Ian Galloway, R (on the application of) v Durham County Council

This is a judicial review case about planning permission and a proposed solar farm development in DH7. The Developer (the Interested Party) is a special purpose company related to Lightsource bp, as was the case in R (Durham County Council) v Levelling Up Secretary [2023] EWHC 1394 (Admin) [2023] PTSR 1735 (see §2). The location (postcode DH7 0RY) is a group of fields described as 93 hectares (230 acres) of land, approximately 230m north of Burnhope, 800m south east of Maiden Law, 1.5km east of Lanchester and 2.2km to the south of Annfield Plain and Stanley. The solar farm would be connected to the National Grid by underground cable, to provide clean energy for up to 13,861 homes. Mr Galloway (the Claimant) lives in Burnhope. His witness evidence tells me about Burnhope, the nearby landscape and footpaths across it.

The Developer made two applications. First, an application dated 30 June 2022 (reference DM/22/01769/FPA) for full planning permission pursuant to s.70 of the Town and Country Planning Act 1990. The Council (the Defendant) by its County Planning Committee resolved on 7 March 2023 to grant that application. It was subsequently granted by a Decision Notice dated 13 July 2023. The March 2023 resolution was preceded by a March 2023 Officer Report by the Council’s Senior Planning Officer (Chris Shields), whose recommendation was accepted by the Committee. Secondly, an application dated 24 October 2023 (reference DM/23/03147/NMA) for a non-material amendment (“NMA”) pursuant to s.96A of the 1990 Act (§33 below). The Defendant by its Head of Planning and Housing (Michael Kelleher) granted that application on 7 November 2023. That November 2023 NMA decision was preceded by a 7 November 2023 Officer Report by Mr Shields and Claire Teasdale, whose recommendation was accepted by Mr Kelleher. The NMA had four aspects (see §§33, 52, 67-68 below). The virtues of open-administration and online-accessibility in the planning world mean that any reader’s thirst for further detail can be quenched by googling the references.

This case has three features with parallels in other recent cases. (1) The Statutory Capacity Threshold for solar farm developments approvable by a local planning authority (§9 below) featured in Durham. (2) An exercise of interpreting a conformity planning condition (§35 below) featured in R (Swire) v Canterbury City Council [2022] EWHC 390 (Admin). (3) An interpretive premise for an NMA decision (§34 below) featured in R (Milne-Skillman) v Horsham District Council [2023] EWHC 2919 (Admin).

The Developer asked the Council to afford substantial positive weight to the level of renewable energy that would be generated by the solar farm, pointing to the Council’s 2019 declaration of a climate emergency and Climate Emergency Response Plan. Objective 17 of the County Durham Plan (“CDP”), adopted by the Council in 2020, is: “Reduce the causes of climate change and support the transition to a low carbon economy by encouraging and enabling the use of low and zero carbon technologies, supporting the development of appropriate renewable energy sources and sustainable and active transport”. Policy 33 says: “Renewable and low carbon energy development in appropriate locations will be supported. In determining planning applications for such projects significant weight will be given to the achievement of wider social, environmental and economic benefits”.

The Committee decided this was an appropriate location. It decided that it was acceptable to have a solar farm development, at this location, sufficient to provide clean energy for up to 13,861 homes. The planning assessment in the March 2023 Officer Report had included these conclusions (§§196-200):

The acceptability of this as a location is a planning judgment which is unimpeachable. But did the Council give planning permission for a solar farm with a “capacity” above 50MW? Did the Council give planning permission for a ‘footprint’ of blue land covered by “the solar arrays” which was bigger than is needed? Was the size of this ‘footprint’ a detail left over for pre-commencement approval? These are among the questions I need to address.

The Agreed Issues are as follows (the labelling is mine):

The ventilation of Agreed Issue [A1.4c] attracted resistance from the Council and the Developer. They characterised it as inappropriate ‘rolling judicial review’. But this resistance subsided. Here is what happened. Mr Kelleher’s November 2023 NMA Decision came just in time to be relied on in the Council’s, and the Developer’s, Detailed Grounds of Resistance. Those documents were the pleaded responses to Mr Galloway’s August 2023 judicial review challenge to the July 2023 Planning Permission. This reliance was the purpose of the NMA. When Mr Galloway promptly applied to amend his judicial review grounds, to challenge the NMA Decision, the Council and Developer persuaded HHJ Klein that Mr Galloway should instead have issued a second judicial review claim. Mr Galloway did so, while also invoking a ‘liberty to apply’ for which HHJ Klein rightly made provision. The Council and Developer sought to rely on the NMA Decision at the hearing before me, while at the same time opposing the Court considering Mr Galloway’s four-page grounds impugning its legality. I found that bewildering. But these clouds blew over. Mr Barrett and Mr Hardy took their stand on the argument that the various aspects of the November 2023 NMA Decision were premised on, and did no more than expressly articulate, what was already the true meaning of the July 2023 Planning Permission (§§34, 53 below). Mr Barrett and Mr Hardy recognised that, insofar as they were wrong in their arguments about that true meaning, the NMA Decision was legally vitiated. Everyone agreed that I had to decide that true meaning anyway. Everyone was able to deal with everything. That is how it should always have been: see Swire at §128 (unfortunately not cited to HHJ Klein).

The Agreed Issues (§7 above) refer to a capacity above 50MW [AI.1], or under 50MW [AI.2], and to a “lawful” capacity [AI.3c]. Here is why. There is a Statutory Capacity Threshold for solar farm developments approvable by a local planning authority. It is prescribed by Parliament in s.15(2)(c) of the Planning Act 2008. That provision says that an onshore non-wind generating station in England is a “nationally significant infrastructure project” (which I will shorten to “NSI Project”) if it is expected, when constructed, to be a generating station whose:

Where it is an NSI Project, a development involving the construction of a generating station does not require planning permission from the local planning authority (2008 Act s.33(1)(a)), but instead needs a development consent from Central Government (s.31). Parliament listed (in s.14) a number of projects – in the fields of energy, transport, water, waste water or waste (see s.14(6)) – which are NSI Projects if they meet relevant statutory criteria. These include (s.14(1)(a)) “the construction or extension of a generating station”, if it crosses the s.15(2)(c) Statutory Capacity Threshold. To carry out a development constituting an NSI Project, without a development consent, is a criminal offence (s.160(1)). In circumstances where a development consent is required from Central Government, Parliament disapplied the requirement that a development needs planning permission (1990 Act s.57), by providing that planning permission is not required for a development “to the extent that development consent is required” (2008 Act s.33(1)(a); 1990 Act s.57(1A)).

All of this makes it important to be able to say whether or not a development of a generating station is, or is not, an NSI Project. It makes it important to say whether or not the generating station is expected to have a capacity more than 50MW. It makes it important to know how to understand and apply “capacity”. After all, the Statutory Capacity Threshold triggers the disapplication of the duty to have planning permission; it triggers the need for the development consent; it delineates the authorising division of labour between local planning authority and Central Government; and it triggers the criminal offence. There are other similar capacity thresholds. The effect of s.15(2)(aa) and s.15(3A)(b) of the 2008 Act is that an onshore wind farm is not an NSI Project; but the effect of s.15(3) is that an offshore generating station is an NSI Project if “its capacity is more than 100 megawatts”. The effect of s.15(2)(c) is that an onshore generating station in England, which is not a wind farm, is an NSI Project if “its capacity is more than 50 megawatts”. The effect of s.15(3A)(c) is that an onshore generating station in Wales, which is not a wind farm, is an NSI Project if “its capacity is more than 350 megawatts”. The statutory wording frames the question as whether, when constructed (or extended), the generating station “is expected to be” one with a capacity more that 50MW (s.15(1)). Capacity matters.

Since the 50MW Statutory Capacity Threshold is so central to the Agreed Issues in the case, the Court was always going to need to understand it. What does it mean? How is it to be reasonably applied? What light is shed by statutory purpose and function? I raised these questions with Counsel. Mr Harwood KC and Mr Hardy were agreed about one thing: Central Government development consent is perceived as a more arduous route than planning permission from a local planning authority. If your solar farm is above the Statutory Capacity Threshold, you face a tougher approval regime.

No party submitted that there is any statutory definition for the s.15(2)(c) Statutory Capacity Threshold. I simply record the following from the 2008 Act. In the context of underground gas storage, Parliament deployed “working capacity” in terms of standard cubic metres (s.17(4)(a)), defined as a facility’s capacity for storage of underground gas ignoring “cushion gas” (s.17(7)). In that context, Parliament also used “maximum flow rate” (s.17(4)(b)), defined as the “maximum rate at which gas is able to flow out of the facilities” when they are “filled to maximum capacity” (s.17(7)). Similar provision was made in relation to liquefied natural gas facilities (s.18), and storage capacity is “to be measured” as if gas were stored “in regasified form” (s.18(5)). In the context of waste water treatment plants, Parliament spoke of a plant expected to have “a capacity exceeding a population equivalent of 500,000” (s.29(1)), alongside “a capacity for the storage of waste water exceeding 500,000 cubic metres” (s.29(1A)(c)). For hazardous waste facilities, Parliament deployed capacity in terms of disposal or recovery of hazardous waste in “tonnes per year” (s.30(2)). Airports were addressed by Parliament – without using the word “capacity” – in terms of air passenger transport services “for at least 10 million passengers per year” or air cargo transport services “for at least 10,000 air transport movements of cargo aircraft per year” (s.23(3)). But I have seen, and was shown, no definition or steer in terms of the megawatt capacity of an electricity generating station. Reference was made to the origin of the Statutory Capacity Threshold for generating stations as being the Electricity Act 1989 s.36 consent regime. But no illumination was said to be derived from that regime or sources relating to it. In Durham, Chamberlain J described the Hulam and Sheraton Farm solar farm developments as each having a capacity of 49.9.MW (§§1-2). That characterisation was uncontroversial. The case was about whether those two solar farms would constitute a single generating station, or two separate generating stations.

Mr Galloway has said, from the start, that capacity means “maximum instantaneous output” and that you measure it by the Combined-Panels Method (§16 below). The Developer has said, throughout the planning process, that a solar farm of up to 50MW in capacity provides electricity for up to 13,861 homes, as is intended here.

In the Developer’s December 2022 communications with the Council, and again in Mr Hardy’s skeleton argument, the Developer has emphasised that the September 2021 Draft National Policy Statement for Renewable Energy (EN-3) (“Draft EN-3”) included these two sentences about typical acreage (§2.47.2):

Draft EN-3 describes a recognised conventional method – for solar farms – which I will call the “Combined-Panels Method”. You count the solar panels and aggregate their maximum power. Mr Hardy accepts that this is what Mr Galloway has done in the present case. Here is the Combined-Panels Method (from Draft EN-3 §§2.48.6 and 2.48.7) (underlining in quotes is my emphasis added):

The Combined-Panels Method has what I will call an “Accompanying Proviso”. It involves a recognition that a degree of “Overplanting” can be acceptable. Overplanting means installing ‘spare’ solar panels for necessary future use, as a ‘back-up’ so as to address light-induced degradation of solar panels. Here, then, is the Accompanying Proviso to the Combined-Panels Method (from Draft EN-3 §2.48.8 and fn.43):

There is another point to be derived from Draft EN-3. It is found in the same paragraph from which the Developer quoted (§15 above). It describes a “typical solar panel” (as at September 2021), as follows (Draft EN-3 §2.47.2):

Draft EN-3 describes another recognised conventional method – being proposed by Central Government for solar farms in the future – which I will call the “Combined-Inverters Method”. You count and aggregate the capacity of the installed inverters (measured in AC). In this case, the March 2023 Officer Report told the Committee that the proposal involved “24 inverters”. No measurement was given for their combined capacity. One of the Approved Plans (§30 below) was: “UK_EPD_INV Inverter”. Here is the Combined-Inverters Method (returning to Draft EN-3 §§2.48.6 and 2.48.7, with a different emphasis):

The Combined-Inverters Method has its own “Accompanying Proviso”. It is the recognition that inverter capacity is not sufficient to address planning acceptability, and consideration should be given to panel size, total area and percentage of ground cover. Here is the Accompanying Proviso (from Draft EN-3 §2.48.9):

Reference was made in the papers before the Court – and in the papers put to the Council by the Developer in seeking the NMA – to a Grid Connection offer for the development, limited to 49.5MW. It could be that ‘grid connection’ up to 50MW explains the reference to electricity for “up to 13,861 homes”. We can imagine electricity exported to the grid through a connection, much as we can think of water flowing to a network through a tap. We can imagine measuring the “capacity” of a water-tank by reference to the flow-rate of the tap. But what if a size threshold brings a more exacting scheme to authorise ‘bigger’ water-tanks? Perhaps no sensible developer would ever build an over-sized tank, so tap-size works for the size threshold. I asked whether a developer could build a huge generating station, with a planning permission referable to a modest start-up grid connection, and then increase the grid connection capacity. That would be like building a huge tank, with a low-rate tap, and then installing a much higher flow-rate. I can see that installing bigger “inverters” could be an extension and further development. But it is not obvious to me that – as Mr Hardy submitted – increasing the grid upload would be “carrying out … development” to require development consent (s.31) and trigger the criminal offence (s.160(1)). I can leave it there. Nobody in this case has argued that it is the grid connection which is the lawful measurement of “capacity” for the Statutory Capacity Threshold in s.15 of the 2008 Act. I am left with the Combined-Panels Method, which is illustrated on the evidence; and the Combined-Inverters Method, which is not.

However capacity is measured, there are other ways we can think about the size of a solar farm. There is the ‘footprint’ of the development. There is the ‘footprint’ of solar panels. This brings us to acreage and what is meant by an acre “occupied” by solar panels. In this case, the development site is 93 hectares. The land to be “occupied by solar panels” has, since January 2023, been described by the Developer as 32 hectares (79 acres). But what does this mean? There are two ways to think about an acre “occupied” by solar panels. Think of a one-acre field divided into horizontal strips of land of equal width. Suppose these strips are, alternately, a solar array (a row of solar panels) or a grass walkway in between. In one sense, if you add up the widths of all the rows of panels, half of this one-acre field is “occupied” by solar panels. In another sense, the whole one-acre field is “occupied” by solar panels. The whole field is what the solar farm “requires”. If you looked at this field, in the countryside, it would look like a field full of solar panels. On a straightforward reading, it seems that Draft EN-3 §2.47.2 (§15 above) was speaking of this as an acre of solar panels, because it was describing the acreage which the solar farm “requires” for the solar panels.

All of which brings me to blue and green areas. This blue/green colour scheme comes from the Developer’s drawings. We are enabled to think of blue areas (fields covered by blue stripes of solar panels including the walkways in between) and green areas (land within the development site which has no solar panels or other structures). It is this world of blue and green which matters most, in the real world, when you are looking at the landscape. I will now explain this world of blue and green areas by describing three drawings, in sequence (see §§24-26 below).

On 14 April 2022, the Developer had asked the Council’s planning officers for a screening opinion, to confirm that there was no need for an environmental impact assessment to accompany the application for planning permission. An accompanying colour drawing was entitled GBR GBR_Burnhope_LP1 – IDL_05, which I will call “IDL-05” for short. IDL-05 was dated 14 April 2022. It was an initial design layout which showed various fields covered with blue lines. The key explained that these blue lines were lines of blue rectangles representing module tables of solar panels (28x4, 14x4 or 7x4). Looking at IDL-05, the field in the south-east of the development area, closest to the village of Burnhope, was shown with a blue area – blue solar panel coverage – right up to the edge of the field constituting the site boundary.

A different drawing accompanied the Developer’s June 2022 application for planning permission. This was entitled GBR_GBR_Burnhope_LP2 – PDL 04, which I will call “PDL-04” for short. PDL-04 was dated 23 June 2022. It was a preliminary design layout, again showing various fields covered with blue lines. As with IDL-05, the key explained that these were lines of blue rectangles representing module tables of solar panels (28x4, 14x4 or 7x4). Looking at PDL-04, the field in the south-east of the development area, closest to the village of Burnhope, was now shown with a blue area – blue solar panel coverage – which would no longer extend to the edge of the field constituting the site boundary. Instead, there was a portion of land between the blue shading and the edge of the fields. This portion was a green area – shaded in green. The key explained that this was a “Screening/Ecological Enhancement Zone”. There was another such green area to the west of the development. PDL-04 was different from IDL-04 in another way: there was an additional field on the west side of the development shaded blue, to be occupied by rows of solar panels. So, June 2022 brought different blue areas; and a reduced blue area nearest the village.

The design layout drawing which became the Approved Plan was entitled GBR_GBR_Burnhope_LP2 – PDL_08, which I will call “PDL-08” for short. It was dated 9 December 2022. Like PDL-04, it was called “preliminary design layout”. Again, it showed various fields covered with blue lines. As with IDL-05 and PDL-04, the key explained that these were lines of blue rectangles representing module tables of solar panels. Looking at PDL-08, the field in the south-east of the development area, closest to the village of Burnhope, was now shown with a blue area – blue solar panel coverage – which stopped even further away from the edge of the field and site boundary. The green area of land (“Screening/Ecological Enhancement Zone”) was bigger. So, December 2022 brought more green, and less blue, nearest the village. PDL-08 was different from PDL-04 in two other ways. First, the module tables described in the key were different (30x2 and 15x2). Secondly, the blue rows were in a narrower configuration, with more rows for each field (see §80 below).

While I am introducing drawings, there are two more that matter in this case. One is the Panel Elevation plan, listed as PNL_2P_25/6839, which I will call “PNL-6839” for short. It is dated 14 December 2022. It contains technical drawings. They are arrays of solar panels, viewed from above, from the side, from the front, and from the back. Detailed measurements and angles are given. Each array has two rows of panels, one row across the top and the other across the bottom. They are positioned on a 25-degree incline. The bottom edge of the bottom row of panels starts 1000mm off the ground. The top edge of the top row of panels ends 3060mm off the ground. Each unit (panel) measures 2384mm top to bottom and 1303mm left to right. The land-space, in between each array of panels, measured from directly under the top edge of one row and the bottom edge of the adjacent row, is 2482mm.

The other is Panel Elevation plan PNL_4L_25/17, which I will call “PNL-25/17” for short. PNL-25/17 was dated 8 June 2022. It contained technical drawings of arrays of solar panels, viewed from above, from the side, from the front, and from the back. Detailed measurements and angles are given. Each array has four rows of panels, positioned on a 25-degree incline. The bottom edge of the bottom row of panels starts 1023mm off the ground. The top edge of the top row of panels ends 3000mm off the ground. Each unit (panel) measures 1133mm top to bottom and 2256mm left to right. The land-space, in between each array of panels, measured from directly under the top edge of one row and the bottom edge of the adjacent row, is 6347mm. PNL-25/17 was superseded by Approved Panel PNL-6839.

The July 2023 Planning Permission is a formal Approval of Planning Permission document. It has three parts. Part 1 identifies the particulars of the application as: “Installation and operation of a ground mounted photovoltaic (PV) solar energy generation system (solar farm), battery storage facility, electrical substation and associated infrastructure”. The location is then also identified. Part 2 identifies the particulars of the decision with this operative wording:

Here is the correct list of the 26 Approved Plans:

Planning Condition 4 and its accompanying “Reason” were as follows:

Planning Condition 12 and its accompanying Reason were as follows:

I am now able to address the first of the four aspects of the November 2023 NMA. I have explained (§29 above) that an incorrect list of 29 plans inaptly included superseded plans and failed to replicate the list of 26 drawings in the March 2023 Officer Report recommendation, which the Committee had resolved to adopt. Subject to a distinct point about one Approved Plan being replaced (see §67 below), everybody agrees that it was a benign virtue of the November 2023 NMA that the correct list of 26 Approved Plans was inserted, with the inapt superseded plans removed, and with titles and descriptions realigned. This was plainly “non-material” pursuant to s.96A of the Town and Country Planning Act 1990 which provides that:

This brings me to the idea of an interpretive premise for an NMA decision. The making of a lawful NMA involves questions of planning judgment about whether the change is “material” (Swire §123; Milne-Skillman §30). But an NMA decision will be vitiated by public law error if there has been a material error of law, including in understanding the correct meaning of the planning permission (Milne-Skillman §§5, 59). If that interpretive premise is legally incorrect, it can vitiate the NMA decision as it did in Milne-Skillman. As I have foreshadowed (§8 above), here the November 2023 Officer Report identified the purpose of the NMA as being “to provide clarity in advance of the JR hearing”; that the amendments were “for clarity only”, making the intention “explicitly clear”; with revisions which “seek only to provide clarity” and could not therefore affect any third party. All of this is true of this first aspect of the NMA. It benignly corrected the list to state only the Approved Plans actually approved by the Committee. The Planning Permission would, in any event, have been interpreted to recognise where one Approved Plan had plainly superseded another.

I am now going to address a central controversy in this case. It concerns the correct interpretation of the ‘conformity’ condition (Planning Condition 4: §31 above), alongside the ‘details-approval’ condition (Planning Condition 12: §32 above). What do they mean, and what is their interrelationship? This is an exercise in interpreting a conformity planning condition. In Swire, Holgate J undertook a similar exercise. He identified the legally correct interpretation – in that case in the context of an outline planning permission – of a ‘conformity’ condition (that the development “shall be carried out in accordance with the following approved plans, in respect of those matters not reserved for later approval”: Swire §36), and a details-approval condition (requiring a pre-development written Masterplan: §37). Holgate J explained that “in accordance with” in the conformity condition had its ordinary and natural meaning of “in agreement or harmony with” (§43) and the legally correct interpretation of that condition was that matters not reserved for later approval needed to “accord or harmonise with” the approved plans, which did not require “rigid adherence” (§§60-61). That analysis of those conditions was case-specific and fact-specific. I now have to undertake a similar exercise in this case, in the context of Planning Conditions 4 and 12.

Mr Barrett for the Council, supported by Mr Hardy for the Developer, argued in essence – as I saw it – as follows. The starting point is that interpretation should be by reference to the natural and ordinary meaning of the words, taking account of the particular context (statutory or otherwise) and in the light of common sense: Lambeth LBC v Housing Secretary [2019] UKSC 33 [2019] 1 WLR 4317 at §19; DB Symmetry Ltd v Swindon BC [2022] UKSC 33 [2023] 1 WLR 198 at §66. It is essential to interpret the Planning Permission as a whole and to read Planning Conditions 4 and 12 together.

Planning Condition 12 plainly provides for an important pre-commencement stage. It involves the “precise details” in relation to “any” above-ground structure to be addressed by submission and approval, under the procedure in Article 27 of the 2015 Order. That will “include” colours and finishes. But it covers all “details” of “any” structure constituting an “above-ground structure”. It is linked to the interests of visual amenity in accordance with the relevant plan policy framework. It is linked to an assessment of the appearance of the development. Condition 12 would be undermined if its function were restricted, to deal with a few narrow remaining matters, with everything else having been fixed by way of rigid adherence to the 26 Approved Plans including PDL-08 and PNL-6839. That cannot have been the objectively-identified purpose. Condition 12 refers to “the approved documents” but this is simply a reference to “the precise details … submitted to and approved in writing” within Condition 12 itself. There is no need for any Condition 12 stage so as to follow-through on compliance with Condition 4, which stands on its own two feet and is enforceable on its own terms.

Planning Condition 4, correctly interpreted, is not requiring rigid adherence with Approved Plans such as PDL-08 and PNL-6839. This is supported by the particular context; and by common sense. The nature of the Planning Permission is that commencement of the development would take place in the future. The Committee’s March 2023 resolution required a biodiversity management agreement (entered into on 12 July 2023). The Planning Permission was designed to authorise a development, permitted (by Condition 1) to be commenced within three years. Condition 12 was a Details-Approval stage, to take place much closer to the actual commencement. This is a solar farm. Technology changes. Take this obvious example. Suppose that – by the time the Developer is ready to commence with the development – there is a commercially available solar panel which is smaller, more powerful, more efficient and more cost-effective than the layout and dimensions depicted in Approved Plan PNL-6839. That is precisely the sort of change of circumstances which is recognisably foreseeable in the context of a solar farm. It would undermine common sense and be contrary to the public interest, and for that matter environmental interests, if the Developer were by virtue of Planning Condition 4 straitjacketed from being able to deploy the newly available technology. It is true that Planning Condition 4 speaks of “strict accordance” with the Approved Plans. However, as a matter of interpretation of Planning Condition 4, there are two features of the Planning Permission read as a whole, whose effect is that “strict accordance” – in context and as a matter of common sense – means Holgate J’s Swire broad “agreement or harmony” rather than “rigid adherence”.

The first feature of the Planning Permission, which has this effect, is the wording on the face of the Approved Plans themselves. The phrase “strict accordance” needs to be interpreted in the light of the very plans which are “the approved plans”. Approved Plans PDL-08 and PNL-6839 are documents which, on their face – including when printed – are expressly non-rigid in nature. PDL-08 is headed “Preliminary Design Layout”. That is what “PDL” stands for. As for the panels (“PNL”), PNL-6839 expressly depicts a “Typical Section”, “Typical Elevation” and “Typical Rear Elevation”. The words “preliminary” and “typical” must have a meaning. They and the drawings have a purpose. The meaning, and the purpose, is non-rigidity. The inherent nature of both of these Approved Plans is they describe aspects of the development whose details would need to be confirmed and may need to be revised prior to commencement. It follows that, for any Approved Plan which uses language such as “preliminary” or “typical”, the accordance in Planning Condition 4 is subject always to the flexibility in having all details capable of being revised under the approved-details mechanism in Planning Condition 12. That is the first feature.

The second feature of the Planning Permission, which has this effect, is the Design and Access Statement. This Statement was one of the supporting documents which accompanied the planning application form, as was statutorily required for this as a “major development” (see Article 9 of the 2015 Order). The Design and Access Statement was incorporated by reference (see principles (2) and (3) in R v Ashford BC, ex p Shepway DC [1998] EWHC 3094 (Admin) at 19), in the operative words in Part 2 of the Decision Notice (§29 above): “in accordance with the application”. There are two key passages in the Design and Access Statement. These would, in any event, be a legitimate aid to construction of an ambiguity in the Planning Permission (principle (4) in Shepway). The two key passages are in Sections 4.3 and 6 and read as follows:

An important reference point is Draft EN-3. It says this, under a heading “Flexibility”:

Viewed in the light of the context, and as a matter of common sense – in interpreting Planning Conditions 4 and 12 straightforwardly, sensibly, purposively and together – matters such as “the dimensions of the panels and their layout and spacing” fall squarely within those matters whose precise details as above-ground structures were required to be submitted and approved, with agreed details considered in the context of visual amenity and the assessment of appearance (Condition 12). Condition 4 “strict accordance” must be construed as allowing this flexibility. That is the argument.

I am unable to accept this interpretation of Planning Conditions 4 and 12. I agree with Mr Harwood KC on this issue. The position, in my judgment, is as follows:

There is a statutory requirement, found in Article 7 of the 2015 Order, that an application for planning permission be accompanied by a plan identifying the land to which the application relates and “any other plans, drawings and information necessary to describe the development which is the subject of the application” (see Barnett v Communities Secretary [2008] EWHC 1601 (Admin) [2009] 1 P & CR 24 at §20). Plans, drawings and information may be revised prior to the consideration and grant of planning permission. A grant of planning permission can involve identifying “approved plans”. There can be a planning condition requiring ‘broad’ conformity (as exemplified in Swire) or a condition requiring ‘strict’ adherence. There can be a subsequent stage of approving the reserved matters in the context of an outline planning permission, or of approving particular details in the context of a full planning permission. We know (Lambeth at §19), that the starting point and usually the end point is to find a natural and ordinary meaning of the words used, viewed in that particular context (statutory or otherwise) and in the light of common sense. This is a full planning permission and there are identifiable Approved Plans (§30 above). But crucially, this is unmistakably a planning permission in which Planning Condition 4 requires that the development:

In Swire, Holgate J described the ordinary and natural meaning of the phrase “in accordance with” as connoting ‘agreement or harmony’ (§43) which he later described as falling short of “rigid adherence” (§§60 and 61). But Holgate J also addressed the use of a word like strict or “strictly”. So, he began (§43)

The Barrett/Hardy Interpretation involves rewriting Condition 4, so that “strict” yields in the face of (a) certain wording in the Approved Plans and/or (b) certain passages in the Design and Access Statement. That is a rewriting of Planning Condition 4 as if it required that the development:

It is quite right that Planning Conditions 4 and 12 must be read in the light of one another, straightforwardly, reading the Planning Permission as a whole. I accept that Planning Condition 12 is a requirement for the “precise details” of “any above-ground structure”. This includes all those above-ground structures which are the subject of the Approved Plans. But it does not follow that the function of Condition 12 is to approve details which can diverge from the Approved Plans. Not all of the “details” of the above-ground structures have been addressed in the Approved Plans. There are gaps. These gaps “include”, but are not limited to, “colours and finishes”. The gaps engage questions about visual amenity, engaging the landscape-related contents of County Durham Plan Policy 39 and National Planning Policy Framework Part 15. All of this is addressed in Condition 12. It makes sense to have the “details” – including those which are already required to accord with the Approved Plans – to see the ‘final product’. There is no requirement or contemplation that the details stage (Condition 12) must involve a referral back to the Committee. Condition 12 has a different ‘when’, and it stands to have a different ‘who’. If a divergence were permissible at a details-approval stage, it would be out of the control of the Committee. If Condition 12 is allowing divergence from the Approved Plans, what is the nature of this permissible divergence? Must it remain within the scope of ‘broad accordance’ (in the Swire sense) with those approved plans (as if deleting the word “strict” for above-ground structures. Is the permitted divergence limited to “details” engaging visual amenity (and the landscape-related contents of Policy 39 and Part 15)? Suppose a divergence in respect of approved plans engaging amenity and pollution (Policy 31). Or a divergence as to “ancillary buildings” engaging Policy 33 (renewable and low carbon energy), which says proposals “should include details” of “ancillary buildings”. These are Policies listed in the Reason for Condition 4, but unmentioned in the Reason for Condition 12. These problems with Condition 12, like the various rewritings of Condition 4, arise on the Barrett/Hardy Interpretation, but not on a straightforward interpretation of the wording and structure. Nothing, in my judgment, turns on whether the first sentence of the Reason for Condition 12 is describing ensuring accordance with new approved documents at the Condition 12 stage. That begs, but does not answer, the questions about gaps and divergence. But, if it did matter, the Reason for Condition 12 is consistent with the Condition 12 stage operating to deal with the gaps (engaging visual amenity and landscape-related policies) and to ensure that the need for strict adherence in Condition 4 has been ‘followed through’ to the finalised details.

It is right that plans and drawings can use words like “preliminary” or “typical” or “draft”. A document which is “draft” becomes final when it is adopted as approved. The same is true of a plan or drawing saying it is “preliminary” or “typical”. What Planning Condition 4 does is to take documents and change their status into “approved plans”. It stamps them: “Approved Plan – Strict Accordance Required”. It fixes them in the sense of imposing the duty of carrying out the development “in strict accordance” with them. It defines the consent. I do not accept that the meaning of Condition 4 turns on whether certain words do, or do not, appear in the document which becomes the Approved Plan for strict accordance.

I do not accept that the Design and Access Statement was incorporated by reference by the Planning Permission; nor that there is an ambiguity which this extrinsic document operates to resolve. The language of Part 2 of the Planning Permission states “in accordance with the application and plans submitted subject to the following conditions and reasons”. So far as concerns the “plans submitted”, it is the “following conditions” which identify “Approved Plans’ and other approved documents (eg. Planning Condition 8). I agree with Mr Harwood KC that the ordinary and natural meaning of “the application” is the application form, which was required to be filed, and which was headed “Application for Planning Permission”. Mr Barrett and Mr Hardy – when I asked them – gave me their list of those documents said to be ‘incorporated by reference’. The list was the application form, the Design and Access Statement, the Planning Statement, and “relevant emails”. I can find no support for that list, or an expanded or contracted version of it, being incorporated by reference. That list would not, for example, include the covering letter dated 22 June 2022 which accompanied the application form. It would not include the 17-item list, in that covering letter, said to have comprised the “submission” (including 18 plans). The Design and Access Statement was not in that 17-item list. Nor was the Planning Statement. There is no reference anywhere in the Planning Permission on any of its Conditions to the Design and Access Statement. It would undermine clarity and straightforwardness if a Planning Permission such as that in the present case were treated as having incorporated by reference unspecified supporting documents, still less “relevant emails” (“correspondence passing between the parties”: Hillside §27), whatever these are. The Design and Access Statement, moreover, was addressing the June 2022 position, including PDL-04 and PNL-25/17. If an informed reader were required – in order to understand the meaning of a Planning Permission – to draw through the history and all the background application documents, they would find that the initially proposed layout and design had been superseded. They would also find that the description within section 6 of the Design and Access Statement (§40 above) of a proposed planning condition was not adopted in the Council’s decision. Nor is it adopted in the Officer Report which was published 5 working days before the meeting and available to everyone. Instead, what was adopted was Condition 4 with its duty of strict accordance, and the express opening words of the reason in Condition 12.

Then there is the reliance on Draft EN-3. What is said in the “flexibility” passage from Draft EN-3 is that, depending on commercial agreements under consideration, a solar farm operator “may not” know “precisely” which panels would be procured for the site until sometime after consent had been granted. It is not said that that will inevitably or invariably be the position. The flexibility described is contingent: “if panel details, or other or any other relevant information, are not available’. It is in that context that “some flexibility should be provided in the consent”. It is said to be “likely” that flexibility will be needed in relation to dimensions of panels and layout and spacing in the case of solar farms. But it is not said to be inevitably or inevitably needed. It is then said that flexibility “may be sought” for energy storage or substitute panels. But the Barrett/Hardy Interpretation would involve a far greater flexibility than this. Draft EN-3 adds that, where details are not available, the applicant must assess the “worst case” effect that the project could have and the maximum impact case scenario. But a reader who delved into the story of this planning application and its evaluative consideration, would find no identification by the Developer or in the Officer Report that what is being addressed is worst-case effects for maximum adverse effects. That course could have been taken, and the Committee would have needed to think about what it was ‘fixing’ and what flexibility was being built in. On the Barrett/Hardy Interpretation, the final “location” of solar panels could, in principle, involve a greater blue area, or a different configuration of blue and green areas. I can see nothing in the paragraphs in Draft EN-3 which demonstrate that it would somehow be contrary to common sense to achieve certainty through strict adherence, followed through prior to commencement. It is for planning officers to decide whether to put forward “strict accordance” conditions, for developers to respond to an officer report and point out issues, and for the Planning Committee to decide what conditions to adopt. There is a statutory mechanism for applications to develop land “without complying with conditions subject to which a previous planning permission was granted” (1990 Act s.73; Milne-Skillman §51). An NMA may be available to “formalise minor differences” as against “approved layout plans” (R (Fulford Parish Council) v City of York Council [2019] EWCA Civ 1359 [2020] PTSR 152 at §44). What is needed is clarity, and the virtues of straightforwardness, in a planning permission and its conditions meaning what they say, so that an informed or interested individual being able readily to understand them and readily find the materials incorporated and approved.

What follows from this analysis is two things. First, that the Planning Permission authorised the solar farm development if – but only if – it was built in “strict accordance” with the Approved Plans, including PDL-08 and PNL-6839. Second, that the Planning Permission did not authorise a later Details-Approval stage at which there could be departures from this “strict accordance” duty.

This takes me directly to a second aspect of the November 2023 NMA. This involved the insertion into Planning Condition of these words at the beginning of Condition 12:

This brings us back to the interpretive premise. The NMA decision-maker Mr Kelleher was told by the November 2023 Officer Report that this amendment was “to provide clarity”, that the scope of Condition 12 was “not curtailed by any detail contained on an Approved Plan under Condition 4”, it being “the Council’s intention when determining the original planning application that details shown on plans approved as part of Condition 4 could be superseded by details provided subsequently to discharge Condition 12”, which the amendment making “this intention explicitly clear”. The argument by Mr Barrett and Mr Hardy as to this aspect of the NMA came to this: “nothing to see here”. They say Planning Conditions 4 and 12, correctly interpreted, already allow divergence from the Approved Plans, at the Planning Condition 12 Details-Approval stage. All that was happening was that this was being spelled out. They accept that – if that is wrong – this aspect of the NMA Amendment cannot stand. In my judgment, for the reasons that I have given, it is indeed wrong. The consequence follows. There was an error of law (as in Milne-Skillman) vitiating any planning judgment. In those circumstances, nothing turns on Mr Harwood KC’s further point, that the wording of the revised Condition 12 being insufficient to permit divergence from Condition 4 “strict accordance”. I disagree with him about that. With this express wording, Condition 12 would be allowing a Details-Approval stage enabling a release from the Condition 4 “strict accordance”. Which links to the materiality of the purported amendment, unrecognised because of the misdirection.

This is a convenient point at which to mention an email dated 19 June 2023, sent by Ms Teasdale, Principal Planning Officer and co-author of the November 2023 NMA Officer Report. Local resident Mr Davies raised a question as to “the number of panels to be installed on the site following the approval” and Ms Teasdale replied:

Mr Galloway has explained, in his witness evidence in these proceedings, what he was able to do by taking Approved Plans PNL-6839 and PDL-08. If you look with care at PDL-08 (§26 above), you can count up the module tables described in the key (30x2 and 15x2), for each of the fields. This gives a count of panels (configured as 30x2=60 and 15x2=30). This panel count gives 110,640 panels. This panel-count can also produce a surface-area count. This is done by taking the dimensions in PNL-6839, with those panels configured – in rows of two – matching the module tables in PDL-08. The panel dimensions from PNL-6839 (2384mm x 1303mm) are simply multiplied by the 110,640 units seen in PDL-08. This gives a surface area count of 343,687m2. All of this was explained, with care and clarity, in Mr Galloway’s witness statement evidence dated 23 August 2023 and 29 November 2023. The Council and the Developer, with every opportunity, have not contested that – if this exercise is undertaken in this way – they, or I, or any member of the public would get the same 110,640 units and 343,687m2.

Mr Galloway accessed PNL-6839 from the planning portal and opened it, as a pdf, on a computer. His witness evidence describes the digital file name for Approved Plan PNL-6839. On this topic, the Council and the Developer have responded, adding evidence of their own. The evidential picture is this. Mr Galloway explains, and shows by means of a screenshot, PNL-6839 as having this electronic file name:

As Mr Galloway’s August 2023 witness evidence established, and as Mr Hardy’s skeleton argument rightly recognises:

Mr Galloway’s Panel-Count of 110,640 panels, multiplied by the maximum power output of 685W for each panel, gives an aggregate of 75.78MW for the solar farm, if built in strict accordance with PDL-08 and PNL-6839 using Trina 685 Wp panels. Mr Galloway’s witness evidence describes this calculation. Nobody has suggested that he has got the maths wrong. When I put it to him, Mr Hardy accepted that this method – used in Mr Galloway’s Capacity-Count – was the Combined-Panels Method, described in Draft EN-3 §§2.48.6 and 2.48.7.

I can now turn to the Agreed Issues (§7 above) Here is the first line of legal challenge. It entails taking Agreed Issue [AI.1] together with [AI.3a] and [A1.3b]. Here they are again:

On analysis, the Harwood Argument on this part of the case has a triple-premise. For the argument to get off the ground, three things must all be correct. (1) One is that “capacity” in the Statutory Capacity Threshold necessarily means as measured by the Combined-Panels Method. (2) Another is that the legally correct interpretation of Planning Conditions 4 and 12 is that development in “strict accordance” with Approved Panels PDL-08 and PNL-6839 permits no divergence at the Condition 12 details-approval stage. (3) Another is that Trina 685 Wp solar panels are incorporated as part of Approved Plan PNL-6839.

Here is the essence of the Harwood Argument on this part of the case, as I saw it. (1) PNL-6839 is an Approved Plan which had been submitted “electronically”, in respect of an application for planning permission which had been “made by electronic communications”, pursuant to Article 7 of the 2015 Order. Documents submitted electronically are uploaded to a planning portal. They are accessible and viewable digitally. Colours can be seen. Details can be seen. Electronic file names can be seen. This is the digital world in which the planning application was made, plans were submitted and accessible, and “Approved Plans” were accessible. (2) The digital file name for the pdf of Approved Plan PNL-6839 includes “Trina 685 Wp” (or “Tr-685” for short) (§56 above). That digital file name forms part of the Approved Plan. This means, on the objectively correct interpretation of the Planning Permission, Condition 4 was requiring the development – if it were to proceed – to be in strict accordance with an Approved Plan identifying this specific solar panel as the product required to be used. (3) Trina 685 Wp is the commercially available solar panel whose dimensions (2384mm x 1303mm) precisely match PNL-6839 (§27 above). Mr Galloway’s Panel-Count and Capacity-Count shows that 110,640 of the Trina 685 Wp panels (each with its maximum power output of 685W) gives an aggregate of 75.78MW (§§55, 58 above). That is the capacity of the solar farm which has been authorised, when built in “strict accordance” with the Approved Plans. That is what Planning Condition 4 requires; and no departure from this is permissible at the Planning Condition 12 details-approval stage. (4) It follows that the Council has purported to grant planning permission for a generating station which cannot in law be authorised by planning permission, and which cannot in law be built out in full. The Court can rule on whether a generating station exceeds the Statutory Capacity Threshold so that development consent was needed (Durham §§28-38), and planning permission was inapt (cf. Newbury District Council v Environment Secretary [1981] AC 578, 599C, 602A-B, 605G, 615C, 624D-E). This purportedly authorised development was not “severable” (cf. Hillside Parks Ltd v Snowdonia National Park Authority [2022] UKSC 30 [2022] 1 WLR 5077 at §§46, 50, 55) and the Planning Permission cannot have a part-authorising effect (cf. Durham at §§51-54). The NMA cannot save an unlawful Planning Permission and, in any event, is flawed in public law terms. The Court should quash the planning permission or declare it to be of no legal effect.

I cannot accept this first line of challenge. There are two independent reasons why. First, I cannot accept premise (3) (see §60 above). On this part of the case, I agree with Mr Barrett and Mr Hardy. Here are my reasons. (1) The Planning Permission, and Planning Condition 4, should be – and can be – given a clear and straightforward interpretation. PNL-6839 was the Approved Plan. Condition 4 required “strict accordance” with it. It was important to be able, straightforwardly, to see and understand what substantive content attracted this obligation of strict accordance. That means by looking at what is on the face of the Approved Plan, whether printed or accessed electronically. This is reinforced by the fact that PNL-6839 says – on its face, when printed – that its “paper size” is “A3”. It is designed and intended to be capable of being understood, complied with, and enforced by being printed in colour on A3 paper. (2) It would undermine the certainty and practical workability of the planning regime if a planning permission could only be understood by investigating the electronic metadata of the pdf file. It would undermine the decision-making process if Committee members were required to take this step, in order to understand what was being recommended to them and what they were doing. It would undermine the enforcement process too. Understanding a planning permission should not be an exercise in metadata detective work. This problem is reinforced by the fact that metadata may look different depending on format (here, pdf) and browser. PNL-6839, when printed, bears no reference to Trina or Trina 685 Wp. Indeed, there was no reference to Trina 685 Wp anywhere in the March 2023 Officer Report, or anywhere else in the Planning Permission. (3) Test it this way. Suppose trinasolar were, now or before commencement of the development, to produce a Trina 400 Wp with a lower power capacity, but the same panel dimensions (2384mm x 1303mm). Or suppose another manufacturer produced a lower-capacity panel with these dimensions. Would the Developer be able to build out the development, in ‘strict accordance’ with PNL-6839 and PDL-08? The answer is “yes”. That means premise (3) has failed. This is fatal to the line of challenge.

Secondly, and independently, I cannot accept premise (1) (§60 above). That is, independently, fatal. I have not been persuaded that the Combined-Panels Method is a ‘legal litmus test’ for identifying the capacity of a solar farm. The position is as I have explained it (§§12-21 above). I was shown no statutory definition, case or commentary supporting the Combined-Panels Method as the sole legally-correct ‘interpretation’ of capacity. Mr Harwood KC did not persuade me that the Combined-Inverters Method, prospectively favoured by Central Government (and now adopted in policy EN-3 from November 2023) is contrary to law. Nor did Mr Hardy or Mr Barrett persuade me that the Combined-Panels Method is contrary to law and the Combined-Inverters Method is the sole legally-correct ‘interpretation’ of capacity. The picture is that there are two recognise methods of measurement, in the reasonable ‘application’ of the Statutory Capacity Threshold, each of which has an important Accompanying Proviso. All of which means that premise (1) has failed. This is, independently, fatal to the line of challenge.

I should make the following clear. I do accept premise (2), for reasons which I have explained (see §§44-51 above). And I do accept that the second aspect of the NMA was not lawfully adopted (§53 above). Had I also accepted premise (1) and premise (3), I would have gone on to conclude that the generating station for which planning permission was purportedly granted by the Council on 13 July 2023 was, by virtue of Condition 4, a development for which development consent was required (2008 Act s.31), and a development excluded from the requirement of planning permission (1990 Act s.57(1A)), by virtue of its expected capacity when constructed (2008 Act s.15(1) and (2)(c)). The appropriate remedy would likely have been a declaration. But that position does not arise.

In Durham, Chamberlain J considered (at §§49-55) whether a local planning authority (and an inspector on appeal) could, in principle, consider the question of planning permission in relation to an NSI Project exceeding the Statutory Capacity Threshold. He held that they could (§§55, 56(c)), for various reasons including the absence of a ‘statutory preclusion’ on planning permission (§51), and utility by reason of this part-authorising effect (§54):

In my judgment, whether this same analysis could apply – on the particular “facts of the present case” – would depend on the legally correct interpretation of the Planning Permission. If, on its correct interpretation, it allows the project to be ‘built-out in part’ then it could lawfully be implemented by constructing something smaller. Sometimes, planning permission will expressly speak of development “up to” a number of units. But sometimes, “multi-unit developments” – as is typically the case with a “housing estate, comprising multiple units” – are “an integrated scheme which cannot be severed” (see Hillside §71). I interpose (as to Durham §51) that, as I see it, Parliament did not use a ‘statutory preclusion’, because it was disapplying (1990 Act s.57(1A)), for an NSI Project, a requirement which otherwise brings planning permission into play (s.57(1)). On the facts of the present case, the correct interpretation of the July 2023 Planning Permission is – in my judgment – for an integrated and non-severable scheme. If this authorised solar farm did exceed the Statutory Capacity Threshold, I do not see how the Developer could simply choose to build a smaller solar farm on part of the fields (eg. the blue areas nearest to the village).

I can now turn to the third and fourth aspects of the November 2023 NMA. The third aspect involved a replacement version of PNL-6839, removing the Trina 685 Wp electronic file name metadata. This NMA was made on the basis that it did not affect the correct interpretation of the Planning Permission. I have concluded (§62 above), as a matter of law, that this was correct.

The fourth aspect of the NMA was the inclusion of this new provision, at the end of Planning Condition 4:

Again, suppose I had found: (a) that Approved Plan PNL-6839 incorporates the Trina 685 Wp; (b) that the Combined-Panels Method is the objectively correct interpretation (or sole reasonable application) of “capacity” in the Statutory Capacity Threshold; and (c) that no decision-making authority could regard 110,640 panels and an overall 75.78MW as acceptable “Overplanting”. On that basis, the fourth aspect of the NMA would not – in my judgment – have saved the Planning Permission. It is based on “export capacity” measured in “AC”.

I return to the interpretive premise. The November 2023 Officer Report told the NMA decision-maker (Mr Kelleher) that this new provision “is commensurate with the details in the planning application and available grid connection and is for clarity only”; it being “clear throughout the application documents … that the electrical output from the solar farm, as proposed and approved, would not exceed 49.9MW”. It was certainly stated throughout the application documents, and stated in the March 2023 Officer Report, that the solar farm was intended to provide clean energy “for up to 13,861 homes”. The Planning Statement and the Design and Access Statement had referred to “the installation of a 49.9MW solar farm”, as did the Officers’ screening opinion. But the March 2023 Officer Report did not make reference to 49.9MW, nor to restricting the solar farm to 49.9MW; still less to an “export capacity” of 49.9MW “AC”. I can see no basis for implying this capped export capacity from the Planning Permission itself. I am not persuaded by the defence of this fourth aspect of the NMA – that it spelled out something which was already there. I can see that, on one view, this amendment is a change which is favourable to Mr Gallagher’s position. But, in my judgment, it is nevertheless a purported NMA which is vitiated by the misdirection in law that it did no more than spell out a position already applicable, on a correct interpretation of the Planning Permission.

I can turn now to the second line of legal challenge in this case. It features Agreed Issues [A1.2] and [A1.3c], which can be taken together. Here they are:

The Planning Permission, correctly interpreted, required “strict accordance” with Approved Plans PDL-08 and PNL-6839. That means 110,640 panels whose dimensions are 2384mm x 1303mm. That is the “very large number and area of solar panels” which “the planning permission, correctly interpreted, approved”. Using Trina 685 Wp, as a commercially available solar panel whose dimensions (2384mm x 1303mm) precisely match Approved Plan PNL-6839, the capacity of the solar farm is 75.78MW. That method is Mr Galloway’s Panel-Count and Capacity-Count. As was accepted by Mr Hardy, this is the Combined-Panels Method. That is the method which was the recognised conventional way of calculating capacity of a solar farm, both at the time of the March 2023 Resolution and the July 2023 Planning Permission (see Draft EN-3 §§2.48.6 and 2.48.7). The same picture would be seen taking the power of the typical solar panel for large-scale developments, described in Draft EN-3 at §2.47.2 (§18 above). PDL-08 and PNL-6938 involve a surface area count of 343,687m2. Draft EN-3 described a typical panel with a surface area of 2m2 and an output of 450W. That means an output of 225W per m2. That gives 77.33MW, at a power of 225W per m2 over a surface area of 343,687m2.

Two things follow from this. The first is that a generating station within the Statutory Capacity Threshold of 50MW – using this recognised method of measuring capacity – would be achieved only if the panel power is reduced by one-third of the power of commercially available or typical solar panels. That is to say, “panels … considerably below the power of panels conventionally used and available”. The second is that a generating station within the Statutory Capacity Threshold of 50MW – using this recognised method of measuring capacity – could be achieved, using commercially available or typical solar panels, with two-thirds of the footprint of solar panel coverage. That is to say, the Council was approving “more panels over a larger area than were required to produce” a generating capacity within the within the Statutory Capacity Threshold of 50MW.

The second of these was a material consideration in planning terms. It was obviously material, in all the circumstances including these: (a) the size of the development was a central issue; (b) the Planning Committee was approving a solar farm to provide clean energy “for up to 13,861 homes” on the basis that this footprint was suitable for such a solar farm; (c) the question was whether solar panel arrays, with the narrower spacing found in PDL-08, could achieve a solar farm of appropriate capacity using less blue space and with more green space; (d) that question could not, on the correct interpretation of Planning Conditions 4 and 12, be addressed at the details-approval stage under Condition 12; (e) the question whether more green space was achievable, and if so where it should be, needed to be addressed by the Planning Committee; (f) the Developer failed to address this important point, which was missed by Officers, and it was not for objectors to raise the point (though Mr Davies, in effect, did). The failure to have regard to this obviously material consideration vitiated the Planning Permission, which cannot be saved by the NMA and should be quashed, so that the Planning Committee can reconsider this aspect. That is the Harwood Argument.

There are two preliminary points to make. First, a principal plank of Mr Barrett and Mr Hardy’s resistance to this second line of challenge was as follows: (1) that the final positioning of solar panels – and questions about how much blue and green and where – would all fall squarely within the Details-Approval stage under Planning Condition 12; (2) that the Council would be duty-bound to consider any point raised by an interested person – as this point has now clearly been by Mr Galloway; and (3) that in real-world terms, the point may well be referred back to the Planning Committee for final Condition 12 approval. That plank in the line of resistance fails. It rests on what I have held to be a legally incorrect interpretation of Planning Conditions 4 and 12 (see §§44-51 above).

Secondly, I have had to consider whether this second line of challenge assumes as a premise that “capacity” for the purposes of the Statutory Capacity Threshold must necessarily mean measurement using the Combined-Panels Method (see §16 above). In my judgment, such an assumption is not a necessary link in the chain of the Harwood Argument. The Combined-Panels Method was the recognised conventional way of calculating capacity of a solar farm, both at the time of the March 2023 Resolution and the July 2023 Planning Permission.

I have identified the plank of resistance, based on the interpretation of Planning Conditions 4 and 12 (see §75 above). In addition to that, Mr Barrett and Mr Hardy submitted, in essence as I saw it, as follows. First, the basic function of the Council as planning authority was to deal with the application that was before them. The Committee had to decide whether a development in these fields and with this footprint of solar panels was acceptable in planning terms, by reference to visual amenity, landscape impacts, effect on rights of way, and so on. Once it was recognised as acceptable with this footprint – as it was – there was no material planning consideration involving asking whether its solar panel footprint could have been smaller.

Secondly, the Committee had amply sufficient information about the sizing of the project. It is true that the December 2022 PNL-6839 used a narrower spacing of solar panel arrays (2482mm) than PNL-25/17 (6347mm). But the Officer Report referred to this (§§9-10):

Thirdly, the question whether the Committee was “approving more panels over a larger area than were required” for a 50MW solar farm was not a material consideration. Officers, and the Committee, could have chosen to consider it. But it was not a mandatory consideration; nor a question which was “obviously relevant”, applying the applicable reasonableness standard (R (Barr) v North Somerset Council [2015] EWHC 1735 (Admin) at §§74-76). Doubtless, this question could and would have been addressed, had it been raised by any third party. But it was not. Which itself undermines the idea that it was “obvious”. Mr Galloway’s Panel-Count and Capacity-Count were undertaken by him, for the first time, in bringing this claim for judicial review. The questions raised by Mr Davies, to which Ms Teasdale responded in the Teasdale Email, came after the Committee meeting and resolution, and did not squarely raise the issue either. The Officer Report was entirely adequate in law, was in no sense materially misleading. There was no failure of reasonably adequate enquiry (Barr §§39-40). The Planning Committee’s decision was not vitiated by the failure to have regard to a material consideration. That is the counter-argument.

Before I turn to my discussion, there are two things I need to do. The first is to explain what it meant to say that “the layout has been amended” with “a gap of approximately 2.5m” (Officer Report §§9-10). I have described PDL-04 (June 2022), with its lines of blue rectangles representing module tables of solar panels (28x4, 14x4 or 7x4): see §25 above. These were the panels configured in 4 rows, whose dimensions were in PNL-24/7 (June 2022): see §28 above. I have described PDL-08 (December 2022), with its narrower lines of blue rectangles, representing different module tables (30x2 and 15x2): see §26 above. These are the panels configured in 2 rows, whose dimensions are in PNL-6839: see §27 above. The closer lines of module tables are clearly visible in PDL-08. The blue rectangles are much tighter together than in PDL-04. They can be counted, for each field. For example, the field furthest north had 38 blue lines shown in PDL-04 but has 57 blue lines shown in PDL-08 (these were counted by Mr Galloway, and the accuracy of the count has not been disputed). For that one illustrative field, this is a 50% increase in the number of rows of module tables. The closer module tables are very clear from PNL-6839. As I explained, in PNL-6839 the land-space, in between each array of panels is 2482mm. As I also explained, in PNL-24/7, it was 6347mm. So, the arrays of solar panels are much closer together. They were 6.3m apart. They are now 2.4m apart. This was the “layout” having been “amended”, with the “approximately 2.5m” between “each line of solar panels”, described in the Officer Report (§78 above).

The second thing I need to do is to identify the applicable legal principle. We are very firmly in the realms of ‘soft’ supervisory review, where primary evaluative judgments were for the planning officers and planning decision-makers. They are afforded considerable latitude. The standard is unreasonableness. There is no room for substitution of judgment; and no room for hindsight. It is a high bar. The following is derived from Barr at §§74-76 (the same point, about the reasonableness standard of review, is made as to sufficiency of enquiry in Barr at §§39-40):

I have not been persuaded by Mr Barrett and Mr Hardy’s counter-argument. In my judgment, Mr Harwood KC’s second line of challenge is correct: there was, applying the high bar of reasonableness, an obviously material consideration as to whether the grant of Planning Permission was “approving more panels over a larger area than were required” for a 50MW solar farm, and as to the implications of that for the blue areas of solar panel coverage and remaining green areas, where those blue and green areas were to be, and who would decide. These are the reasons why I have arrived at that conclusion:

First, there was an important Statutory Capacity Threshold, found in the statutory scheme, governing whether the solar farm was authorisable by the Council has a local planning authority (§§9-11 above). Unless the effect of the Planning Permission were to approve a ‘severable’ development (see §66 above), the purported planning permission would be of no effect. If the development were a solar farm with a capacity over the Statutory Capacity Threshold of 50MW, it needed the rigours of Central Government development consent.

Secondly, there were two recognised methods for measuring the capacity of a solar farm, for the purposes of applying the Statutory Capacity Threshold (§§16-20 above). One was the conventionally acceptable Combined-Panels Method, with its Accompanying Proviso. The other was an approach which Central Government was proposing for future adoption, namely the Combined-Inverters Method, also with its Accompanying Proviso. There were descriptions of these methods, readily available in public domain materials. They addressed an important question about whether the solar farm could reasonably be said to be within the Statutory Capacity Threshold.

Thirdly, size was important. The size of the blue footprint of solar panel coverage was a principal controversial issue. There was a recognised “degree of landscape harm” with a policy “conflict … in respect of public rights of way”. There were letters of objection and concerns against the proposal. But these were being assessed as outweighed by the benefits, in terms of energy supply, of having a generating station capable of uploading green energy for up to 13,861 homes (§5 above). The March 2023 Officer Report referred (at §§63-64) to 466 objections, nearly all of which raised impact to landscape as the main issue. The amount of the blue areas (solar panel coverage), and the remaining green areas, was significant. It was significant as to: (a) how much blue and how much green; (b) if there could be less blue and greener, where should that be; and (c) who should be making that decision.

Fourthly, the Developer had been communicating with planning officers about size of the footprint, and about the efficiency of the layout. When the revised layout (PDL-08) and panel elevation (PNL-6839) were put forward in December 2022, the Developer said it had listened to the feedback, gained an understanding of the main concerns, and amended the scheme with the revised layout. That meant a reduced blue area and a greater green area nearest the village – described as “greater visual containment of the scheme”, achieving “further reductions in landscape and visual effects”. The Developer also described “a highly efficient layout and use of the land”. It did so, moreover, by making reference to the acreage description in Draft EN-3 §2.47.2. The quotation (§15 above) referenced a solar farm requiring “2 to 4 acres for each MW of output”. The point was made by the Developer that this proposed development of 230 acres (93 hectares) would involve 79 acres (31.1 hectares) of solar panel coverage. There are two problems with this. First, as Mr Harwood KC pointed out in his reply, the description in Draft EN-3 §2.47.2 is surely to fields ‘covered with solar panels’ (see §22 above). And as Mr Galloway’s witness evidence explains, the 31.1 hectares “occupied by the solar panel arrays” (Officer Report §9) is apparently based on adding together all those strips of land over which the module tables stand. That does not appear to be comparing like with like. Secondly, none of this is a method to measure capacity. The methods to measure capacity appear – in the very same source Draft EN-3 – under the relevant heading and on the very next page. See §§15-20 above. There is also – in the same §2.47.2 as was partially quoted by the Developer – a description of a typical solar panel (225W per m2): see §18 above. None of this was brought to the attention of planning officers by the Developer. There was no exploration of how capacity was being, or should be, measured.

Fifthly, it was the very revision of the layout, relied on in December 2022 for its “greater visual containment”, “further reductions in landscape and visual effects” and “highly efficient layout and use of the land” which contained the layout change which posed the all-important question. The Developer emphasised the reduction in the blue area in PDL-08, by contrast with PDL-04. Less blue; more green (see §§25-26 above). But there was also a change in the spacing between lines of solar panels (see §80 above). This was the new 2.4m (2482mm), in place of the previous 6.3m (6347mm). Planning officers were onto this. The March 2023 Officer Report specifically drew attention to the “gap of approximately 2.5m” (§10), and the “changes … made” (§197): see §§78 and 5 above. It is fair to say that the Teasdale Email (§54 above) went on to include references to both the old, and the new, spacing. Anyhow, the all-important question was this: now that you recognise, and we recognise, that your solar panels can be so much more closely configured – 2.4m apart rather than 6.3m apart – why do you still need so much blue to get to capacity? That was never addressed.

Sixthly, on the evidence before me, all of this can really matter. We have three typical, commercially available solar panels. One is the Trina 650 Wp (§57 above). Its dimensions perfectly match those in PNL-6839 of December 2022. It is 650W per panel, or 221W per m2. Another is the Draft EN-3 §2.47.2 “typical solar panel” of September 2021 (§18 above). It is 450W per panel, and 225W per m2. The third is a panel called the Longhi 540W. Its dimensions match, perfectly, the dimensions of the panels in PNL-25/17 (1133mm by 2256mm) of June 2022. The Longhi 540W was named in the electronic file name for the pdf of PNL-25/17. Its power, as the name reflects, is 540W per panel. That is 211W per m2. So, the three ‘typical’ panels we have are 221W per m2, 225W per m2, and 211W per m2. Now take the Panel-Count (§55), which was the conventionally recognised method of measuring capacity for a solar farm (§16) described in the same source relied on by the Developer (§15). You can get everything you need from PDL-04 and PNL-25/17. You can just count them up. Mr Galloway did. This was the picture at June 2022:

Seventhly, none of this was being deferred to the Planning Condition 12 details-approval stage. That is because Planning Condition 4 was designed to require “strict accordance” with the Approved Plans. The details-approval stage did not allow for the blue coverage to get bigger; but nor did it allow for the blue coverage to get smaller. The questions of blue and green needed to be addressed by the Planning Committee, or the proposed Planning Conditions needed to be designed differently and the Committee needed to be clear as to what required “strict accordance” and could be deferred for later approval based on broad “accordance”. See §§44-51 above.

Based on the combination of these seven features, and on the very particular facts of this individual case, there was in my judgment a public law unreasonableness in not addressing whether the grant of Planning Permission was “approving more panels over a larger area than were required” for a 50MW solar farm, and as to the implications of that for the blue areas of solar panel coverage and remaining green areas, where those blue and green areas were to be, and who would decide. Mr Galloway’s second line of challenge succeeds. An important – obviously material – question was left unexplored. It could have made a difference. I cannot say that it is highly likely that the outcome for Mr Galloway would have been substantially no different. That would be to speculate and to step into the shoes of the merits-evaluative planning judgment of the planning decision-maker.

Mr Harwood KC submitted that the Developer fell short of its duty of candid disclosure in failing to explain to the Court how, internally, it calculated the capacity of the proposed solar farm. He says an adverse inference is justified, namely that the Developer had a Combined-Panels Method calculation which showed an oversized capacity, which it concealed from the Council and the Court. I agree with him on the first point, but not on the second. A method of calculating capacity must have existed and been adopted. It could have been the Grid Connection Offer. It could have been the Combined-Panels Method (with its Accompanying Proviso). It could have been the Combined-Inverters Method (with its Accompanying Proviso). In my judgment, given the centrality of capacity to this claim for judicial review, the Developer should have drawn attention to the methods of measurement of capacity found in the very source (Draft EN-3) on which it was relying for typical acreage. The Developer should have explained, with evidence, why its solar farm was and is said to be within the Statutory Capacity Threshold. I see the force of the Developer saying that the only directly relevant material was what planning officers and then the Committee were told; not material ‘internal’ to the Developer. But the Durham case held that the Court could itself consider the question of capacity. Furthermore, some ‘internal’ material has been disclosed by the Developer. The Developer disclosed an email exchange from April 2022, when an updated layout of “50.37MW” was amended to “49.9MW” in what became IDL-05. But what was the methodology for “50.37MW” and then “49.9MW”? Was this the Combined-Panel Method, before the blue coverage was adjusted in PDL-04? I received no evidence from the Developer about any calculation of capacity, at any stage. But, as I have explained, I do not draw any adverse inference that the Developer was ‘sitting on’ some damning Combined-Panels Method calculation. I do not know, cannot say, and should not speculate. This was a Developer who was saying, throughout, that it wanted to produce green energy for up to 13,861 homes. It has held a Grid Connection Offer of 49.5 MW, as the documents record. It may be that a developer would have no commercial interest in building an oversized or under-efficient generating station. The Developer was prepared to propose an express amendment to the Planning Permission, to restrict the export capacity to 50MW, but this was, deliberately, an “export capacity” in “AC”. In all the circumstances, no adverse inference is appropriate; nor is one necessary. But I do think Mr Galloway, and the Court, should have received a more transparent, evidenced explanation.

I can deal with the final Agreed Issue shortly.

Having rejected the first line of challenge, and having accepted the second line of challenge, I can now record my answers – for the reasons set out in this judgment – to the Agreed Issues (reordered them for clarity):

In the light of the line of challenge which has succeeded, I will quash the Planning Permission and the NMA, so that the decision is remitted to the Planning Committee, as primary decision-maker, to ask itself this question: in granting the Planning Permission, are we approving more panels over a larger area than are required to produce a 49.9MW solar farm? It may be that the Developer will want now to provide planning officers and the Committee with information about its approach to measuring capacity; about the application here of the Combined-Panels Method and its Accompanying Proviso (Overplanting); about the application here of the Combined-Inverters Method and its Accompanying Proviso (including total area and percentage of ground cover); and about whether its use of acreage occupied by solar panels is the same, or a different, measurement from the one on which it has been relying. The Committee will be able to revisit the question of what ‘accordance’, and what ‘strict accordance’, it wishes to impose; and what flexibility it wishes to allow. Everyone will want to think about whether less blue, and more green, is possible and appropriate; and if so, where. As I have explained, the Council and the Developer have argued that all of this was open for further consideration, and could be referred to the Committee, at the Condition 12 Details-approval stage. In the event, it will need to be considered more directly, on remittal.

Having circulated this judgment as a confidential draft, I am able to deal here with the appropriate order and any consequential matters. I am making the following Order: Under AC-2023-LDS-000229 (the main judicial review): (1) The claim for judicial review is allowed. (2) The planning permission (reference DM/22/01769/FPA) granted by the Council on 13^th July 2023 for a solar farm at Burnhope is quashed. (3) The Council do pay Mr Galloway’s costs of the proceedings summarily assessed in the sum of £32,000. Under AC-2023-LDS-000290 (the second claim for judicial review): (4) Permission to apply for judicial review is granted. (5) The Council and the Developer’s summary grounds of resistance do stand as detailed grounds and all time periods are abridged. (6) The claim for judicial review is allowed. (7) The non-material amendment (reference DM/23/03147/NMA) made by the Council on 7^th November 2023 is quashed. (8) The Council do pay Mr Galloway’s costs of the proceedings summarily assessed in the sum of £9,000.

The only contested issue in all this was the quantum of costs. Mr Barrett accepted that there should be costs orders, assessed by me, but asked me to assess the Council’s costs liability at £22,380.21 (50% of Mr Galloway’s costs of £44,760.42) and £5,000 (against £9,670.80 claimed). Mr Harwood KC submitted that no costs reduction should take the main judicial review costs below the cap of £35,000, and that the whole of £9,670.80 should be awarded for the second judicial review. I arrived at £32,000 for the main judicial review on the basis of an overall 25% reduction in Mr Galloway’s costs, where I am not awarding costs on an indemnity basis. True, the evidence and materials were relevant to the winning line of analysis. True, this would always have been a one-day substantive hearing. But the first of two main lines of argument failed, and it failed for two independent reasons (§§62-63 above). As to the second judicial review, these are reasonably-apportioned costs which were entirely avoidable. The Council should have embraced the amended grounds (§8 above). It provoked the costs of Mr Galloway’s Leading Counsel, and the partner at his firm of solicitors; and the £770 continuation fee (Judicial Review Guide 2023 p.189). The modest broad-brush reduction down to £9,000 is because I am not awarding costs on an indemnity basis.