]” (paragraph 1.3). It advocated a practical method for calculating how nutrient neutrality could be achieved, based on “best scientific knowledge” but subject to revision as further evidence was obtained (paragraph 1.4).

The “best available up-to-date evidence” indicated that some of the protected sites were “widely in unfavourable condition due to existing levels of nutrients” and “at risk from additional nutrient inputs” (paragraph 2.3). In Natural England’s view, there were likely significant effects on several internationally designated sites “due to the increase in wastewater from the new developments coming forward” (paragraph 2.4). Nutrient neutrality would allow local planning authorities to comply with their duties under regulation 63 (paragraph 2.5), and provide “a means of ensuring that development does not add to existing nutrient burdens” (paragraph 2.6).

In section 4, “Nutrient Neutrality Approach for New Development”, it was stated that “[achieving] nutrient neutrality is one way to address the existing uncertainty surrounding the impact of new development on designated sites”, and that “[this] practical methodology provides advice on how to calculate nutrient budgets and options for mitigation, should this be necessary” (paragraph 4.1). It suggested this approach to calculating “nutrient budgets” (in paragraphs 4.6 to 4.9):

“4.6

For those developments that wish to pursue neutrality, Natural England advises that a nitrogen budget is calculated for new developments that have the potential to result in increases of nitrogen entering the international sites. A nutrient budget calculated according to this methodology and demonstrating nutrient neutrality is, in our view, able to provide sufficient and reasonable certainty that the development does not adversely affect the integrity, by means of impacts from nutrients, on the relevant internationally designated sites. This approach must be tested through the ‘appropriate assessment’ stage of the Habitats Regulations Assessment. The information provided by the applicant on the nutrient budget and any mitigation proposed will be used by the local planning authority, as competent authority, to make an appropriate assessment of the implications of the plan or project on the designated sites in question. …

4.7

The nutrient neutrality calculation includes key inputs and assumptions that are based on the best-available scientific evidence and research. It has been developed as a pragmatic tool. However, for each input there is a degree of uncertainty. For example, there is uncertainty associated with predicting occupancy levels and water use for each household in perpetuity. Also, identifying current land/farm types and the associated nutrient inputs is based on best-available evidence, research and professional judgement and is again subject to a degree of uncertainty.

4.8

It is our advice to local planning authorities to take a precautionary approach in line with existing legislation and case-law when addressing uncertainty and calculating nutrient budgets. This should be achieved by ensuring nutrient budget calculations apply precautionary rates to variables and adding a precautionary buffer to the [total nitrogen] calculated for developments. A precautionary approach to the calculations and solutions helps the local planning authority and applicants to demonstrate the certainty needed for their assessments.

4.9

By applying the nutrient neutrality methodology, with the precautionary buffer, to new development, the competent authority may be satisfied that, while margins of error will inevitably vary for each development, this approach will ensure that new development in combination will avoid significant increases of nitrogen load to enter the internationally designated sites.”

For development which would drain to the mains network, the suggested method would involve four stages. In the first stage, which was to calculate the total nitrogen derived from the development which would leave wastewater treatment works, the first step was to “Calculate [the] additional population” arising from the development. Relevant here is the advice given on occupancy rates:

“4.18

New housing and overnight accommodation can increase the population as well as the housing stock within the catchment. This can cause an increase in nitrogen discharges. To determine the additional population that could arise from the proposed development, it is necessary that sufficiently evidenced occupancy rates are used. Natural England recommends that, as a starting point, local planning authorities should consider using the average national occupancy rate of 2.4, as calculated by the Office for National Statistics (ONS), as this can be consistently applied across all affected areas.

4.19

However competent authorities may choose to adopt bespoke calculations tailored to the area or scheme, rather than using national population or occupancy assumptions, where they are satisfied that there is sufficient evidence to support this approach. Conclusions that inform the use of a bespoke calculation need to be capable of removing all reasonable scientific doubt as to the effect of the proposed development on the international sites concerned, based on complete, precise and definitive findings. The competent authority will need to explain clearly why the approach taken is considered to be appropriate. Calculations for occupancy rates will need to be consistent with others used in relation to the scheme (e.g. for calculating open space requirements), unless there is a clear justification for them to differ.”

The second step in the first stage was to “Confirm water use”. In Natural England’s view, planning authorities ought to impose conditions for maximum water usage of “110 litres per person per day” on new developments (paragraphs 4.11 and 4.22). The advice here was that “[the] water use figure is a proxy for the amount of wastewater that is generated by a household”, that “[new] residential development may be able to achieve tighter water use figures” (paragraph 4.23), and that “while new developments should be required to meet the 100 litres per person a day standard, the risk of standards slipping over time and the uncertainty inherent in the relationship between water use and sewage volume should be addressed by the use in the calculation of 110 litres per person per day figure” (paragraph 4.25).

The third step in the first stage involved identifying the “[wastewater] treatment works” into which water from the development would drain. Natural England adopted a precautionary approach, stating that “[where] there is a permit limit for Total Nitrogen, the load calculation will use a worst case scenario that the [wastewater treatment works] operates at 90% of its permitted limit” (paragraph 4.29).

The fourth step in the first stage was to “Calculate Total Nitrogen (TN) in Kg per annum that would exit the [wastewater treatment works] after treatment derived from the proposed development”. It was noted that “[natural] reductions in nitrogen concentrations, mainly through de-nitrification processes, also occur within watercourses”. But there was “[insufficient] evidence … to properly evaluate de-nitrification rates within the greater Solent catchments”; so that factor was not included. Natural England took the view that this provided “an additional precautionary factor for the methodology” (paragraph 4.42).

The second stage was to “Adjust nitrogen load to account for existing nitrogen from current land use”. This advice was given:

“4.45

This next stage is to calculate the existing nitrogen losses from the current land use within the redline boundary of the scheme. The nitrogen loss from the current land use will be removed and replaced by that from the proposed development land use. The net change in land use will need to be subtracted from or added to the wastewater Total Nitrogen load.

4.46

Nitrogen-nitrate loss from agricultural land can be modelled using the Farmscoper model. …

4.47

If the development area covers agricultural land that clearly falls within a particular farm type used by the Farmscoper model then the modelled average nitrate-nitrogen loss from this farm type should be used. …

…

4.51

It is important that farm type classification is appropriately precautionary. It is recommended that evidence is provided of the farm type for the last 10 years and professional judgement is used as to what the land would revert to in the absence of a planning application. In many cases, the local planning authority, as competent authority, will have appropriate knowledge of existing land uses to help inform this process.

4.52

There may be areas of a greenfield development site that are not currently in agricultural use and have not been used as such for the last 10 years. In these areas as there is no agricultural input into the land a baseline nitrogen leaching value of 5 kg/ha should be used. This figure covers nitrogen loading from atmospheric deposition, pet waste and nitrogen fixing legumes.”

The third stage was to adjust the nitrogen load to account for land uses in the proposed development.

The fourth stage was to “Calculate the net change in the Total Nitrogen load that would result from the development”. The advice was this:

“4.67

It is necessary to recognise that all the figures used in the calculation are based on scientific research, evidence and modelled catchments. These figures are the best available evidence but it is important that a precautionary buffer is used that recognises the uncertainty with these figures and in our view ensures the approach prevents, with reasonable certainty, that there will be no adverse effect on site integrity. Natural England therefore recommends that a 20% precautionary buffer is built into the calculation.

4.68

There may be instances where it is the view of the competent authority that an alternative precautionary buffer should be used on a site-specific basis where sufficient evidence allows the legal tests to be met.”

Since the judge’s decision in the court below, Natural England has, in March 2022, issued further guidance. This does not bear on the claim with which we are concerned.

On 9 June 2020, as statutory consultee under regulation 63(3), Natural England gave its advice to the council on “nutrient neutrality” for the proposed development. It did so in the light of the council’s “Nitrogen Budget”, dated 11 May 2020. The “Nitrogen Budget” was based on an occupancy rate of 2.4 persons per dwelling and included a precautionary buffer of 20%, both of which were subsequently used in the council’s appropriate assessment. Natural England said that “[provided] the council, as the competent authority, [was] assured and satisfied [that] the site areas [were] correct and that the existing land uses [were] appropriately precautionary”, it raised “no further concerns with regard to the nutrient budget”. Nor did it raise concern about the use of average land use figures for calculating the baseline nitrogen deposition from the site, about the 2.4 occupancy rate, or about the 20% precautionary buffer applied.

Mr Wyatt and his wife submitted several letters of objection to the council. In his “further comments” dated 15 June 2020, Mr Wyatt addressed the use of the 20% precautionary buffer, arguing that it appeared “irrational” because there was “no evidential basis explaining why a 20% buffer has been used”. He also expressed his concern about the use of the occupancy rate of 2.4 persons per dwelling. He noted that the council had used its “discretion to vary this figure” when considering a proposal of “16 age related apartments” in Station Road, Portchester, for which it had “used what [it] termed an overall “cautious average” occupancy rate of 2”, which was “in line with the 2011 Census figure”. He expected the council to be consistent. The 2011 Census gave an average occupancy rate of 3.4 persons per household for houses of the size proposed, which would be “a more appropriate figure”. If the council was “consistent” and used “the correct land use figures and a more realistic occupancy rate”, it would “reject the application on the grounds that it will be in deficit and therefore cannot meet the nitrate neutrality regulations”.

Natural England gave further advice to the council on 18 August 2020, now in the light of the council’s draft appropriate assessment. It did not doubt the conclusions of the draft appropriate assessment or the likely efficacy of the proposed mitigation measures. Again, it raised no concerns about the use of average figures, the occupancy rate of 2.4 persons per dwelling, or the use of the 20% precautionary buffer.

In the appropriate assessment presented to the council’s Planning Committee on 19 August 2020, it was acknowledged (on p.2) that “[all] new housing development within 5.6 km of the Solent SPAs is considered to contribute towards an impact on the integrity of the Solent SPAs”, and (on p.4) that “[the] proposed development is within 5.6 km of the Solent & Southampton Water SPA”. The likely nitrogen output of the proposed development was identified, and the proposed mitigation measures described and considered. These conclusions were stated (on p.17):

“The project being assessed will result in a positive nitrogen output of 10.5 kg/TN/yr and therefore the waste water from the development will add to the nitrogen levels within the Solent. … The pathway is via the wastewater treatment works. Therefore, the surplus in the nitrogen output would need to be mitigated. … In order for the development proposal to demonstrate nitrogen neutrality, an on-site wetland will be created on site. The proposed wetland would remove nitrates from surface water and roof water drainage through a combination of physical, chemical and biological processes via interactions between the water, substrate and micro-organisms such as algae. The wetland would in turn provide a reduction of 11.51 kg/N/yr meaning there would be an overall reduction in nitrates being discharged from the site. The mitigation will be secured through a Section 106. …”.

and (on p.18):

“In conclusion, the application will have a likely significant effect in the absence of avoidance and mitigation measures on [the protected sites] … This represents the authority’s Appropriate Assessment as Competent Authority in accordance with requirements under Regulation 63 of the [Habitats Regulations], [and] Article 6 (3) of the Habitats Directive … .

The authority has concluded that the adverse effects arising from the proposal are wholly consistent with, and inclusive of the effects detailed in the Solent Recreation Mitigation Strategy. The authority’s assessment is that the proposed mitigation package complies with this Strategy and that it can therefore be concluded that there will be no adverse effect on the integrity of the Solent and Southampton Water SPA.”

In his report to the committee, the officer considered the possible impact of the development on the European protected sites in the Solent, under the requirements in regulation 63 of the Habitats Regulations and in the light of the appropriate assessment. He reminded the members that “[regulation 63] provides that planning permission can only be granted by a ‘competent authority’ if it can be shown that the proposed development will either not have a likely significant effect on designated [protected sites] or, if it will have a likely significant effect, that effect can be mitigated so that it will not result in an adverse effect on the integrity of the designated [protected sites] …” (paragraph 8.26). He referred to Natural England’s advice, explaining the concept of “nutrient neutrality” and the need for local planning authorities to take a “precautionary approach” (paragraphs 8.32 and 8.33); to the “nutrient budget” (paragraph 8.34); and to the existing land use (paragraphs 8.35 to 8.37).

He then came to the “assumed occupancy rate” (in paragraphs 8.38 to 8.42):

“8.38

Natural England recommends that, as a starting point, local planning authorities should consider using the average national occupancy rate of 2.4 persons per dwelling as calculated by the Office for National Statistics (ONS), as this can be consistently applied across all affected areas. However competent authorities may choose to adopt bespoke calculations where they are satisfied that there is sufficient evidence to support this approach.

8.39

Concern has been raised by third parties over the use of the average occupancy rate of 2.4 for this development of eight houses. Some have expressed the view that a higher occupancy rate ought to be applied since the houses are likely to be larger than average dwellings (although it should be noted that the application is in outline form and scale and layout of the development are reserved matters). Third parties have noted that the Council used bespoke calculations when determining a recent planning application for a sheltered housing development elsewhere in the Borough.

8.40

It is acknowledged that some houses will have more than the average number of occupants. It is also of course the case that some will have less. The figure of 2.4 is an average based on a well evidenced source (the ONS) and which has been shown to be consistent over the past ten years. As stated above the Natural England methodology allows bespoke occupancy rates however to date the Council has only done so to lower, not raise, the occupancy rate and where clear evidence has been provided to demonstrate that the proposed accommodation has an absolute maximum rate of occupancy. In the case of sheltered housing which is owned and managed by the Council for example it has been previously been considered appropriate to apply a reduced occupancy rate accordingly.

8.41

In all instances it is the case that the Natural England methodology is already sufficiently precautionary because it assumes that every occupant of every new dwelling (along with the occupants of any existing dwellings made available by house moves) is a new resident of the Borough of Fareham. There is also a precautionary buffer of 20% applied to the total nitrogen load that would result from the development as part of the overall nutrient budget exercise.

8.42

Taking the above matters into account, Officers do not consider there to be any specific justification for applying anything other than the recommended average occupancy rate of 2.4 persons per dwelling when considering the nutrient budget for the development.”

The judge had before him in evidence a witness statement, dated 25 February 2021, of Dr James O’Neill on behalf of Mr Wyatt, and three witness statements, dated 9 December 2020, 4 February 2021 and 12 April 2021, of Ms Allison Potts on behalf of Natural England. Dr O’Neill is the Principal of James O’Neill Associates, an environmental consultancy. Ms Potts is the Acting Area Manager of Natural England’s Thames Solent team.

In his witness statement, Dr O’Neill said that “[if] an incorrect occupancy rate is used then it will cause the total nitrogen figure … to be wrong”, and lead to a figure “which has a real risk of significantly underestimating and therefore downplaying the actual nitrogen output of the development in question”. Natural England suggested that an occupancy rate of 2.4 persons per dwelling should be used as a starting point, but that a “bespoke calculation” would be appropriate in some cases (paragraph 20). However, in Dr O’Neill’s view, as the proposed dwellings would have four or five bedrooms, the national average occupancy rate would not represent best scientific evidence available. A “specific dataset for four to five bedroom dwellings” was available for Fareham, which would have given an average occupancy rate of 3 for such dwellings, not 2.4, broadly in alignment with the national average of 3.14. The use of the national average was not, therefore, justifiable (paragraphs 23 to 29).

On the “use of averages in the land classification”, Dr O’Neill said the “selection of the correct land use for the site is a matter of judgement which [he was] not qualified to make an assessment of here” (paragraph 36). But he made four points: first, existing land use figures should be “sufficiently precautionary” (paragraph 38); second, there was “reasonable doubt in respect of the land classification employed” in the appropriate assessment (paragraph 39); third, the “Farmscoper model” relied on average data, rather than using site-specific data; and fourth, the inaccuracy introduced by the use of the 2.4 occupancy rate would be compounded by the use of average land use figures (paragraph 42).

Finally, he addressed the 20% precautionary buffer. He said that, “for such a buffer to be valid, the level of uncertainty associated with each step of the calculation must be known” (paragraph 47). There was a range of statistical methods for quantifying that uncertainty, but “no evidence that they have been applied in respect of the impugned calculation”. The buffer applied was “insufficiently precautionary” (paragraph 48).

Ms Potts did not agree with Dr O’Neill’s opinion. In her evidence she said that his “focus on the 2.4 figure takes no account of the precaution that is built into the methodology as a whole” (third witness statement, paragraph 5). Natural England had considered the available data and concluded that the national average occupancy rate was “the best available scientific evidence for use in the methodology when applied to development within the Solent catchments.” (third witness statement, paragraph 7).

Seven reasons were given for that conclusion. First, the 2.4 figure was often used, and it would only be necessary for a local planning authority to adopt bespoke figures in an “extreme occupancy scenario”. The fact that a development consisted of larger houses was not in itself enough to warrant the adoption of a bespoke rate (second witness statement, paragraphs 23 to 25). Second, Natural England had concluded that reliance on the “finer grain detail” would have introduced “unnecessary and unwieldy complication”. It “would have required using 65 different occupancy rates across the area (13 ONS areas x 1-5+ bedroom rates)”, and it would also have been “necessary to use a per bedroom water usage rate”. These figures were “not easily obtainable” (third witness statement, paragraph 8). Third, Natural England had assumed “100% inward migration”, whereas in reality “some occupants of new dwellings will be moving within the affected catchments, so do not represent an entirely new burden” (second witness statement, paragraph 31, and third witness statement, paragraph 9). Fourth, “while larger properties tend towards having more occupants than smaller properties (but not in a linear relationship to the number of bedrooms), occupancy and dwelling size are not very highly correlated” (second witness statement, paragraph 26). Fifth, the occupancy rate of 2.4 persons per dwelling was part of a broader “strategic solution”, and a “standardised approach” (second witness statement, paragraphs 32 and 33). Sixth, the data showed that houses with higher occupancy rates had significantly lower water use figures per occupant (second witness statement, paragraph 27). And seventh, the data was “suggestive of a decline in average occupancy over time”. The 2.4 occupancy rate was meant to account for “the nutrient impact arising from the proposed development in perpetuity” (second witness statement, paragraphs 28 to 30).

Ms Potts drew attention to three “precautionary” elements in the method used for estimating “water use”: first, “[the] water use figure used (110 l/p/d) is 10% higher than Southern Water’s target required …”; second, “[all] new build development will have meters and water use in metered properties is significantly lower than non-metered properties”; and third, “[water] supply is less than water return to [wastewater treatment works] reflecting use of water to wash cars, water gardens”. She also referred to two “precautionary” elements in the consideration of “Wastewater Treatment Work Operations”: first, that it “[assumes wastewater treatment works with total nitrogen] permits operate at the maximum possible within legal limits”, whereas the “Solent [wastewater treatment works] with [a total nitrogen] permit are currently on average performing 25% more effectively than [the] assumed level”; and second, that “[an] unknown proportion of nitrogen discharged seaward of the international sites will be lost to sea and will not affect the designated sites” (second witness statement, Table 1).

Ms Potts also referred to the 20% precautionary buffer, and explained in detail how the correct figure was arrived at. She said the development of the buffer had “involved consideration of the likelihood, severity, duration and tendency of potential impacts”, and “in determining the level of the buffer, each component was assessed individually, as well as evaluating the relationships between each component, the risk of exceedances and the severity of such exceedance”. Among the factors taken into account were “the degree of known variability for each component” and “the fact that not all risks are fully known”. Ms Potts emphasised that defining the buffer had involved “expert judgement”, and the choice of 20% as the appropriate figure was considered to be commensurate with “no reasonable doubt” about the absence of adverse effects on the integrity of the protected site (second witness statement, paragraphs 56 to 64, and third witness statement, paragraphs 19 to 21).

Jay J. was critical of the approach to occupancy rates in Natural England’s technical guidance note, and of the council’s use of an occupancy rate of 2.4 persons per dwelling in this case. But adopting the degree of deference he thought right in the circumstances, and approaching the matter on a Wednesbury basis, he concluded that the use of the 2.4 occupancy rate was sufficiently precautionary. He concentrated, in particular, on two “precautionary elements” of the appropriate assessment that could “legitimately be brought into account”: first, that “the relationship [between occupancy rates and water usage] is not one of direct proportionality”, and second, that “the algorithm assumes 100% migration to the area” (paragraph 84 of his judgment). He was “satisfied that there was an adequate precautionary leeway afforded by [these] two key factors” (paragraph 86). He added, however, that the technical guidance note would need to be reviewed in the light of his judgment (paragraph 87).

The judge did not accept that the use of average land use figures was inappropriate, or that site specific measurements should have been taken. He thought that site specific measurements would provide “no more than a snapshot of existing land use”, and it was not clear that “the overly rigorous approach recommended by Dr O’Neill would in fact yield more protective data” (paragraph 110).

On the use of the 20% precautionary buffer, the judge concluded that the lack of “any arithmetical calculation or other algorithm” in the calculation of the buffer was not fatal to it. He thought that there was “room for debate between reasonable scientists, using their judgment, expertise and experience, as to whether the figure should be, say, 10%, 20% or 30%”. And he found “no place for judicial intervention on any Wednesbury basis” (paragraph 111).

Those were the judge’s principal conclusions on this part of the case. I shall also refer to some other passages in his judgment when I come to the argument put forward on the first main issue.

Mr Gregory Jones Q.C., for Mr Wyatt, submitted that the judge made two fundamental errors in his approach to the legal framework governing appropriate assessment. First, he accepted Ms Potts’ evidence on the soundness of the method used by the council in conducting the appropriate assessment. He ought to have looked at the underlying evidence and considered, for himself, whether the figures used were sound. This, submitted Mr Jones, follows from the CJEU’s judgment in Dutch Nitrogen, in particular at paragraph 101. He accepted that the court must adopt a Wednesbury approach, but he submitted that the approach should be more stringent given the high level of certainty required under regulation 63. Given that much of the water environment in the Solent was in unfavourable or failing status, the level of certainty required was higher. A distinction must be made between evidence considered by decision-makers at the time, and expert evidence produced later in explanation. Secondly, the judge had erred in his approach to the “precautionary principle”. He should have accepted that where data is uncertain, the “reasonable worst-case scenario” must be assessed. This follows from the European Commission’s guidance on the precautionary principle, cited with approval in Bayer CropScience v Commission (Case T-429/13), and it would be consistent with the approach taken in environmental impact assessment (see R. v Rochdale Metropolitan Borough Council, ex parte Milne (No.2) (2001) 81 P. & C.R. 27).

On ground 1 of the appeal Mr Jones argued that Natural England’s technical guidance note invited error when it said local authorities “may choose to adopt bespoke calculations” for occupancy rates. In particular, he criticised the first sentence of paragraph 4.7 of the technical guidance note, the first sentence of paragraph 4.19, and paragraph 4.42. It would never be permissible for an authority to adopt the national average occupancy rate unless it was the correct occupancy rate for the development proposed. Authorities must adopt bespoke calculations. Otherwise, their decisions will not be based on the “best scientific knowledge”. The correct occupancy rate was not a matter of expert judgment; it was a simple and readily ascertainable fact. In this case it was common ground that the occupancy rate of 2.4 persons per dwelling was inaccurate, and that an occupancy rate of 3 would have been accurate for the four-bedroom houses proposed. The council’s decision to adopt an occupancy rate of 2.4 was therefore wrong, and unlawful. Each factor in the calculation of the nitrogen budget had to be precautionary, and based on the best available evidence. It was not permissible to rely on the precautionary nature of other factors, or on the precautionary buffer applied at the end, to justify using an insufficiently precautionary occupancy rate. Once the judge had concluded that using an occupancy rate of 2.4 did not represent the “best scientific knowledge”, he could not hold that its use was lawful (see Holohan, at paragraph 33). He should not have found it sufficiently precautionary on the strength of the two factors mentioned by Ms Potts; there was no evidence that they would counteract the error. It was also inconsistent to conclude, as he did, that Natural England’s technical guidance note would need to be reviewed in the light of his judgment but that the decision in this case, based on the advice given in that document, was nonetheless sound.

On ground 2 Mr Jones criticised the use of average figures, and, in particular, the use of “average land use figures” in calculating the baseline nitrogen deposition from the site. Average figures relied on speculation about what might happen in the future, and so were necessarily contrary to the requirement for certainty under regulation 63. In Dutch Nitrogen the CJEU had made it clear that reliance on average values was impermissible (paragraphs 55 and 147 of the Advocate General’s opinion, and paragraph 119 of the judgment). Using average land use figures to calculate the baseline nitrogen deposition from the site was insufficiently precautionary. Both in adopting these average figures and in its use of the Farmscoper model, Natural England’s advice in paragraphs 4.45 to 4.52 of the technical guidance note was flawed, and so was the council’s application of that advice in the appropriate assessment.

On ground 4 Mr Jones argued that the 20% precautionary buffer applied by the council was unlawful, because it lacked any evidential basis. The purpose of the buffer was not merely to provide an extra level of protection but to ensure that the whole exercise met the required standard of scientific certainty. To remove “all reasonable scientific doubt” about the effects of the proposed works on the protected site concerned”, the uncertainty inherent in the initial steps must first be quantified, and then an appropriate buffer applied in light of that uncertainty (see People Over Wind, at paragraph 38). Natural England’s relevant advice (in paragraphs 4.8, 4.9 and 4.67 of the technical guidance note) was flawed, and so was the council’s application of that advice in the appropriate assessment.

Those arguments all go to the contention that the council erred in law when performing its duty under regulation 63(5) of the Habitats Regulations not to grant planning permission unless it had ascertained that the proposed development would “not adversely affect the integrity of the European site”. I do not accept that contention. The council’s conclusion on the crucial question under regulation 63(5) was, ultimately, an evaluative judgment for it to make as “competent authority”. And in my view the conclusion it reached, as a matter of evaluative judgment, was legally sound. I therefore agree with the decision in the court below on this part of the claim.

I cannot fault the judge’s self-direction on the relevant legal principles (in paragraphs 29 to 39 of his judgment), and in my opinion he went on to apply those principles appropriately. I do not think he made the fundamental errors of which he is accused.

The first of those alleged errors, essentially, is that the judge simply accepted the evidence of Ms Potts without question. I do not think he did that. On the contrary, he examined in appropriate depth and detail the evidence of the expert witnesses on either side (in paragraphs 58 to 72, and paragraphs 106 to 111).

On occupancy rates, he approached the evidence before him with care. He expressed his own concerns about some of that evidence (in paragraphs 75 to 80). He did not rely on the parts he found less than convincing (paragraph 84). He reminded himself of “[the] need for judicial deference in a domain of technical and scientific expertise” (paragraph 81). He acknowledged that the figures used by the competent authority must “have the effect of removing all scientific doubt “based on complete, precise and definitive findings”” (paragraph 79).

Nor did he accept unquestioningly the use of the 20% precautionary buffer, in the way in which it had been applied, merely because Ms Potts said that this was appropriate. In her evidence, she had explained, at length, the justification for using the 20% buffer (her second witness statement, paragraphs 56 to 64, and her third witness statement, paragraphs 18 to 21). She did not simply assert that its use was correct. And the judge, for his part, did not simply take her evidence at face value. He considered the reasons she had given in support of the 20% buffer, and he concluded, in the light of that evidence, that there was no justification for the court’s intervention “on any Wednesbury basis” (paragraph 111). I agree.

More generally, it seems to me that the judge adopted the correct approach in his consideration of the council’s appropriate assessment as a whole. He understood that the Wednesbury standard of review had to be deployed with suitable rigour in the legislative context here. He knew that he had to establish whether, in all the circumstances, the council had reached a reasonable and lawful conclusion, as a matter of its own exercise of evaluative judgment, in ascertaining whether the high threshold set by regulation 63(5) had been surmounted. He applied an appropriately intense standard of scrutiny, consistent with the proper application of Wednesbury principles in the light of the jurisprudence to which he had referred.

I reject the submission that the judge ought to have given greater weight than he did to the unfavourable status of the water environment in parts of the Solent. This was a fact explicitly acknowledged and taken into account by Natural England when issuing the advice in its technical guidance note – advice on which the council relied in its appropriate assessment. And there is no support either in the habitats legislation itself or in the relevant authorities for the proposition that the unfavourable status of a protected site raises the level of certainty which has to be achieved if the proposed development is to be approved, or for the proposition that the standard of review the court should adopt in those circumstances is more demanding. In this case, Natural England’s technical guidance note, to which the council had regard in undertaking the appropriate assessment, took into account the fact some of the protected sites in the Solent were “widely in unfavourable condition due to existing levels of nutrients” and “at risk from additional nutrient inputs” (paragraph 2.3).

Whatever the particular circumstances in a given case, the basic duty of the competent authority under regulation 63 is, and remains, to grant planning permission only if satisfied that the proposed development “will not adversely affect the integrity” of the European protected site. The duty of the court is, and remains, to ensure that the authority’s evaluative judgment on that question was lawfully exercised.

In doing that, the court must keep in mind the difference between evidence of what was considered by a decision-making authority at the time of its decision and evidence put forward after the event to explain or justify that decision (see R. (on the application of United Trade Action Group Ltd.) v Transport for London [2021] EWCA Civ 1197). It is trite, for example, that later evidence of a decision-maker’s thinking cannot be used to contradict the original reasons given or to provide wholly new reasons (see R. v Westminster City Council, ex parte Ermakov [1996] 2 All E.R. 302, and Inclusion Housing Community Interest Company v Regulator of Social Housing [2020] EWHC 346 (Admin), at paragraph 78). But that has not been done in this case. Ms Potts’ evidence goes no further than to amplify the reasons why Natural England decided to adopt the approach it did, and reached the view it did, at the time of its consultation by the council. The evidence was properly admitted, and the judge was entitled to rely on it as he did.

As for the second fundamental error of which the judge is accused, I do not think he adopted too lax an understanding of the precautionary principle, either generally or as it applied in this case, or that he wrongly discounted the concept of the “reasonable worst-case scenario”, contrary to the CJEU’s reasoning in Bayer CropScience and the High Court’s in ex parte Milne. In Bayer CropScience the CJEU cited the European Commission’s guidance, “Communication on the precautionary principle” (2000) Annex III, which advises that in cases of doubt a “worst-case” hypothesis “could” – not must – be assessed (paragraph 114 of the judgment). But it did not treat the guidance as if it had the status of law. It adopted the established approach, consistent with its own judgment in Waddenzee. Nor does the principle referred to in ex parte Milne – that a proposal requiring environmental impact assessment must be sufficiently detailed to allow for proper assessment – bear on the question here, which is whether any uncertainty in the data involved in an appropriate assessment under regulation 63 must always be resolved by using a “reasonable worst-case scenario”. In Waddenzee, as Jay J. said (in paragraph 32 of his judgment), the CJEU accepted that national authorities do not need to be “absolutely certain” that there will not be adverse effects on the integrity of the protected site, but must be “satisfied that there is no reasonable doubt as to the absence of adverse effects”. The judge asked himself “whether “reasonable worst case scenario” is an apt synonym for “precautionary””, but he did not think it was necessary to come to a decisive view on the point (paragraph 47). I do not think he needed to do so. In my view it was legitimate for him to conclude that, at least in this case, the “reasonable worst-case scenario” did not have to be assessed if the precautionary principle was to be satisfied.

Turning to ground 1 of the appeal, I do not think there can be any proper challenge in these proceedings to the lawfulness of the advice given by Natural England in its technical guidance note, which seems to have been the real target for much of the argument advanced on behalf of Mr Wyatt.

It should be remembered that the technical guidance note is not statute. It does not create some additional legal requirement or test. It is an advisory document, which is neither mandatory in effect nor prescriptive of a single correct procedure to be followed. It contains guidance, whose purpose is to assist competent authorities in performing their functions under the habitats legislation. It does not assert that the approach it suggests is the only means of conducting an appropriate assessment. On the contrary, it expressly acknowledges that this approach is only “a means” or “one way” of undertaking that task (paragraphs 1.3, 2.6 and 4.1).

The Supreme Court has recently confirmed that there are only limited grounds on which a policy can be challenged as itself being unlawful (see R. (on the application of A) v Secretary of State for the Home Department [2021] UKSC 37, and also the recent decision of this court in R. (on the application of Pearce) v The Parole Board [2022] EWCA Civ 4). In R. (on the application of A) Lord Sales and Lord Burnett C.J. stressed that it is “not the role of policy guidance to eliminate all uncertainty regarding its application and all risk of legal errors” (paragraph 34). The appropriate question for the court is this: “does the policy in question authorise or approve unlawful conduct by those to whom it is directed?” (paragraph 38).

Where Natural England’s advice on the appropriate occupancy rate is concerned, the answer to that question would clearly be “No”. At the level of generality at which the technical guidance note was suggesting it, the use of an occupancy rate of 2.4 persons per dwelling cannot be said to be unlawful on the ground that it is inconsistent with the “best scientific evidence”. The technical guidance note did not misstate the legal position under regulation 63 (see R. (on the application of A), at paragraphs 46 and 47). It did not “authorise or approve”, let alone prescribe, the use of that occupancy rate by all local planning authorities in every case, regardless of the circumstances. It did not remove or reduce the onus on those authorities to be sure, beyond “all reasonable scientific doubt”, that the integrity of the protected site would not be adversely affected (see paragraphs 1.4, 2.5, 4.6 to 4.9, and 4.18 to 4.19 of the technical guidance note).

Nor do I accept the criticism made of the council’s use of an occupancy rate of 2.4 persons per dwelling in the particular circumstances of this case. Although an appropriate assessment must be based on “best scientific knowledge”, the question for the court is not whether each individual figure used in it is intrinsically the “best scientific knowledge” when considered on its own, divorced from the full context in which it is used. As Mr David Elvin Q.C. submitted for Natural England, the court must take a “holistic” view on the question whether the assessment methodology as a whole represents “best scientific knowledge”.

When that is done here, it is, I think, plain that the council understood its duty under regulation 63 correctly. This much is clear from the summary of the law which the officer set out in his report (in particular, at paragraph 8.26), and from the equivalent summary in the appropriate assessment itself (in particular, at pp.15 to 19).

The council consulted Natural England twice. As the judge said (in paragraph 81 of his judgment), Natural England had specifically considered “the application of more size-sensitive datasets but rejected the need for [that]”. It does not seem to have intended that the occupancy rate of 2.4 persons per dwelling should always be only a “starting point”. It evidently took the view that there were sound reasons in consistency, given the nature and availability of other datasets, to use that occupancy rate for development in the Solent (Ms Potts’ third witness statement, paragraphs 7 and 8). This was, on the face of it, a carefully considered judgment. And in any event, the council’s committee considered objections to the use of an occupancy rate of 2.4 for the proposed development, but rejected them in the light of Natural England’s response to consultation. Tellingly, Natural England did not oppose the use of that occupancy rate in this particular case, rather than the adoption of a bespoke figure. It had seen the council’s nitrogen budget before responding to consultation, and it knew therefore that an occupancy rate of 2.4 was being used in that nitrogen budget. Had it been concerned about this, one would have expected it to make that clear, but it raised no such concern. And as Mr Timothy Mould Q.C. submitted for the council, compelling reasons would have been required for the council to depart from Natural England’s position.

In the circumstances it was, I think, open to the council to rely on the precautionary nature of several factors in the nitrogen budget to ground its own judgment that the use of an occupancy rate of 2.4 persons per dwelling was consistent with a sufficiently precautionary approach in this instance.

The judge recognised the strength in two of the points made by Ms Potts in her evidence – that the relationship between occupancy rates and water usage was “not one of direct proportionality”, and that the algorithm “assumed 100% migration to the area” (paragraph 84 of the judgment). He did not confine himself to reliance on these two reasons alone – he merely said (in the same paragraph) that these two reasons “have force” and that he found the other reasons “less persuasive”.

There were, I think, at least six other factors identified by Ms Potts in her second witness statement which, in combination with the two considerations on which the judge focused, were capable of justifying the conclusion that the use of the 2.4 occupancy rate would be consistent with the precautionary principle here. First, the water use figures for the proposed development were themselves precautionary. They were “10% higher than Southern Water’s target required”, and they did not take into account the fact that “[all] new build development will have meters and water use in metered properties is significantly lower than non-metered properties”. Second, the water use figures were based on “water supply”, which would in fact be “less than water return to [wastewater treatment works] reflecting use of water to wash cars, [and] water gardens”. Third, the wastewater treatment works were “performing 25% more effectively” than the level assumed in the nitrogen calculations (see also paragraph 4.29 of Natural England’s technical guidance note). Fourth, there would be natural reductions in nitrogen concentrations, mainly through de-nitrification processes, which were unquantifiable and so were not taken into account in the calculation (see also paragraph 4.42 of the technical guidance note). Fifth, there was not a high degree of correlation between occupancy rates and dwelling sizes. And sixth, an unknown proportion of the nitrogen discharged to the sea would not affect the protected sites.

The council was also entitled to rely, as the planning officer did in paragraph 8.41 of his report, on the use of the 20% precautionary buffer applied at the end of the calculation to strengthen the conclusion that the use of the 2.4 occupancy rate was appropriate in the particular circumstances of this case. Obviously, the application of such a buffer at the end of a calculation would not excuse a general lack of precaution in the figures used in the calculation itself. This could lead to impermissible “double-counting”. But that is not what happened here. Even though, in this respect, the approach adopted by the council seems not to have been what the technical guidance note contemplated, the precautionary buffer was not used here to justify a general lack of precaution in the exercise, but to strengthen the justification for using an occupancy rate of 2.4 in the calculation. As Mr Mould submitted, it was not Wednesbury unreasonable to use it in this way. It was not the sole justification for the council’s conclusion, as a matter of judgment, that an occupancy rate set at this level was consistent with a sufficiently precautionary approach in the appropriate assessment for this proposed development. It was one element in the broader justification for the use of that occupancy rate, to be seen in the context of the assessment methodology as a whole.

It would not be right for the court to intervene in a case of this kind simply because there is a divergence of expert opinion on some of the figures used in the appropriate assessment. Sometimes, perhaps often, there may not be a consensus of expert opinion. If that is so, there is nothing in law to compel the competent authority in making an appropriate assessment, or the court in reviewing the authority’s decision taken in the light of that appropriate assessment, to default to the most conservative or cautious view propounded.

The argument advanced by Mr Jones does not demonstrate that in this case it was inappropriate or unlawful for the council to adopt the occupancy rate of 2.4 persons per dwelling on the ground that it was, in one expert’s view, insufficiently precautionary – or for any other reason. As Mr Elvin submitted, this is a paradigm case of expert witnesses differing on matters of scientific judgment, in which the court would need to be shown some conspicuous error in the competent authority’s own evaluation of the expert advice it received at the time of its decision before that decision could properly be overturned. For the court to upset a decision when it has not been shown that the competent authority’s own exercise of evaluative judgment was so defective as to be Wednesbury unreasonable but where there is disagreement between experts on the correct ingredients of the appropriate assessment, would involve the court stepping beyond its proper supervisory jurisdiction into the realm of the competent authority’s own remit under the habitats legislation.

I think Mr Jones’ criticism of the judge’s reasoning on this issue is mistaken. The judge accepted that “[an] occupancy rate of 3 would be the best available scientific evidence for 4-5 bedroom houses in the Fareham region” (paragraph 83). This, however, was not fatal to his essential analysis. Reading the relevant passage of his judgment fairly as a whole, I do not think it can be said that he fell into error. He was recognising the fact that, taken in isolation, an occupancy rate of 3 would generally be appropriate for four and five-bedroom houses in the area. Nowhere did he suggest, however, that in this case the appropriate assessment as a whole was inconsistent with “best scientific knowledge”. He found that the method used in the appropriate assessment, taken in its entirety and thus including the occupancy rate of 2.4, complied with the precautionary principle. He accepted as lawful the council’s conclusion, as a matter of its own judgment, that in the circumstances here an assessment using that occupancy rate was sufficiently precautionary. And in my view he was right to do so.

Lastly on this ground, I do not think the judge’s view that Natural England’s technical guidance note would have to be reviewed in the light of his judgment is inconsistent with his view that the approach taken to the occupancy rate in this case was legally defensible. In effect, he was pointing out that the technical guidance note, as drafted, could be liable to misinterpretation or misapplication in other cases, and suggesting that Natural England might describe more clearly the general approach it suggested to this part of the calculation.

Ground 2 is also, in my view, unmeritorious – for two reasons. First, I see no objection in principle to the use of average land use figures to calculate the baseline level of nitrate deposition from the site of the proposed development.

And secondly, I cannot agree with the reading of the CJEU’s judgment in Dutch Nitrogen urged on us by Mr Jones.That case concerned the question of whether “programmatic legislation” – where the appropriate assessment for certain types of project was carried out in advance at a general level and those projects would then be exempt from the requirement for individual assessment – was compatible with article 6(3) of the Habitats Directive. This was the issue to which the observations of the court and Advocate General Kokott on the use of average figures were directed. When the Advocate General said (in paragraph 55 of her opinion), that “[it] would not be sufficient merely to show rough averages and to ignore local or temporary peak load values where those peak values are likely adversely to affect the conservation objectives of the site”, she meant, I think, that it was not appropriate to use averages for all projects of a certain type where some projects of that type might exceed those averages and thus damage the integrity of the protected site. She made clear (in paragraph 147 of her opinion) that the difficulty with using an average value for a number of sites was that it might fail to “guarantee that there are no significant effects on any single protected site”. To the same effect, the court said (in paragraph 119 of its judgment) that “[an] average value is not, in principle, capable of ensuring that there are no significant effects on any single protected site”.

Those statements about the use of average values in that context must be viewed with care in a case such as this, which is not concerned with “programmatic legislation” but with the individual assessment of the particular effects of a specific project. Nothing said in Dutch Nitrogen implies that in this situation the use of averages is inherently objectionable. It is true that the use of average figures will necessarily involve the exercise of judgment on their validity in the particular context. But this does not mean that using them is, in principle, contrary to the requirement for the necessary degree of certainty, as amplified in Waddenzee. The use of average figures may sometimes be conducive to sufficient certainty, sometimes not. Whether that is so in a particular case will be a matter of judgment for the competent authority.

Nothing suggests that in this case either Natural England or the council misunderstood the degree of certainty required by the precautionary principle. Nor is there any evidence to show some justiciable error in the conclusion reached, as a matter of judgment, that the use of average land use figures was, in this case, suitable for the appropriate assessment, and sufficiently robust. Dr O’Neill’s evidence does not demonstrate that there was such an error. Indeed, he recognised (in paragraph 36 of his witness statement) that the selection of the correct land use for the site was “a matter of judgement”, on which he “did not feel qualified to make an assessment”.

I do not accept that the judge held that the use of average land use figures was impermissible but failed to carry that conclusion through to a finding of legal error. What he did (in paragraphs 75 to 77 of his judgment) was to point out that the relevant advice in Natural England’s technical guidance note might be misconstrued in some other case in which the circumstances were different. One should not infer from what he said that in his view the use of average figures would always be impermissible, or that this was so in the circumstances here.

Coming finally to ground 4, I do not think there can be any serious dispute that, in a particular case, the use of a 20% precautionary buffer can ensure that the appropriate assessment meets the required standard of scientific certainty. As the judge said (in paragraph 111 of his judgment), the 20% figure is “not derived from any arithmetical calculation or other algorithm”. There is, however, no legal requirement that every element of an appropriate assessment be based on arithmetic or algorithm. That would be a fallacy. If a precautionary buffer is employed, it should be set at a reasonable level, to help achieve adequate certainty that the high threshold in regulation 63 is crossed. But as Mr Mould submitted, to think that reasonable scientific judgments in undertaking an appropriate assessment can only be reached through arithmetical calculation would be to take too narrow a view of rational enquiry. Such judgments can be formed, and sometimes will best be formed, without resort to arithmetic. This will not, in principle, expose the appropriate assessment to the charge that it suffers from “lacunae” or that it lacks “complete, precise and definitive findings”, as required by the CJEU (see the CJEU’s observations in People Over Wind, at paragraph 38).

The fact that the 20% precautionary buffer was not the product of arithmetic, but of judgment, does not mean that it lacked an adequate basis. As Ms Potts made clear, the appropriate figure to adopt as a buffer was considered carefully by Natural England, knowing the nature of the risks and uncertainties involved (second witness statement, paragraphs 56 to 64, and third witness statement, paragraphs 19 to 21).

Once again, the essence of the complaint is that there is an expert witness – Dr O’Neill – who, in his evidence to the court, has disagreed with a particular figure used in the calculation. That disagreement does not automatically equate to evidence of serious scientific doubt about an appropriate figure for a precautionary buffer. No doubt Dr O’Neill’s evidence shows that, for the reasons he gave, some experts might have adopted a more generous buffer than 20%. This does not mean, however, that the court is bound to find that the buffer actually chosen by Natural England and applied by the council as competent authority was insufficiently precautionary. As Ms Potts’ evidence effectively confirmed, the choice of 20% as the appropriate figure represented the expert regulatory body’s judgment on the level of precautionary buffer consistent with “no reasonable doubt” that the integrity of the protected site would not be adversely affected. It was made with that level of certainty explicitly in mind (third witness statement, paragraph 21). Neither the selection of that figure in Natural England’s technical guidance note nor its use in the appropriate assessment undertaken by the council in this case is open to attack on any legal grounds.

Section 38(6) of the 2004 Act provides:

“(6)

If regard is to be had to the development plan for the purpose of any determination to be made under the Planning Acts, the determination must be made in accordance with the plan unless material considerations indicate otherwise.”

This provision and its predecessor, section 54A of the Town and Country Planning Act 1990, are the subject of ample authority, including several decisions at the highest level and in this court. The relevant principles do not need to be set out at length yet again. They have been stated and restated many times (see, for example, the leading judgment in this court in Secretary of State for Communities and Local Government v BDW Trading Ltd. (trading as David Wilson Homes (Central, Mercia and West Midlands) [2017] PTSR 1357, at paragraphs 19 to 23). A decision-maker must always heed the statutory priority given to the development plan, but is free to assess what weight to give to its policies and to all other material considerations in deciding whether the decision should be made, as the statute presumes, in accordance with the plan (see the speech of Lord Clyde in City of Edinburgh v Secretary of State for Scotland [1997] 1 W.L.R. 1447, at pp.1458 and 1459). If the decision-maker fails to have regard to a relevant policy in the plan or to interpret it properly, conscious that relevant policies in the plan may pull in different directions, the court can act (Lord Clyde’s speech in City of Edinburgh, at p.1459D-F, and the judgments of Lord Reed and Lord Hope of Craighead in Tesco Stores Ltd. v Dundee City Council [2012] PTSR 983, at paragraphs 19 and 34 respectively). But there is no prescribed method for discharging the section 38(6) duty, such as a two-stage approach. This is left to the decision-maker’s good sense in the particular circumstances of the case in hand (Lord Clyde’s speech in City of Edinburgh, at pp.1459 and 1460).

In R. (on the application of Hampton Bishop Parish Council) v Herefordshire Council [2014] EWCA Civ 878, Lord Justice Richards said (in paragraph 28) that “[it] is up to the decision-maker how precisely to go about the task, but if he is to act within his powers and in particular to comply with the statutory duty to make the determination in accordance with the development plan unless material considerations indicate otherwise, he must as a general rule decide at some stage in the exercise whether the proposed development does or does not accord with the development plan”. As Mrs Justice Patterson emphasised in Tiviot Way Investments Ltd. v Secretary of State for Communities and Local Government [2015] EWHC 2489 (Admin) (at paragraphs 27 to 36), with the later endorsement of this court in BDW Trading Ltd. (at paragraph 21), the decision-maker must ascertain whether there is compliance or conflict with the development plan “as a whole”.

It is axiomatic that the interpretation of a development plan policy is ultimately a matter for the court, but that the application of policy is for the decision-maker, subject to the court’s review on public law grounds. The court will intervene on a misconstruction of policy by the decision-maker if satisfied that this has had a material bearing on the decision. But it will only upset a local planning authority’s decision based on an officer’s exercise of planning judgment in assessing compliance with policy if it is convinced that a public law error has been committed (see the judgment of Lord Reed in Tesco v Dundee City Council, at paragraph 19).

At the time of the decision to grant planning permission, the development plan for the borough of Fareham comprised the adopted Fareham Borough Core Strategy and the adopted Fareham Local Plan Part 2: Development Sites and Policies Plan. In his report to the committee the officer identified a number of relevant policies, including Policy CS2 (“Housing Provision”), Policy CS6 (“The Development Strategy”) and Policy CS14 (“Development Outside Settlements”) of the core strategy, and Policy DSP6 (“New Residential Development Outside of the Defined Urban Settlement Boundaries”) and Policy DSP40 (“Housing Allocations”) of the local plan (paragraph 4.1 of the officer’s report).

Policy CS14 of the core strategy says that “[built] development on land outside the defined settlements will be strictly controlled to protect the countryside and coastline from development which would adversely affect its landscape character, appearance and function”, and that “[acceptable] forms of development will include that essential for agriculture, forestry, horticulture and required infrastructure …”.

The first part of Policy DSP40 of the local plan refers to the sites allocated for residential development, sites with planning permission for residential development, and sites safeguarded from other forms of development. The second part of the policy deals with the situation where the requisite five-year supply of land for housing is lacking. It states:

“Where it can be demonstrated that the Council does not have a five year supply of land for housing against the requirements of the Core Strategy (excluding Welbourne) additional housing sites, outside the urban area boundary, may be permitted where they meet all of the following criteria:

i.

The proposal is relative in scale to the demonstrated 5 year housing land supply shortfall;

ii.

The proposal is sustainably located adjacent to, and well related to, the existing urban settlement boundaries, and can be well integrated with the neighbouring settlement;

iii.

The proposal is sensitively designed to reflect the character of the neighbouring settlement and to minimise any adverse impact on the Countryside and, if relevant, the Strategic Gaps;

iv.

It can be demonstrated that the proposal is deliverable in the short term; and

v.

The proposal would not have any unacceptable environmental, amenity or traffic implications.”

When considering the implications of the five-year housing land supply, the officer advised the members that section 38(6) of the 2004 Act was the “starting point for the determination of this planning application” (paragraph 8.8 of the report), and that “there is a presumption in favour of policies of the extant Development Plan, unless material considerations indicate otherwise”. He also reminded them that “[material] considerations include the planning policies set out in the [National Planning Policy Framework (“NPPF”)]” (paragraph 8.9).

He then referred to several policies of the NPPF, including the policy for the “presumption in favour of sustainable development” in paragraph 11 (paragraph 8.12), and the policy in paragraph 177, which states that “[the] presumption in favour of sustainable development does not apply where the plan or project is likely to have a significant effect on a habitats site (either alone or in combination with other plans or projects), unless an appropriate assessment has concluded that the plan or project will not adversely affect the integrity of the habitats site” (paragraph 8.14).

On the question of the proposal’s acceptability as residential development in the countryside, the officer concluded that it was in conflict with Policy CS14 and several other policies of the development plan, stating “[the] site is clearly outside of the defined urban settlement boundary and the proposal is therefore contrary to Policies CS2, CS6 and CS14 of the adopted Core Strategy and Policy DSP6 of the adopted Local Plan Part 2: Development Sites and Policies Plan” (in paragraph 8.22).

The officer quoted Policy DSP40 of the local plan in full (in paragraph 8.52) and then dealt with it in a series of paragraphs (paragraphs 8.53 to 8.65), in which he addressed each of the five criteria in the second part of the policy. On the second criterion, he said this (in paragraph 8.55):

“8.55

The site is considered to be sustainably located within a reasonable distance of local schools, services and facilities at nearby local centres (Warsash and Locks Heath). This part of the northern arm of Brook Avenue is located outside of the urban area, the existing urban settlement boundary being approximately 140 metres east of the site. The proposal is not therefore adjacent to the urban settlement boundary.”

He found compliance with each of the other four criteria (paragraphs 8.54 and 8.56 to 8.65.).

When he came to the “planning balance”, the officer said (in paragraph 8.78):

“8.78

Section 38(6) of [the 2004 Act] sets out the starting point for the determination of planning applications …”.

He then quoted section 38(6), and continued (in paragraphs 8.79 to 8.83):

“8.79

This application has previously been the subject of a favourable Committee resolution to grant planning permission. The revised application proposes additional measures to address the matter of nutrient neutrality but is otherwise the same.

8.80

The site is outside of the defined urban settlement boundary and the proposal does not relate to agriculture, forestry, horticulture and required infrastructure. The principle of the proposed development of the site would be contrary to Policies CS2, CS6 and CS14 of the Core Strategy and Policy DSP6 of Local Plan Part 2: Development Sites and Policies Plan.

8.81

Officers have carefully assessed the proposals against Policy DSP40: Housing Allocations which is engaged as this Council cannot demonstrate a 5YHLS. In weighing up the material considerations and conflicts between policies; the development of a greenfield site weighted against Policy DSP40, Officers have concluded that the proposal is relative in scale to the demonstrated 5YHLS shortfall (DSP40(i)), can be delivered in the short-term (DSP40(iv)), and would not have any unacceptable environmental, traffic or amenity implications (DSP40(v)). Whilst there would be harm to the character and appearance of the countryside the unsightly derelict buildings currently on the site would be demolished. Furthermore, it has been shown that the site could accommodate eight houses set back from the Brook Avenue frontage and an area of green space to sensitively reflect nearby existing development and reduce the visual impact thereby satisfying DSP40(iii). Officers have however found there to be some conflict with the second test at Policy DSP40(ii) since the site is acknowledged to be in a sustainable location but is not adjacent to the existing urban area.

8.82

In balancing the objectives of adopted policy which seeks to restrict development within the countryside alongside the shortage in housing supply, Officers acknowledge that the proposal could deliver 8 dwellings, as well as an off-site contribution towards affordable housing provision, in the short term. The contribution the proposed scheme would make towards boosting the Borough’s housing supply would be modest but is still a material consideration in the light of this Council’s current 5YHLS.

8.83

There is a clear conflict with development plan policy CS14 as this is development in the countryside. Ordinarily, officers would have found this to be the principal policy such that a scheme in the countryside should be refused. However, in light of the Council’s lack of a 5YHLS, development plan policy DSP40 is engaged and officers have considered the scheme against the criteria therein. The scheme is considered to satisfy four of the five criteria and in the circumstances, officers consider that more weight should be given to this policy than CS14 such that, on balance, when considered against the development plan as a whole, the scheme should be approved.”

The officer went on to consider relevant policies in the NPPF (in paragraphs 8.84 to 8.87). He referred to the fact that an appropriate assessment had been undertaken and had “concluded that the development would not have an adverse effect on the integrity of the sites”; noted that in these circumstances paragraph 177 of the NPPF said “the presumption in favour of sustainable development imposed by paragraph 11 … is applied” (paragraph 8.84); confirmed that officers had “therefore assessed the proposals against the ‘tilted balance’ test set out at paragraph 11 of the NPPF” (paragraph 8.85), and considered there to be “no policies within the [NPPF] that protect areas or assets of particular importance which provide a clear reason for refusing the development proposed” and that “any adverse impacts of granting permission would not significantly and demonstrably outweigh the benefits, when assessed against the policies in the [NPPF] taken as a whole” (paragraph 8.86). He recommended that planning permission be granted, subject to a section 106 obligation and suitable conditions (paragraph 8.87).

Jay J. found it clear that the officer had advised the committee that the proposed development did not accord with the development plan in a number of respects. However, paragraph 8.83 of the report gave rise, in his view, to “a degree of interpretative challenge”, and “its various strands are difficult to identify and disentangle” (paragraph 159). That paragraph, he thought, was “somewhat elliptical” and “a degree of benevolence” was required. The issue was “how much?” (paragraph 160).

On that question the judge said that in paragraph 8.83 the officer “was dealing with the first stage of the s.38(6) analysis”, and “considering the extent of compliance with the development plan and the ordering of policies within that plan”. The officer had “found, as he was entitled to, that policy DSP40 was more important in this case than CS14, owing to the shortfall in housing supply, and that the failure to satisfy the second criterion did not undermine this conclusion”. In the judge’s view, “[the] final clause in para 8.83 could be better worded, but it sets out the planning officer’s conclusion on the first stage”. It was “not a conclusion on the s.38(6) issue tout court, still less the planning application as a whole” (paragraph 160).

Having concluded that paragraphs 8.84 to 8.87 of the officer’s report dealt with “the second stage of the s.38(6) exercise”, the judge described paragraph 8.86 of the report as a “composite conclusion on all remaining material considerations in the light of the tilted balance [in NPPF policy]”. The officer’s “overall conclusion” in paragraph 8.87 was, he said, was “legally unexceptionable” (paragraph 161).

Mr Jones submitted, as he did before the judge, that the council had failed to comply with section 38(6). The officer’s advice in paragraphs 8.78 to 8.87 of his report did not contain a conclusive view on the question of whether the proposed development was in accordance with the development plan as a whole – an essential part of the decision-making process, as Patterson J. had said in Tiviot Way Investments (paragraph 27). There was at least “substantial doubt” over the council’s performance of its duty (see the judgment of Lord Justice Elias in Secretary of State for Communities and Local Government v Calderdale Metropolitan Borough Council [2011] J.P.L. 412, at paragraph 46). The officer had not dealt properly with the “nature and extent” of the proposal’s conflict with the plan, and the significance of that conflict (see the judgment of Lord Reed in Tesco v Dundee City Council, at paragraph 22). In the final sentence of paragraph 8.83 of the report, it was not clear whether he was saying that the proposal accorded with the plan as a whole, or that, despite not being in accordance with the plan, it should be approved because “other material considerations [indicated] otherwise”. Having recognised the ambiguity in the officer’s assessment, the judge should have found there was “substantial doubt” sufficient to justify his quashing the planning permission. He went beyond the “benevolence” appropriate in the reading of a planning officer’s report.

Mr Mould supported the judge’s analysis. The court, he submitted, should not read the officer’s report with undue rigour, but with “reasonable benevolence” and bearing in mind it was written for councillors with local knowledge (see R. (on the application of Mansell) v Tonbridge and Malling Borough Council [2019] PTSR 1452; [2017] EWCA Civ 1314, at paragraph 42). Reading the report fairly, it could not conclude that the members had been materially misled. The officer understood the priority to be given to the development plan. He was clearly satisfied that, on balance, the proposal was in accordance with the plan. Because of the shortfall in the housing land supply, he gave more weight to Policy DSP40 of the local plan than to Policy CS14 of the core strategy. In paragraph 8.83 of the report he concluded, in effect, that the limited conflict with Policy DSP40, a partial conflict with only one of its five criteria, when added to the conflict with other plan policies, did not prevent him from finding the proposal in accordance with the plan “as a whole”. This was a reasonable and lawful exercise of planning judgment. The following four paragraphs of the report, paragraphs 8.84 to 8.87, were devoted to “other material considerations” arising from the NPPF, which, again as a matter of planning judgment, the officer found not to indicate the refusal of planning permission. Both limbs of section 38(6) were properly dealt with. And the officer’s ultimate conclusion on the “planning balance”, in paragraph 8.87, was not irrational or otherwise unlawful.

This is an issue to be dealt with in the spirit of realism and common sense to which this court has often referred (see, for example, what was said in Mansell, at paragraph 42; and in St Modwen Developments Ltd. v Secretary of State for Communities and Local Government and others [2017] EWCA Civ 1643, at paragraph 7).

Like the judge, I am not persuaded by Mr Jones’ argument here. On a fair reading of the officer’s report, in particular the passages which embody the performance of the decision-maker’s duty under section 38(6), I would accept that the assessment may, in part, be infelicitously expressed, but not that it is, in substance, unlawful. This is not to ignore the well-known principles governing the approach to planning officers’ reports to committee stated by this court in Mansell, but only to apply those principles sensibly in the circumstances here. When that is done, I do not think one can conclude that there was any “material defect” in the officer’s advice justifying interference by the court (see Mansell, at paragraph 42(3)).

Unlike several cases which have recently found their way to the Court of Appeal or above (see, for example, Braintree District Council v Secretary of State for Communities and Local Government [2018] EWCA Civ 610), there is no issue of policy interpretation for the court to resolve here. The meaning and effect of the relevant policies of the development plan are uncontentious.

In any event, I do not think it can be said that this is one of those cases in which the officer, or the members, misinterpreted any of the relevant policies (see R. (on the application of Corbett) v The Cornwall Council [2020] EWCA Civ 508, at paragraphs 65 to 67). The officer recognised that the proposal was in conflict with Policy CS14 of the core strategy because it would be development in the countryside which did not fall into any of the acceptable forms of development identified in that policy. Indeed, he accepted that there was a “clear conflict” with that policy, “as this is development in the countryside”, and that this conflict would “ordinarily” have led to the refusal of planning permission (paragraph 8.83 of his report).

I agree with Mr Jones that we can put to one side the general quality of the officer’s report, and the obvious care he took in other parts of his planning assessment. As Mr Jones submitted, the judge’s observations praising the officer for the way in which he dealt with other matters could not override a finding that he went wrong in handling the requirements of section 38(6). But I also accept Mr Mould’s submission that those observations of the judge played no part in his conclusions on those parts of the officer’s report where the officer applied the policies of the development plan and took other material considerations into account.

It cannot be suggested that either the officer or the committee was unaware of section 38(6) and the need to perform the duty it states. The officer quoted that provision at the beginning of his consideration of “the planning balance”, in paragraph 8.78 of his report. He obviously had it in mind as he went about that assessment. So this is not a case where it is unclear whether the decision-maker had in mind the words of the statute and proceeded in the light of them. Here, the officer plainly did that. The question is whether he did so lawfully.

As Mr Mould submitted, the structure of the officer’s section 38(6) assessment, in paragraphs 8.78 to 8.87 of his report, is divided into two parts. In the first, comprising paragraphs 8.78 to 8.83, the officer addressed the first limb of the duty – to ascertain whether the proposal was or was not “in accordance with the development plan”. In the second part, which comprises paragraphs 8.84 to 8.87, he turned to “other material considerations”, in particular the policy for the “presumption in favour of sustainable development” in paragraph 11 of the NPPF. He did not have to split the assessment in this way, there being no statutory requirement to do so. But he was entitled to do it, and was thus able to divide his conclusions on the two limbs more distinctly than if he had combined them in a single sentence or paragraph.

I consider, as the judge did, that the officer reached a clear conclusion on the compliance of the proposal with the development plan as a whole. That conclusion appears in the final sentence of paragraph 8.83 of the report. It is true that the officer did not express it in the language used in the first limb of the section 38(6) duty. He did not say, explicitly, that the proposal was “in accordance with the development plan”. He said that “on balance, when considered against the development plan as a whole, the scheme should be approved”. This corresponds to the first limb of the statutory duty. In the context of the officer’s consideration of the four policies of the development plan to which he referred in paragraph 8.80 and his consideration of Policy DSP40 of the local plan and Policy CS14 of the core strategy in paragraphs 8.81 to 8.83, it was, in my view, a sufficiently clear conclusion that the proposal was in accordance with the plan as a whole. To hold otherwise would be to rob the officer’s conclusion of its real meaning, and to undo the committee’s acceptance of it in resolving as it did.

Was the officer lawfully entitled to reach the conclusion that the proposed development accorded with the development plan as a whole? In my view he was. So long as he did not lapse into a misunderstanding of any relevant policy of the plan – which he did not – the accordance of the proposal with the plan as a whole was a matter of planning judgment for him.

What the planning officer did here, as one sees in paragraph 8.22 of the report, was to acknowledge that the application site was “clearly outside … the defined urban settlement boundary”, so that the proposal was “contrary to” several policies of the core strategy and also Policy DSP6 of the local plan. However, because of the absence of a five-year supply of housing land under the requirements of the core strategy it was Policy DSP40 of the local plan on which the officer focused, as the policy of central relevance to the proposal. There can be no complaint about that. This was a classic case of two policies of the development plan pulling in different directions: Policy CS14 of the core strategy pointing to a refusal of planning permission for housing development in the countryside, and Policy DSP40 creating, in its second part, a different and permissive approach to such proposals in the absence of a five-year supply of housing land, subject to the criteria set out. In those circumstances the two policies would obviously be in tension with each other. A proposal satisfying the criteria in the second part of Policy DSP40 would accord with the policy formulated specifically for the situation which arose here, but would likely be in conflict with Policy CS14. In that situation, the decision-maker would have to consider which of these two policies should prevail, the general policy for development in the countryside or the policy deliberately crafted for housing development in the countryside where there is not a five-year supply of housing land. In this case the officer effectively gave precedence to Policy DSP40, as he was clearly entitled to do.

The part of Policy DSP40 which fell to be applied here, because of the absence of a 5-year supply of housing land, sets out what is, in effect, a self-contained policy approach to the determination of applications for planning permission for housing development in those circumstances. The five criteria in the policy encapsulate considerations to which the council will need to have regard when determining such an application.

The officer quoted the relevant part of Policy DSP40 in paragraph 8.52 of his report, and then went through the five criteria, one by one, in paragraphs 8.53 to 8.65. He did not suggest that any of those criteria could be left out of account. He found that four of them – the first, third, fourth, and fifth – were fully complied with. There is no criticism of his consideration of those four criteria. The other criterion – the second – he dealt with in paragraph 8.55, reaching the significant conclusion that the site was “sustainably located within a reasonable distance of local schools, services and facilities at nearby local centres …”. But because the urban settlement boundary was “approximately 140 metres east of the site”, the development would “not … [be] adjacent to [that] boundary”. Thus the proposal complied partially with the second criterion, though not totally. It was non-compliant only to the extent that the site was 140 metres from the urban settlement boundary, not “adjacent” to it. There is, however, no definition of the concept of adjacency in the policy. This is left to the decision-maker’s planning judgment on the facts of the particular case. In summary, therefore, the proposal was fully compliant with four of the five criteria in the policy and substantially compliant with the other.

Those conclusions were picked up later in the officer’s report, and distilled in paragraph 8.81, where he concluded that there was compliance with the first, third, fourth and fifth criteria of Policy DSP40, but “some conflict” with the second criterion, “since the site is acknowledged to be in a sustainable location but is not adjacent to the existing urban area”. None of that part of the officer’s assessment betrays any misunderstanding of Policy DSP40, nor any unlawful application of it.

The advice in the following paragraph (paragraph 8.82) is also unimpeachable. It refers to the contribution that the proposed development would make towards the provision of housing and affordable housing in the situation to which the second part of Policy DSP40 is directed – the absence of a five-year housing land supply.

In paragraph 8.83 the officer recognised the “clear conflict” with Policy CS14 of the core strategy, because this would be “development in the countryside”. That policy, however, was not “the principal policy” because the lack of a five-year housing land supply meant that Policy DSP40 was engaged, and that the proposal was to be considered under the criteria in that policy. Having stated that position, the officer then returned to his assessment of the proposal’s compliance with Policy DSP40. He concluded that “in the circumstances, … more weight should be given to this policy than CS14 such that, on balance, when considered against the development plan as a whole, the scheme should be approved”.

That clearly was an expression of planning judgment, having regard to the role of Policy DSP40 as the main policy of relevance, and the degree of compliance the officer had found with it. One can readily infer that in his view some provisions of the development plan pulled in opposite directions (see Lord Clyde’s speech in City of Edinburgh at p.1459 D-F, and the judgments of Lord Reed and Lord Hope in Tesco v Dundee City Council respectively at paragraphs 19 and 34, and the judgment of Mr Justice Sullivan, as he then was, in ex parte Milne, at paragraphs 48 to 50). Policy CS14 of the core strategy was in tension with Policy DSP40 of the local plan, but the latter prevailed because there was not a five-year supply of housing land. The proposal substantially complied with the relevant part of Policy DSP40, satisfying all five criteria save for its limited conflict with the second criterion. And that limited conflict with one element of a single criterion in the policy was not, in the officer’s view, enough to prevent a finding of compliance with “the development plan as a whole”. In other words, the degree of conflict with the policy was not, overall, of such significance as to prevent approval of the scheme being in accordance with the plan for the purposes of section 38(6). This conclusion too, was a matter of planning judgment for the officer and is not assailable on any public law grounds.

There was, in my view, no misunderstanding or unlawful misapplication of development plan policy, and the path was open to the officer to reach the conclusion he did in the final sentence of paragraph 8.83 – that, “on balance”, when it was “considered against the development plan as a whole”, the proposal ought to be approved. Though not perhaps expressed with perfect clarity, this was a rational conclusion in the exercise of planning judgment, consistent with the relevant passages of the officer’s report read fairly together, and plain in its meaning in that context. In short, it was lawful.

The officer’s conclusions on other material considerations in paragraphs 8.84 to 8.87 were predicated on his conclusion on the first limb of the section 38(6) duty – that a decision to grant planning permission for the proposed development would be in accordance with the development plan. Those conclusions, whose import was that “material considerations” did not indicate that planning permission should be refused, were clearly stated, and are sufficient, in my view, to comply with the second limb of section 38(6). I agree with the judge’s conclusions to that effect.

In my view, therefore, the officer’s assessment under section 38(6), regarded with realism and common sense, is not flawed by any error of law. The reality here is that in the conscious performance of the section 38(6) duty, he undertook every necessary exercise of planning judgment for that duty to be complied with, and none of those planning judgments are infected by legal error. In substance, the officer’s assessment, accepted by the members, was not materially defective.

For the reasons I have given, I would dismiss the appeal.

I agree that this appeal should be dismissed for the reasons given by the Senior President of Tribunals.

I agree with the judgment of the Senior President of Tribunals on grounds two, four and five, concerned respectively with the use of average land use figures in the calculation of baseline nitrogen deposition, the use of a 20% buffer in the budget calculation, and section 38(6) of the Planning and Compulsory Purchase Act 2004. On those issues I have nothing to add.

On the first ground of appeal, which is concerned with the use of the average national occupancy rate of 2.4 persons per dwelling in calculating a nutrient budget for a development of 4-5 bedroom houses, I agree with what the Senior President has said and with his conclusion that the appeal should be dismissed. However, I think it necessary to spell out that, in my view at any rate, the Council’s appropriate assessment was not in accordance with the procedure set out in the technical guidance issued by Natural England, but was nevertheless lawful because there was a good reason not to follow that procedure. In short, that good reason was that the Council consulted Natural England, making clear that it had used the 2.4 persons occupancy rate, and Natural England had no objection to this. I set out my reasoning in this judgment.

Council Directive 92/43/EC (“the Habitats Directive”) was transposed into domestic law by the Conservation of Habitats and Species Regulations 2017 (“the Habitats Regulations”). In the case of a proposed development which is likely to have a significant effect on a protected site, Regulation 63 imposes three relevant obligations on a planning authority (referred to in the Regulations as a “competent authority”). It is common ground that the development here was likely to have such an effect, and that Regulation 63 is therefore engaged.

Those three obligations are as follows. They are mandatory. First, the planning authority must make an “appropriate assessment” of the implications of the proposed development for that site (para (1)). Second, it must consult the appropriate nature conservation body, in this case Natural England, and have regard to any representations made by that body (para (3)). Third, it must refuse planning permission if the conclusion of the “appropriate assessment” is that the proposed development will adversely affect the integrity of the site in question (para (5): strictly, para (5) says that permission may only be granted if the development will not adversely affect the integrity of the site, but this amounts to the same thing.

This latter obligation is subject to an exception, not applicable here, if the planning authority is satisfied that there are “no alternative solutions” and that there are “imperative reasons of overriding public interest” for the grant of permission (see Regulation 64). But that is the only circumstance in which permission may be granted for a development when an “appropriate assessment” carried out by the planning authority indicates an adverse effect on the site in question. The existence of an exception in these very limited circumstances, but not otherwise, demonstrates the importance which the legislature has attached, as a matter of policy, to the protection of endangered habitats. Unless a proposed development qualifies as necessary for imperative reasons of overriding public interest, with no alternative solution, planning permission must be refused for a development which will adversely affect the integrity of the site. There is no balance to be undertaken, weighing protection of the environment against (for example) the need for housing, however acute that need may be. Unless Regulation 64 applies, the planning authority has no discretion to exercise once it has concluded, by means of an “appropriate assessment”, that the effect of the proposed development will be adverse – and that is equally so even if the adverse effect is only modest.

Accordingly Fareham Borough Council had an obligation in the present case to carry out an appropriate assessment, to consult Natural England and to have regard to any representations which it made. The decision whether to grant permission in the light of that “appropriate assessment” remained that of the Council as the planning authority. But that decision was constrained by the outcome of the “appropriate assessment”. If the assessment was unfavourable, permission had to be refused and the grant of permission would necessarily be unlawful. If the assessment was favourable, the Council would have to make a planning judgment in the usual way, with which the court would only interfere on Wednesbury grounds.

Thus in a case where Regulation 63 (but not Regulation 64) applies, the task for the planning authority is not merely to undertake an overall evaluation of all the circumstances, giving such weight to each as it thinks fit. Rather, a favourable “appropriate assessment” is a necessary gateway through which an application must pass before the grant of permission can be considered.

Accordingly the nature of the “appropriate assessment” which a planning authority is obliged to carry out and the degree of rigour which it must bring to bear may be of critical importance. A more rigorous assessment may show an adverse effect which a less rigorous assessment would not. The question therefore arises, who decides what should be done by way of “appropriate assessment” and how it should be carried out? The answer is that, in general, it is left to the planning authority to decide for itself what steps should be taken to investigate the impact of the proposed development on the protected site. This was explained by Lord Carnwath, giving the judgment of the Supreme Court, in R (Champion) v North Norfolk District Council [2015] UKSC 52, [2015] 1 WLR 3710:

“41.

The process envisaged by article 6(3) should not be over-complicated. As Richards LJ points out, in cases where it is not obvious, the competent authority will consider whether the ‘trigger’ for appropriate assessment is met (and see paras 41-43 of Waddenzee). But this informal threshold decision is not to be confused with a formal ‘screening opinion’ in the EIA sense. The operative words are those of the Habitats Directive itself. All that is required is that, in a case where the authority has found there to be a risk of significant adverse effects to a protected site, there should be an ‘appropriate assessment’. ‘Appropriate’ is not a technical term. It indicates no more than that the assessment should be appropriate to the task in hand: that task being to satisfy the responsible authority that the project ‘will not adversely affect the integrity of the site taking account of the matters set out in the article. As the court itself indicated in Waddenzee the context implies a high standard of investigation. However, as Advocate General Kokott said in Waddenzee [2005] All ER (EC) 353, para 107:

‘the necessary certainty cannot be construed as meaning absolute certainty since that is almost impossible to attain. Instead, it is clear from the second sentence of article 6(3) of the Habitats Directive that the competent authorities must take a decision having assessed all the relevant information which is set out in particular in the appropriate assessment. The conclusion of this assessment is, of necessity, subjective in nature. Therefore, the competent authorities can, from their point of view, be certain that there will be no adverse effects even though, from an objective point of view, there is no absolute certainty.’

In short, no special procedure is prescribed, and, while a high standard of investigation is demanded, the issue ultimately rests on the judgment of the authority.”

Accordingly, and in general, so long as the planning authority makes rational choices as to the steps which it will take to investigate the impact of the proposed development, the court will not interfere. Those rational choices must include application of the precautionary principle, which is implicit in the Regulations, and must involve a high standard of investigation, but precisely what that means in practice in any given case is left to the judgment of the planning authority, subject only to review by the court on Wednesbury grounds.

In the present context, however, Natural England as the appropriate nature conservation body has published specific guidance to planning authorities as to the nature of the “appropriate assessment” which they should carry out, which is precisely applicable to the proposed development in this case. The relevant Advice was its “Advice on Achieving Nutrient Neutrality for New Development in the Solent Region (Version 5 – June 2020)” (“the 2020 Advice”).

The 2020 Advice sets out “a practical methodology to calculating how nutrient neutrality can be achieved”, which is said to be “based on best available scientific knowledge”. The methodology consists of calculating a “nutrient budget”, by which the amount of nutrient deposition on protected sites resulting from a proposed development can be estimated. It is, however, important that the 2020 Advice states repeatedly that it is “one way” (or “one means”) of addressing this question (see paras 1.3, 2.2 and 4.1). It does not purport to prescribe a calculation which planning authorities in the Solent region must perform in all circumstances in order to carry out a lawful “appropriate assessment”.

The 2020 Advice begins by explaining the importance for wildlife of the water environment within the Solent region and the existing (and in some cases increasing) deterioration of protected sites. The methodology which it sets out does not seek to reverse the deterioration. Rather, it has the more limited ambition that new developments should not make things worse. In that context it emphasises repeatedly that planning authorities should take a precautionary approach when addressing uncertainty and calculating nutrient budgets. For example:

“1.4

… It is our advice to local planning authorities to take a precautionary approach in line with existing legislation and case-law when addressing uncertainty and calculating nutrient budgets.”

The 2020 Advice goes on to explain that this precautionary approach must be adopted separately at two stages, first when determining each of the “key inputs and assumptions” underpinning a nutrient budget, one of which is the prediction of occupancy levels for a new development, and then again when adding a precautionary buffer to the Total Nitrogen (“TN”) which has been calculated:

“4.7

The nutrient neutrality calculation includes key inputs and assumptions that are based on the best-available scientific evidence and research. It has been developed as a pragmatic tool. However, for each input there is a degree of uncertainty. For example, there is uncertainty associated with predicting occupancy levels and water use for each household in perpetuity. Also, identifying current land / farm types and the associated nutrient inputs is based on best-available evidence, research and professional judgement and is again subject to a degree of uncertainty.

4.8

It is our advice to local planning authorities to take a precautionary approach in line with existing legislation and case-law when addressing uncertainty and calculating nutrient budgets. This should be achieved by ensuring nutrient budget calculations apply precautionary rates to variables and adding a precautionary buffer to the TN calculated for developments. A precautionary approach to the calculations and solutions helps the local planning authority and applicants to demonstrate the certainty needed for their assessments.”

The 2020 Advice explains at para 4.12 that the proposed methodology “is for all types of development that would result in a net increase in population served by a wastewater system, including new homes, student accommodation, tourism attractions and tourist accommodation”.

The methodology contains a number of stages for developments which will drain to the mains network. The first stage is to calculate the Total Nitrogen (measured in kilograms per annum) derived from the development that would exit the Wastewater Treatment Works after treatment. Within this first stage are three steps, the first of which is to calculate the additional population resulting from the proposed development. This is dealt with at paras 4.18 and 4.19, on which much of the argument focused:

“Stage 1 Step 1 Calculate additional population

4.18

New housing and overnight accommodation can increase the population as well as the housing stock within the catchment. This can cause an increase in nitrogen discharges. To determine the additional population that could arise from the proposed development, it is necessary that sufficiently evidenced occupancy rates are used. Natural England recommends that, as a starting point, local planning authorities should consider using the average national occupancy rate of 2.4, as calculated by the Office for National Statistics (ONS), as this can be consistently applied across all affected areas.

4.19

However competent authorities may choose to adopt bespoke calculation tailored to the area or scheme, rather than using national population or occupancy assumptions, where they are satisfied that there is sufficient evidence to support this approach. Conclusions that inform the use of a bespoke calculation need to be capable of removing all reasonable scientific doubt as to the effect of the proposed development on the international sites concerned, based on complete, precise and definitive findings. The competent authority will need to explain clearly why the approach taken is considered to be appropriate. Calculations for occupancy rates will need to be consistent with others used in relation to the scheme (e.g. for calculating open space requirements), unless there is a clear justification for them to differ.”

As is apparent from these paragraphs, the occupancy rate of 2.4 persons per dwelling is the average national occupancy rate for all kinds of dwellings, calculated by the Office for National Statistics. It is derived from the 2011 Census.

I would make two observations on what is said in these paragraphs. First, Natural England’s recommendation is that this occupancy rate should be “considered” by competent authorities, not that its use is in any way mandatory. It is described as no more than “a starting point”. Second, the Advice states that competent authorities “may” choose to adopt a different rate, tailored to a particular area or particular scheme, but that where they do so, the occupancy rate adopted must be evidence-based, clearly explained and consistent with other calculations used in relation to the proposed development.

Mr Timothy Mould QC for the Council and Mr David Elvin QC for Natural England emphasised the use of the word “may”, submitting therefore that competent authorities can be under no obligation to use another occupancy figure. Mr Gregory Jones QC for the appellant objectors submitted that the word “may” should be read as “must”. I would not accept either of these submissions. In my judgment the advice to planning authorities is to begin (“a starting point”) by considering whether the average national occupancy rate of 2.4 is appropriate to use for the development in question. As it is a national average rate over all kinds of dwelling, it is likely that it can appropriately be used where a development consists of mixed housing, including both larger and smaller properties. In such cases, the starting point may well also be the finishing point. But it is common sense that a new development consisting exclusively of larger houses is likely to have a higher occupancy rate than the national average. In such a case, there seems to me to be a powerful argument that it is not appropriate to use the 2.4 rate, which a planning authority needs to consider. I would read these paragraphs as encouraging planning authorities to consider whether there is an alternative evidence-based occupancy rate for which a clear justification can be stated. In fact, such an alternative rate would not have been difficult to find in this case: the Office for National Statistics, which is the source of the 2.4 rate, also publishes an average occupancy rate for four-bedroom houses based on the same source (i.e. the 2011 Census), namely 3.14 persons per dwelling.

Once the occupancy rate for the nitrogen budget calculation has been determined, the next step is to determine the estimated water use for the proposed development. The 2020 Advice recommends using a figure, itself described as precautionary, of 110 litres per person per day. There was nothing to indicate any circumstances in which a lesser usage figure should be used, for example that some occupants of the new dwellings might already be residents within the catchment area.

Stages 2 to 4 of the calculation need not be considered in any detail for the purpose of this ground of appeal. Stage 2 is to adjust the nitrogen load to account for existing nitrogen from current land use; Stage 3 is to adjust the nitrogen load to account for land use with the proposed development; and Stage 4 is to calculate the net change in the Total Nitrogen load that would result from the development. It is at this last stage that a precautionary buffer is recommended:

“4.67

It is necessary to recognise that all the figures used in the calculation are based on scientific research, evidence and modelled catchments. These figures are the best available evidence but it is important that a precautionary buffer is used that recognises the uncertainty with these figures and in our view ensures the approach prevents, with reasonable certainty, that there will be no adverse effect on site integrity. Natural England therefore recommends that a 20% precautionary buffer is built into the calculation.”

Thus the 20% precautionary buffer is not a substitute for use of the best available evidence-based figures for the previous stages of the methodology. On the contrary, it is an additional protection which assumes that the best available figures have been used in those previous stages.

Mr Jones submitted that the guidance set out in the 2020 Advice was unlawful, although it is fair to say that his primary attack in this court was that it had not been properly applied. I would reject the submission that the 2020 Advice was itself unlawful. It is a rational methodology recommended by the appropriate nature conservation body.

The question then arises whether a planning authority in the Solent Region must carry out an “appropriate assessment” in accordance with the 2020 Advice – or to put it another way, whether any departure from the methodology set out in the 2020 Advice would render an “appropriate assessment” unlawful, such that a grant of planning permission based on such an assessment would be Wednesbury unreasonable. In my judgment that cannot be the case. The 2020 Advice itself makes clear that it is only one way of carrying out an “appropriate assessment” and that its use is not mandatory. The true position is that a planning authority ought to follow the methodology contained in the 2020 Advice unless it has good reason not to do so. That is for the same reason, explained by Lord Justice Sales in Smyth v Secretary of State for Communities and Local Government [2015] EWCA Civ 174, [2015] PTSR 1417, that a planning authority must place considerable weight on the response of Natural England in response to a consultation under Regulation 63(3):

“85.

Moreover, the authorities confirm that in a context such as this a relevant competent authority is entitled to place considerable weight on the opinion of Natural England, as the expert national agency with responsibility for oversight of nature conservation, and ought to do so (absent good reason why not): Hart, supra, [49]; R (Akester) v DEFRA [2010] Env LR 33, [112]; R (Morge) v Hampshire County Council [2011] UKSC 2, [2011] 1 WLR 268, [45] (Baroness Hale); R (Prideaux) v Buckinghamshire County Council [2013] EWHC 1054 (Admin), [2013] Env LR 32, [116]. The Judge could not be faulted in giving weight to this consideration in the present case, at para. [165] of her judgment.”

One potentially good reason not to follow the methodology in the 2020 Advice precisely would be that Natural England itself has raised no concerns about a proposed development, despite appreciating that the methodology has not been precisely followed.

This brings me to the obligation, contained in Regulation 63(3), to consult Natural England and to have regard to its view. As explained in the passage from Lord Justice Sales’ judgment in Smyth quoted above, the Council was both entitled and required to place considerable weight on the opinion of Natural England, unless there was good reason not to do so.

In this case the Council did consult Natural England. Although it did not draw specific attention to the use of the national average occupancy rate of 2.4 persons per dwelling for a development consisting of 4-5 bedroom houses, it provided information about the proposed development to Natural England from which the nature of the development and the use of the national average occupancy rate were both readily apparent. We can safely proceed on the basis that Natural England understood this. That is apparent from its stance and evidence in this action, opposing the claim for judicial review. Natural England made clear in its response to the consultation that it had no concerns about the proposed development, including the use of the average national occupancy rate, provided that certain conditions were imposed. That was so even though using the occupancy rate of 2.4 resulted in a nitrogen budget calculation which was only just positive, from which it would have been apparent that taking any higher occupancy rate would have meant that the assessment was negative and that permission would necessarily have had to be refused.

The Officers’ Report for the proposed development, dated 19^th August 2020, noted that the application was for eight detached dwellings which were likely to be larger than average. It summarised accurately the content of paras 4.18 and 4.19 of the 2020 Advice, noting that Natural England recommended that, as a starting point, local planning authorities should consider using the average national occupancy rate of 2.4 persons per dwelling, but that they might choose to adopt bespoke calculations where satisfied that there is sufficient evidence to support this approach. Referring to the concern of objectors that a higher occupancy rate ought to be applied since the houses were likely to be larger than average dwellings, the Report concluded as follows:

“8.40

It is acknowledged that some houses will have more than the average number of occupants. It is also of course the case that some will have less. The figure of 2.4 is an average based on a well evidenced source (the ONS) and which has been shown to be consistent over the past 10 years. As stated above the Natural England methodology allows bespoke occupancy rates however to date the Council has only done so to lower, not raise, the occupancy rate and where clear evidence has been provided to demonstrate that the proposed accommodation has an absolute maximum rate of occupancy. In the case of sheltered housing which is owned and managed by the Council for example it has previously been considered appropriate to apply a reduced occupancy rate accordingly.

8.41

In all instances it is the case that the Natural England methodology is already sufficiently precautionary because it assumes that every occupant of every new dwelling (along with the occupants of any existing dwellings made available by house moves) is a new resident of the Borough of Fareham. There is also a precautionary buffer of 20% applied to the total nitrogen load that would result from the development as part of the overall nutrient budget exercise.

8.42

Taking the above matters into account, Officers do not consider there to be any specific justification for applying anything other than the recommended average occupancy rate of 2.4 persons per dwelling when considering the nutrient budget for the development.”

For my part, and without (I hope) reading the Report in an unduly legalistic way, I do not think that these paragraphs represent a correct application of the methodology contained in the 2020 Advice. The Report treats the average national occupancy rate as the rate “recommended” by Natural England, to be applied unless there is a “specific justification” for taking some other rate. But that is not what the 2020 Advice says. What it says is that the 2.4 rate should be considered, but it does not suggest that it is anything more than a starting point.

Moreover, the Report’s justification for using the 2.4 figure was that the Natural England methodology “is already sufficiently precautionary”. The first reason for this view was that the methodology assumes that every occupant of every new development would be a new resident of the borough. On this point the Report is mistaken. There is nothing to that effect in the 2020 Advice. It does not suggest, for example, that in the case of a mixed development where it might be expected that occupancy will be in line with the average national rate, some adjustment should be made to take account of this factor. The second reason was that there was also “a precautionary buffer of 20% applied to the total nitrogen load”. But the existence of that buffer is not a justification for using anything other than the best available evidence-based occupancy rate for the development concerned. Rather, the 20% buffer is intended to be an additional protection, over and above the use of the best available evidence as to the “key inputs and assumptions” underpinning the nutrient budget. It is applied only after the four stages of the methodology have been completed. In my view, therefore, the Report departs from the methodology set out in the 2020 Advice on the question of occupancy rate.

Despite this, however, I consider that the use of the national average occupancy rate of 2.4 persons per dwelling did not render the “appropriate assessment” carried out by the Council unlawful. The question for the Council was not whether it had followed precisely the methodology set out in the 2020 Advice, but rather whether it had carried out a sufficient “appropriate assessment” for the purpose of the Habitats Regulations. It was not mandatory to follow precisely the methodology set out in the 2020 Advice and the use of the national average occupancy rate was not questioned by Natural England when consulted about the proposed development. Rather, Natural England stated that it had no concerns. That was a view to which the Council was entitled and required to have regard. It provided a good reason not to follow precisely the methodology set out in the 2020 Advice. In those circumstances we can only interfere with the conclusion of the Council, based on the assessment which it had undertaken, that the proposed development would not contravene Regulation 63 of the Habitats Regulations, if that conclusion was Wednesbury unreasonable. That is a demanding test and I am not persuaded that it is satisfied here.

I would add that we have been provided with the latest version of Natural England’s Advice to planning authorities, issued in March 2022 and updated expressly in the light of (among other things) the judgment of Mr Justice Jay in this case. This Advice is not limited to the Solent region.

Interestingly, the 2022 Advice emphasises the importance of local conditions in selecting an occupancy figure, and the need to focus on the particular project being assessed. It recognises that the average national occupancy rate of 2.4 persons per dwelling (which it notes will be subject to change when the results of the 2021 Census become available) may not be appropriate for certain types of development:

“Occupancy rates based on dwelling type

Should the nature or scale of development associated with a particular project proposal suggest that the use of an average occupancy rate is not appropriate, then the Local Planning Authority may decide to adopt an occupancy rate based on the dwelling types proposed for that particular project, provided it meets the criteria outlined above …”

Those criteria include that the rate selected reflects local conditions, is sufficiently robust and appropriate for the project being assessed, and is derived from a reliable source which can show trends over a protracted period of time, such as data from the Office for National Statistics.

For the future it is the 2022 Advice which planning authorities will need to consider.