(Sheriff MacLeod, D J Houston, J A Smith)
(Application RN SLC/82/10 – Order of 15 June 2011)
AGRICULTURE – RURAL PAYMENTS – APPEAL AGAINST REDUCTION OF SINGLE FARM AND LESS FAVOURED AREA PAYMENTS FOR OVER-DECLARATION OF ELIGIBLE LAND – EXTENT OF “PERMANENT PASTURE” – WHETHER MARRAM GRASS “HERBACEOUS FORAGE” AND PERMANENT PASTURE FOR SFP AND LFASS PURPOSES
In terms of article 34(1) of European Council Regulation (EC) No 73/2009, support under the single payment scheme is granted upon activation of a payment entitlement per eligible hectare of land. Paragraph 2(a) of Article 34 provides that eligible hectares comprise “any agricultural area of the holding … that is used for an agricultural activity …” and paragraph 2(h) defines “agricultural area” as “any area taken up by arable land, permanent pasture or permanent crops”. Article 2 of European Commission Regulation (EC) No 1120/2009 defines “permanent pasture” as “land used to grow grasses or other herbaceous forage naturally (self-seeded) or through cultivation (sown) and that has not been included in the crop rotation of the holding for five years or longer …”.
This was an appeal against penalties imposed by the respondents in respect of over-declaration of eligible land in the appellants’ 2009 Single Farm Payment and Less Favoured Area Support Scheme applications. The penalties related to a field extending to 52.26 ha comprising undulating sand dunes populated for the most part, in various densities, by marram grass and a variety of fescues but also including areas of bare, unvegetated or minimally vegetated, sand. The appellants had claimed the full extent of the field as being eligible for SFP and LFASS purposes but following an inspection the respondents decided that only 13 ha were eligible. In the course of the appeal the appellants conceded that 15 ha of bare, unvegetated or minimally vegetated, sand should not have been included in their declaration of eligible land but contended that the remainder of the field was eligible. It was unclear at the outset of the appeal whether the respondents accepted that marram grass, however dense, could constitute “herbaceous forage” and, therefore, “permanent pasture” for the purposes of the foresaid Regulations. However in the course of the appeal procedure the respondents conceded that marram grass constituted herbaceous forage and that, in appropriate densities, it could, therefore, constitute permanent pasture. At proof, however, it became apparent that the inspector who had carried out the inspection had been under the impression that marram grass could not constitute permanent pasture whatever its density.
HELD that the inspector’s misapprehension as to the status of marram grass vitiated the respondents’ decision and that said decision ought therefore to be set aside and the case remitted to the respondents to impose penalties confined to 18 ha of land which the Court, on the evidence, found to be the extent of bare, unvegetated, or minimally vegetated, sand present in the field.
The Note appended to the Court’s Order is as follows:-
 This is an appeal under The Rural Payments (Appeals) (Scotland) Regulations 2009 (“the Appeal Regulations”).
 Although the appeal proceeds in the name of Foveran Farms, the appellant is in essence Mr Raymond Davidson, who farms land owned by a family trust, of which he is one of the trustees, at Mains of Foveran, Aberdeenshire.
 On 2 December 2009 the farm was the subject of a Scottish Government Rural Payments and Inspections Directorate (SGRPID) inspection. The inspection was carried out by Mr Alan Henry, an Agricultural Officer at SGRPID’s Thainstone office. In the course of his inspection Mr Henry became concerned that there may have been a substantial over declaration of eligible land in respect of one of the fields on the farm (field 13 on production 22) in the appellant’s 2009 Single Farm Payment (SFP) and Less Favoured Area Support Scheme (LFASS) application. The field in question runs along the coast and essentially comprises an undulating sand dune landscape. It extends to 52.26 ha, all of which had been claimed as eligible for both SFP and LFASS by Mr Davidson in 2009
 As a result of his concerns, Mr Henry consulted a colleague, Mr John Reid, then a Senior Agricultural Officer at Thainstone. Mr Henry and Mr Reid then visited the field in question, in the company of Mr Davidson, on 11 December 2009. They concluded that there had indeed been a large over-claim and informed Mr Davidson that substantial penalties would follow. They did not at that time quantify the extent of the over-claim. Instead that was the subject of further work, including what was described as part of a “technical assessment”, back at their office. The eventual result was that they determined only 13 of the 52.26 ha as eligible and the rest ineligible. That resulted in the whole of the LFASS payment being withheld and a substantial reduction in SFP for that year. Mr Davidson applied for an internal SGRPID review of that decision as provided for in the Appeal Regulations. That review upheld the decision and Mr Davidson has now appealed to us. We heard the appeal at Aberdeen on 5 and 6 April 2011 when the appellants were represented by Dr Gerald Banks, Agricultural Consultant, and the respondents by Mr Donald Cameron, Advocate.
 The full hearing of the appeal had been preceded by a debate, or something more in the nature of a pre-proof hearing as it turned out to be, at Edinburgh on 23 February. That hearing identified the relevant law and it is set out in the Note which accompanied our Order of 7 March 2011 appointing the full hearing of the appeal. There is no need to repeat it here. Suffice to say that the case essentially turns on the extent of “permanent pasture” (within the meaning of Articles 2 of European Regulation (EC) No 73/2009 and European Commission Regulation (EC) No 1120/2009) present in the field in question at the time of said inspection.
 We heard evidence on that matter from the appellant, Dr. Banks, and both inspectors. We need not set out that evidence at length but will confine ourselves to the more salient and relevant points made by each.
 Before we do that, however, we would say something about the physical characteristics of the field in question. As we have said it comprises an undulating sand dune landscape. It is part of the Foveran Links Site of Special Scientific Interest. In terms of its plant life it is unevenly populated with a variety of plants, including herbaceous species, prominent among which are marram grass and fescues. In terms of the density of that plant life, it ranges from an area of around 6 ha near the middle of the field which is sufficiently densely populated to have been described in evidence as “fescue-rich”, all of which is undoubtedly eligible for SFP at least, and areas of bare sand, or sand with minimal vegetation, extending to at least 15 ha which was agreed by Dr Banks prior to the February hearing as not being eligible for either SFP or LFASS (see the appellant’s response to our Order of 17 February 2011). The rest – extending to around 31 ha – is mixed or, as Dr Banks described it, a mosaic of sand and vegetation. Completely accurate measurement of this area for eligibility purposes is perhaps impossible due to lack of precision in the definition of permanent pasture in the legislation and certainly impractical in terms of the time the exercise would take. Nevertheless in inspections such as took place here, the best assessment practically possible has to be made.
 Mr Davidson gave evidence that he had farmed Mains of Foveran for 29 years. The field in question had been used by stock throughout that time. At one time it had been used for grazing/wintering cows. It was presently used as a “buffer area” for wintering sheep from about January or February up until just before lambing time.
 In 2009 he had claimed the whole of the field for SFP and LFASS in the belief that it was wholly eligible. For LFASS purposes it was classified as severely disadvantaged.
 Where sheep ate the marram grass that encouraged other plant species to grow. If it was heavily stocked it would eventually all go into grass like the adjacent fields to the west which had once been as field 13 is now. Field 13 comprised a dynamic system, the physical features of which changed over time although the percentage of vegetated areas as against unvegetated sand did not change that much. The vegetative cover other than marram grass died down a little in the winter and revived in the spring.
 Asked as to the use made of field 13, Mr Davidson explained that although the sheep accessed the area themselves they also chased them into it particularly on bad nights. That was because it offered dry, comfortable shelter. When there their feed was supplemented by silage and cobs but they got some nutrition from the grasses growing in the field. Cattle had grazed the field far better than sheep. Cattle got more nutrition from marram grass than sheep did but sheep would it eat it too. If other sweeter grasses were available in adjacent fields the sheep – which were free to come and go as between field 13 and the fields immediately adjoining it – would prefer those. But when they were driven onto field 13 at night they would stay there all night, so they must be getting some nutritional value from it.
 So far as the inspection of 2 December was concerned, he had not accompanied Mr Henry. However he had accompanied Mr Henry and Mr Reid on their inspection of 11 December. They had been in field 13 for only 15 minutes or so. Maybe the whole time spent on the farm that day had extended to 45 minutes but not their time in field 13. Mr Davidson described the areas of the field inspected on that occasion. At the end of the inspection he had been told that there would be a deduction in his payments due to ineligibility. He had subsequently visited the Thainstone office and had been shown mapping photographs on a screen.
 Returning to the use made of field 13, Mr Davidson described it as the most valuable piece of ground he had for his farming enterprise. He explained that most of his land comprised heavy clay soil and if sheep were not taken off that from January to March you got no grass.
 Dr Banks, who runs an agricultural consultancy, gave evidence that Mr Davidson had come to him with his problem in January 2010. By that time Mr Davidson had already applied for a review of the penalty decision.
 In Dr Banks’ opinion the only way one could get a clear impression of the extent of the unvegetated land in a field such as field 13 was by aerial photography. Because of the contours of the land and the size of the field it was too big a picture to be able to see it from one, or many, points on the ground. Aerial photographs allowed one to see the whole picture. He had engaged a consultancy to carry out such an exercise using Google maps but part of the area had been obscured by cloud and the images had therefore been inadequate for that purpose. However he had been aware that the respondents themselves had been carrying out aerial photography of farms and he had been able to get a first class aerial photograph of Foveran from them. Productions 20 and 22 were based on that photograph. Under reference to these productions Dr Banks then described a “desk exercise” he had carried out, separating the vegetated from the unvegetated areas and measuring the unvegetated areas using a planimeter. The result of that exercise, shown on production 22, was that he calculated that 15 ha of field 13 were unvegetated sand. Everything shown as vegetated on the photograph should be eligible for LFASS and SFP. What was shown as bare sand was certainly ineligible for LFASS but may be eligible for SFP. He would leave that decision to the Court. His position was that the areas of bare sand, extending to 15 ha, were ineligible for LFASS and possibly for SFP but that everything else was eligible for both. [This submission by Dr Banks was a departure from the express concession made in response to our Order of 17 February 2011, referred to at para  above. As such we have not entertained it but, for what it is worth and without having heard full argument on the eligibility of the 15 ha for SFP, we think that Dr Banks’ concession was correct.]
 In cross-examination Dr Banks accepted that he had not carried out an “organic assessment” of the field. His conclusions were based on what the aerial photograph showed. It showed what was vegetated and what was not. Everything vegetated was eligible, everything unvegetated, or minimally vegetated, was not. He accepted that an aerial photograph would not differentiate between the species present, for instance what was marram and what was not. Some of the land was shown as “speckled”, partly green partly yellow, and he had made an assessment as to how much of that should be regarded as pure sand and, therefore, ineligible. His “best guess estimate” had been that 2.38 ha of the speckled area were ineligible. That was a guess but it was a reasonable one.
 Mr Cameron put to the witness that it was not possible, with a field such as this, to do more than take an educated guess. Dr Banks disagreed. The inspectors with their GPS (Global Positioning System) equipment ought to be capable of more precise measurement than he had been with his planimeter. He had carried out his planimeter measurements several times and then taken an average of the results. Use of GPS technology would have produced more accurate results. Assessing the extent of eligible forage on this field was not entirely guesswork. There were areas that could be measured reasonably precisely. He agreed that it was impossible, in practical terms, to get an absolutely accurate measurement on the ground and certainly such an exercise would not be financially worthwhile. But more accurate results could be achieved than had been achieved at inspection here and the best way to get these results was to use aerial photography.
 Asked about the inspection carried out on 11 December, Dr Banks said that all it involved was a 15 minute visit and a “bisect” of the field. Mr Reid would have seen a “snippet” of the habitats in the central part of the field. On a 52 ha field that was not a comprehensive assessment. It was not representative enough. Whatever use had been made by the inspectors of maps, photographs and GPS measurements, it did not amount to a comprehensive assessment because the data was not sufficiently representative of the field as a whole.
 When Mr Cameron explained the basis upon which Mr Henry and Mr Reid had arrived at their decision that only 13 ha of the field were eligible Dr Banks responded that that was news to him and to his client. There had been no suggestion at the review meeting that the figure of 13 ha comprised an area of 6 ha shown green on production 32 and an allowance of 25 per cent of the areas shown blue on that plan which was equivalent to 7 ha. Dr Banks did not accept that only 25 per cent of the areas shown blue on production 32 or green on production 22 was eligible. All of the areas shown green on production 22 were eligible.
 Mr Henry explained that the purpose of his visit of 2 December had been to verify the areas claimed by the applicant under the SFP, LFASS, Rural Priorities and Countryside Premium schemes.
 He explained the maps he had had available to him. He also confirmed that aerial photographs were available at Thainstone. These had been taken in June 2006. Production 22 appeared to be based on one of these photographs. He described his methodology in terms of taking measurements during an inspection but we need not go into the detail of that. He demonstrated, using his laptop computer and what are now productions 44 and 45, how he carried out his inspection of field 13, including the parts of the field visited. None of that is disputed or in any way controversial for present purposes.
 Mr Henry also spoke to productions 32 and 42, the first being a map he had prepared showing the various categories into which he had divided field 13 for eligibility calculation purposes and the second being photographs taken during the inspection of 2 December.
 His first impression had been that the whole field was ineligible for SFP as well as for LFASS purposes. That was because the vegetative cover could not support agricultural production. He had then realised that there was in fact a more productive area of the field, situated in a low basin in the central part of the field, where the land was flatter. His opinion had been that that area certainly had some value. However, overall, as it appeared that a large deduction was going to be involved, he had wanted someone else to come out and give a second opinion before completing his technical assessment. He explained that a technical assessment was an exercise carried out using technical skills and knowledge of the subjects gained at inspection.
 He had returned to Thainstone and spoken to John Reid, Senior Agricultural Officer. They had visited the farm together on 11 December. Mr Reid had been there to assess the habitats in the field. Three habitats had been identified: (i) bare sand, (ii) an area with a high content of fescue grasses, and (iii) a mixed habitat in which marram grass predominated. John Reid had been there to assess the grasses, an exercise which had involved getting down on his hands and knees to examine the plant life closely. They had been in the field for 45 minutes to an hour. John Reid had taken a view as to what was eligible. They had told Mr Davidson – who had accompanied them – that there would be a large deduction from the declared area and a severe financial penalty. But they had not at that time formed a view as to the precise extent of the deduction or amount of the penalty. In order to do that they needed the office GIS (Geographical Information System)equipment so as to work out the area of pure sand. They had plotted the edges of the clearly bare areas and the system calculated the extent of these areas. They had done the same for the fescue dominated area. The result of that exercise had been that 18 ha had been measured as pure sand, 6 ha had been measured as clearly eligible and the remaining 28.26 ha as mixed. Of this mixed habitat 25 per cent was eligible. That resulted in a finding that 13 ha of the field were eligible. In reaching that conclusion account had been taken of the photographs taken on 2 December, the aerial photographs and John Reid’s findings from his visit on 11 December. The decision arrived at was a joint decision.
 Asked whether he thought that was a fair decision, Mr Henry replied that he thought it was a generous one. There was more sandy ground than could be seen in the aerial photograph. That photograph did not give the full picture. He and Mr Reid had made a good assessment; they had used all the tools available to them.
 Asked by Dr Banks as to why the area of the mixed habitat deemed to be eligible was 25 per cent and not 35 or 50 per cent, Mr Henry replied that it was from knowledge gleaned from being on the site and having assessed the density of the sward. The extent of sand on the ground was much more than was shown on the aerial photograph.
 In answer to questioning from the Court, Mr Henry confirmed that he had been concerned to identify vegetative cover that was capable of supporting agricultural activity. Referred to photographs 6 and 8 of production 42, he confirmed his view that these showed areas which were not capable of supporting agricultural production. That was because what they showed was marram grass. Mr Henry confirmed that he would deem any area of ground where there was nothing but marram grass ineligible. He would do so both for SFP and LFASS purposes. He then explicitly confirmed to the Court that in order for an area of land to be eligible it had to have something beyond marram grass growing on it.
 Dr Banks pursued the point about the eligibility of marram grass. He asked Mr Henry what his view would have been of an area which was 50 per cent rushes. Mr Henry replied that rusheswould normally be accepted as eligible. He had seen livestock grazing rushes.
 Mr Reid explained that at the material time he had been the Scheme Manager for the SFP scheme at Thainstone. He was referred to production 43, a Scottish Government publication containing “SFP Land Based Inspection Guidance” for the use of its agricultural officers. At paragraph 5.2.4, on page 11, examples are given of land which is SFPS ineligible. The examples are “rock, scree, roads, yards, buildings, hard-standings, water/water courses, trees/forestry, and dense vegetative cover such as impenetrable gorse/scrub and monoculture stands of bracken.” The guidance then goes on “It has been confirmed that areas with vegetative cover such as open grazeable mosaics of gorse, open scrub, rush and heather are SFPS eligible”. Mr Reid confirmed that this material corresponded with his own understanding of matters.
 Alan Henry had visited Foveran on 2 December 2009 and had come back unsure about this particular field in which he had identified three different areas, (i) mixed habitat, (ii) bare sand and (iii) areas of marram grass in a dune system.
 Asked about his own understanding as to the eligibility of marram grass, Mr Reid explained that it was not the marram grass itself that was ineligible; it was rather the habitat that could make it ineligible. A mosaic of mixed grasses was definitely eligible. The cover did not necessarily have to be 100 per cent but a good green cover was required. Sand like a beach was definitely ineligible. On areas of rolling dunes where only marram grass was present a technical assessment had to be made to assess the area with green cover and, on that basis, the percentage of the dune area which was eligible.
 Mr Reid confirmed having visited Foveran on 11 December. He had not been concerned with the measurement of boundaries; all the “leg work” had been done by Mr Henry on his first visit. So far as measurements within the field were concerned, GPS would not have assisted. To try to delineate the individual areas of distinct habitat would have been well nigh impossible. To deal with that situation they had been trained to carry out a technical assessment. The area of mixed fescues had been measured by GPS. He himself had looked only at individual areas: Mr Henry had been over the whole site. As part of the technical assessment the area office GIS and GPS had been used and an assessment had been made of the percentage eligibility using evidence gained by “hands and eyes and having a good look at what’s there”. Aerial photographs were also frequently used in technical assessments.
 Before going on site he had spoken to Mr Andrew Struthers in the Land Based Inspections team at SGRPID’s Saughton office, Edinburgh so as to seek his opinion as to how to deal with a dune system and ensure a consistent approach. Mr Struthers had said that a technical assessment would be required to investigate the extent of cover.
 They had spent about 45 minutes to an hour in the field. On production 22 he identified the dark green area of fescues, the mixed dune system and a mixed habitat near the line of the fence with the adjoining field. The Court clarified with him his definition of the three different habitats and their content. By “dune system” he meant areas of sand with only marram as vegetation. The area of mixed habitat was not part of the dune system and nor was the fescue rich area. Asked about the extent of the mixed habitat area, Mr Reid said that it comprised only a narrow strip, examples of which could be seen in photographs 13 and 15 of production 42. These were areas of agricultural grasses as well as marram. Photograph 14 was a photograph of the area of fescues and photographs 9, 10, 11 and 12 were of the dune system. Photographs 5, 6 and 8 looked like areas of marram grass only. He would classify these as part of the dune system.
 He had looked at all three habitats, looking at ground conditions, plants present and extent of cover. You had to get down on your hands and knees to do that properly. Otherwise you would not know what was there. It was the amount of herbaceous cover that mattered. Having done that he had come to a concluded view as to eligibility. The fescue area was undoubtedly eligible. It had a good cover of agricultural grasses. “Good cover” was difficult to define but there had to be enough cover to be productive and to form a sward. It had to be like what you would expect in other areas where animals grazed. He and Mr Henry had not been looking for 100 per cent cover. He had also been satisfied that an area next to the fence forming the western boundary of the field was eligible. He described this as a mixed habitat area. In answer to the Court he confirmed that he had treated all of the mixed habitat area as eligible. The dune area was not all ineligible. The areas of pure sand within it were ineligible and an assessment had to be made as to the rest. There was a lot of ineligible land in the dune system. When they had returned to their office that day they had assessed 25 per cent of the dune area as being eligible.
 The final result had been that the 6 ha low-lying area of fescues was eligible and that 25 per cent of the rest was eligible. It had worked out at 13 ha being eligible. The additional 7 ha had come from the assessment of the dune system and mixed habitat area. Mr Henry had measured the areas of bare sand at his own desk back at the office and had brought the results to him. He had also measured the fescue area. The 18 ha area was pure sand and the 6 ha area fescues. The remaining 7 ha came from the large area of mixed habitat and dune system which was left and had been arrived at by applying a factor of 25 per cent eligibility to that area. He and Mr Henry had agreed that the total eligible area in the field was 13 ha.
 Dr Banks asked the witness whether there had been any mention of that total comprising two parts, one of 6 ha and one of 7 ha, at the review meeting. Mr Reid could not remember what had been said.
 Dr Banks also sought to clarify with him what he was saying about the extent and locations of the different habitats. He confirmed that the fescue area extended to 6 ha. The area of mixed habitat was along the western edge of the field and extended to maybe 2 ha and the rest was the dune system and included the 18 ha of pure sand.
 An assessment made on the basis of an aerial photograph alone was insufficient. An assessment on the ground was essential.
 The issues in the case had been clarified by the Court’s earlier Note. The respondents had confirmed that marram grass could constitute eligible forage. Mr Henry and Mr Reid had sought to identify the various habitats present in the field and the issues which arose in relation to those. Mr Reid had described the composition of the field as being 6 ha of fescues, 2 ha of mixed habitat at the edge of the field, 18 ha of pure sand, leaving 26 ha comprising the dune system.
 To suggest that only 25 per cent of the large dune area was vegetated to any extent “took a lot of swallowing” because production 22 suggested that the whole of that area was vegetated in the summer. He accepted that within that there would be some “speckling” but there were also sporadic patches of vegetation within the larger areas of otherwise bare sand. To suggest that 75 per cent of the green area shown on production 22 was sand was completely unacceptable. In any event no satisfactory explanation had been given for the choice of 25 per cent rather than 50 per cent or even 90 per cent.
 His own calculation on the basis of the photograph at production 22 was undoubtedly rough and ready and the planimeter methodology old fashioned. He did not have the facilities which were available to the respondents. But it had been a “shot” at identifying the area of vegetated land and there was still nothing from the respondents to compare it with. They had said that they had measured the areas of sand using the office AOGIS (Area Office Geographical Information System), so why had they not printed out and produced the results, if not immediately then at least in time for the review meeting or for this appeal? They had still not shown the Court exactly how they had assessed the 18 ha of unvegetated sand.
 Until the previous day the appellant had been under the impression that the whole of the 13 ha found by the respondents to be eligible for SFP and LFASS was located in the dark green area between the coastal and western dunes shown on production 22. It had never been suggested that there was an additional eligible area somewhere else. In particular Mr Davidson had not been made aware of that on 11 December nor when he had subsequently met Mr Reid at the Thainstone office. Similarly nothing had been said about this at the review meeting. Indeed the report of the review meeting had concluded that the ineligible area was ineligible because it comprised sand and marram grass. There had been no suggestion until Mr Reid’s evidence that the three habitats were not bare sand, marram grass areas and the fescue rich area. It now appeared that there was the dune system, a mixed area restricted to the western fringe of the field and the fescue rich area. Mr Reid had said that the mixed area was restricted to that western fringe and that there were 2 ha to 4 ha of land there which were eligible as a side effect of the improved land in the adjacent fields. Dr Banks was surprised to find “at this late stage” that a significant area of the land accepted by the respondents as eligible was along that western edge of the field.
 If there were 6 ha of fescue rich ground, 2 ha of eligible land along the western fringe and 18 ha of bare sand, that left 26 ha within the dune system which had not been classified as open sand. We had seen Mr Henry’s photographs of these areas, particularly numbers 5, 6 and 8 of production 42. These had been taken in mid-winter and showed the impact of winter, with the grass bleached by the salty winds. But it was easy to imagine that these would be green in the summer, as indeed they were in the aerial photograph on which production 22 was based. Generally green meant chlorophyll and chlorophyll meant plants. In this case a large part of that green area happened to be marram grass but marram grass was eligible forage. The marram grass on the top of the dunes was so thick that it prevented the total erosion of that part of the dunes. That was the role of marram grass; without marram the whole area would move and change far more rapidly. The habit of marram grass was to invade new areas. Clearly it would not be so densely concentrated in newly colonised areas but, over time, the stability produced by the marram grass made it possible for other species to survive in those areas, particularly if they were grazed. Some of Mr Henry’s photographs showed other species surviving even in mid-winter. The Macaulay Institute had listed the various species found on Foveran Links for the purposes of the SSSI citation. So it was not a monoculture of marram grass. Yet one of the first things Mr Henry had said in evidence was that his first impression on going into the field had been that the whole field was ineligible because it seemed to comprise only marram grass. He had still been of the opinion that marram grass was ineligible when he had given his evidence.
 Mr Henry’s and Mr Reid’s whole approach seemed to be that the cover had to be capable of providing sustenance to livestock. But that was to neglect the “third leg” of eligibility, which was environmental management. It worried Dr Banks that Mr Henry and Mr Reid had been carrying out inspections throughout this area under that misunderstanding. It was difficult to see where that misunderstanding had come from: it was not contained in the inspectors’ written guidance.
 Returning to the 26 ha referred to at paragraph  above, Dr Banks accepted that it would contain some areas of sand. That it might contain 10 per cent bare sand was an assessment he “could live with” but to suggest that 75 per cent of an area showing green on the aerial photograph was bare sand was not credible. No proper technical assessment could have produced that result. Off the top of his head he would suggest that that a reasonable figure for the eligible area would be not 13 but 31.5 ha, made up of 6 ha of fescues, 2 ha along the western edge and 23.5 ha of the 26 ha.
 Mr Cameron began his submissions by advising us to be careful to distinguish between what was submission and what was the giving of evidence in what Dr Banks had said. We should rely on our own recollection of the evidence rather than that of Dr Banks.
 Moving on to substantive matters, Mr Cameron was content to restate that the respondents accepted that marram grass was a “grass or other herbaceous forage”. To the extent that Mr Henry disputed this and said that areas of pure marram grass were ineligible, that was not in accordance with the respondents’ position.
 The respondents’ position was that inspections such as these required a qualitative and quantitative assessment of the grass present. It also had to be a “contextualised assessment” which looked at the extent of vegetative cover, grazeability, density, the proportion of other species present and other factors relative to the field’s value as agricultural land. Such inspections involved the respondents’ officials on the ground carrying out such an assessment. There was no magic formula for calculating the extent of eligible land. No formula was contained in the legislation or guidance. An absolutely objective assessment was impossible.
 We were concerned with the question of what, in this field, constituted “permanent pasture”. Mr Davidson had given clear and credible evidence that sheep do not find marram grass attractive. They tended to nibble at it, they had to be forced onto this field and they had to be fed supplements while they were there. This supported the respondents’ position which was that marram grass had limited value as forage.
 In relation to the measurement of what was eligible and what was not, a truly accurate picture was impossible in a situation such as this. A broad brush approach had to be taken. It was ultimately a matter of opinion as to how much of a mixed area was eligible and how much was not. Dr Banks had accepted that his own method had been rough and ready. Reliance solely on aerial photographs was insufficient. The Court should prefer the respondents’ position which was that one needed to go in amongst the vegetation rather than simply assume that what showed as green on an aerial photograph was vegetation. Photographs could be deceptive, so we should not rely on them alone. Dr Banks himself had accepted that a system of on-the-ground inspection plus aerial photographs gave a better understanding than aerial photographs alone. The respondents’ approach was more scientific and objective. They had used a combination of an aerial photograph taken in summer and two inspections in the middle of winter. Mr Henry had been on the farm all day when doing his first inspection. He had taken photographs and in giving evidence had demonstrated a very rigourous and highly precise system of inspection. He had been an impressive witness.
A full GPS assessment would not have been cost-effective, as Dr Banks had accepted.
 Mr Reid had been there to look at the grasses and habitats. Mr Cameron accepted that Mr Henry and Mr Reid had taken different views in their categorisation of the various habitats. If their evidence could not be reconciled, we should prefer Mr Henry’s because he was the person who had carried out the principal inspection. In doing so, he had walked all round the field and he had taken photographs. Mr Reid had been there only to look at the grasses. They had concurred on the 18 ha which were ineligible as bare sand and on the 6 ha central area which was clearly eligible. Therefore the disputed area, on both their analyses, was the same. Beyond that, Mr Reid had identified 2 ha of eligible land on the western fringe of the field which fell into Mr Henry’s “mixed” category. On Mr Henry’s application of a factor of 25 per cent to an area of 28 ha, he had come up with 7 additional hectares as being eligible. Mr Reid had applied the same factor to 26 ha, producing 6.5 ha so the difference between them was only 0.5ha.But if we had to choose between them, we should prefer Mr Henry’s evidence.
 It was not true to say that Mr Henry had not given a reason for selecting a factor of 25 per cent. It had come from his knowledge of the field. And 25 per cent was not minimal; it was a quarter. In Mr Henry’s view that had been a generous assessment. The whole 13 ha found to be eligible amounted to a quarter of the whole field.
 Dr Banks had criticised the respondents for not having provided a breakdown of their figures earlier. But we should accept the evidence of Mr Henry as to how the 13 ha had been arrived at. Mr Henry had been a frank, honest and professional witness. Similarly Dr Banks’ criticism that no explanation had been given in evidence as to how the 18 ha of ineligible sand had been measured should be rejected. Mr Henry had given evidence that he had zoomed in on the aerial photograph using the AOGIS system and had been able to get a line round the areas of pure sand. Furthermore Mr Davidson had asked for such an assessment of these areas and had been provided with it. Page 4 of the review report also explained how the figure of 18 ha had been reached.
 Mr Cameron’s final submission was that the decision to allow only 13 ha as eligible was quintessentially a matter of discretion, in that the extent of eligible land was not capable of accurate ascertainment. It was therefore a matter of subjective judgement and on plain and well known principles of public governing the exercise of discretion we should be very slow to disturb the inspectors’ decision unless it was perverse, procedurally improper or ultra vires. The question for us was whether the decision to assess only a quarter of the 28 ha described by Mr Henry as “mixed” was irrational, improper or illegal. “Unlikely to be correct” was not a sufficient basis for disturbing the decision.
 Asked by the Court as to the relevance of the agricultural value of the grasses present and in particular whether there had to be a minimum agricultural value, Mr Cameron repeated Mr Henry’s formulation; that the vegetative cover had to be capable of sustaining agricultural production, adding the words “at whatever level”.
 Dr Banks immediately took issue with this insistence on agricultural value. It was, he submitted, totally at odds with the thrust of EC Regulation 73/2009, which was away from agricultural production. But if agricultural value was important, Mr Davidson had given evidence that this area was one of the most valuable assets of his business in terms of agricultural activity. His sheep were outwintered there in good high welfare conditions, with no need for sheds and the illnesses to which that could lead and no footrot which might result from keeping them on the heavy land.
 Dr Banks concluded by moving for the expenses of the application. The respondents had failed to respond to Notes of Arguments which had been lodged and failed to meet other deadlines. They had failed to come to any accommodation with the appellant prior to the hearing of 23 February 2011 and failed to lodge productions vouching their position prior to that hearing. They should be found liable for the appellant’s whole expenses.
 We are satisfied that the respondents’ decision cannot stand. That is because Mr Henry’s calculation of the eligible area cannot be relied upon because of his understanding that marram grass is not eligible forage. That understanding is at odds with the position the respondents take in this appeal, although, in fairness to Mr Henry, there was a suggestion at an earlier stage that the respondents themselves were taking that position. It was only at the hearing of 23 February that the matter was put beyond doubt (see paragraph  of our Note of 7 March) and that position was reaffirmed by Mr Cameron on behalf of the respondents at the conclusion of the proof, as has been recorded above.
 Mr Henry, however, was of the view, both when he visited the field and when he gave evidence, that marram grass on its own, no matter how dense, was ineligible as permanent pasture. That he was under that impression when he carried out the inspection is seen from the entry he made in his report of 22 December 2009 (production 8) where, in the section for comments including any on GAEC and cross-compliance issues, he wrote “Deduction due to sand dunes and Marram grass”. He repeated that understanding at the start of his evidence and again, in response to detailed questioning by the Court probing precisely this point, towards the end of his evidence. Marram grass on its own was not eligible. Something else had to be present. Fescues, however, were eligible on their own. The test he applied was whether there was vegetative cover capable of sustaining agricultural production. What he was looking for was a mixture of grasses such as is present in the low lying area extending to 6 ha which everyone is agreed is eligible. Implicit in his approach was a requirement for, as the Court put it in questioning Mr Cameron, some minimal level of agricultural (or, perhaps, nutritional) value.
 The decision arrived at by Mr Henry as to the extent of eligible land in the field is consistent with that understanding but the understanding itself is at odds with the respondents’ declared position. That position, as stated on 23 February and again at proof, was not subject to any qualification such as “provided other species are also present so as to create a sward capable of supporting agricultural production”. Accepting the declared position of the respondents before this Court as correct, it follows that Mr Henry’s decision was fundamentally flawed. He was not allowing for the fact that marram grass in sufficient density, although on its own, is eligible. That is obviously a very serious misapprehension to be under given the prominence, possibly even predominance, of marram grass in this field.
 Mr Reid’s evidence does not retrieve the position. Indeed we were invited, for good reasons, to prefer Mr Henry’s evidence to Mr Reid’s on any question as to what the eligible land in the field comprised. That was because there were conflicts between the two witnesses as to how, precisely, the 13 ha they had found to be eligible was made up. These conflicts were as described by Dr Banks in his submissions. Had Mr Henry’s conclusion, as the principal inspecting officer (although not the more senior of the two) been reliable we would not have been inclined to make very much of these differences and would probably have resolved matters in the way Mr Cameron invited us to, but where the principal conclusion is not itself reliable the effect of these inconsistencies is that we cannot rely on Mr Reid’s evidence as showing that the result arrived at by Mr Henry was correct, notwithstanding Mr Henry’s misapprehension as to the status of marram grass.
 Parties were agreed that identifiable areas of bare sand with no, or minimal, vegetation were not eligible. Dr Banks measured those at 15 ha and Mr Henry at 18 ha. On that matter we prefer Mr Henry’s evidence. Dr Banks frankly admitted that the exercise he had carried out was rough and ready and relatively unsophisticated, in that he did not have available to him the full panoply of technology available to Mr Henry. No criticism attaches to Dr Banks for that. He did what he could and his underlying methodology was consistent with Mr Henry’s. But given the fact that Mr Henry measured these areas making use of more sophisticated equipment, his conclusion on balance has to be preferred to that of Dr Banks.
 The result of all of that is that the respondents’ decision has to be set aside save to the extent of 18 ha of the field being ineligible.
 The appellant’s position is that the whole of the rest of the field is eligible. We do not feel able to make a positive finding to that effect on the basis of Dr Banks’ methodology and the limited resources available to him. In our Note of 7 March 2011 we said (at para ) that at proof the onus would be on the appellants (in that Note we spoke of Foveran Farms, rather than Mr Davidson, as being the appellants) “to prove their case”. That has been done to the extent of showing that the decision of the respondents cannot stand. We are not satisfied that the appellant requires to go beyond that and prove that his own position is correct. That is because penalties simply cannot be allowed to stand when the basis for them is vitiated. Had there been some other compelling evidence in the case to show that, notwithstanding his misapprehension as to the status of marram grass, Mr Henry’s conclusion was nevertheless correct – for example if it were shown that there are no areas of the field in which marram grass on its own is present in sufficient density to be eligible – the position would be different. But where the decision is itself vitiated and that is not cured by other evidence the penalty imposed cannot stand.
 Regulation 9(2) of the Appeal Regulations provide that in determining an appeal the Court may (a) confirm the decision of the Scottish Ministers, (b) amend or alter that decision in any respect which it considers appropriate, or (c) substitute for that decision any decision which it considers appropriate. There is no express power to remit to the Scottish Ministers to have the exercise on which their decision was based re-done but even if there were we would not consider that appropriate: we have doubts as to whether a June 2011 inspection, let us say, can simply be substituted for a December 2009 one for the purposes of the appellant’s 2009 SFP and LFASS claims.
 Accordingly, what we have done is set aside the respondents’ decision in so far as it goes beyond a finding that 18 ha of this field were ineligible for aid purposes and remitted the matter to the respondents to calculate the applicable penalties on that basis.
 Before leaving the case we would like to say a number of things. Firstly, we want to put it on record that, despite his misunderstanding of the position in relation to the eligibility of marram grass, we intend no criticism of Mr Henry. Dr Banks himself spoke highly of him and, having seen and heard him as a witness, we share that view. He seemed to us to be a good, conscientious and professional officer who was applying the rules of the relevant schemes as he understood them. He was a good and credible witness. So far as his misunderstanding is concerned, as we have said above, he was not alone in that. The true position seems to have been realised by the respondents only in the course of this appeal. He should regard himself as unimpugned by anything we have said in this Note. The same is true of Mr Reid.
 Secondly, we endorse Dr Banks’ criticisms of the respondents as to their delay in setting out fully and clearly the basis of their inspectors’ decision. It would have been helpful to Dr Banks and to the Court had that been done and had the productions upon which they intended to rely been produced at an earlier stage. Their failure to do so left them open to risk of suspicion that some sort of retrospective justification of the decision was being sought. We are sure that that is not the case but the respondents’ own interests, as well as everyone else’s, would have been better served by earlier and fuller disclosure of their position. The respondents came to proof on little more than an averment that “the density of marram plants on the area in dispute … was not in a sufficient density to permit the area to meet the statutory definition of eligible land” (Answers, Page 6, paragraph B) and that the area “did not meet the qualifying standard to be considered permanent pasture on the basis of a technical assessment of the areas forming part of the field” (Respondents’ Note in response to our Order of 17 February, paragraph 8). This was not helpful to the appellant or to the Court.
 We recognise that this approach may have been thought to be justified on the basis of Mr Cameron’s submission that the decision of the inspectors should be treated by the Court as an administrative decision only to be set aside if demonstrably wrong. We did not hear full submissions on that point and the question of the extent of permanent pasture came to be treated as a matter of fact for the Court to determine on the evidence. We expected to hear evidence and submissions focussed on what density of vegetative cover is necessary to constitute permanent pasture but, for the reasons given, we have been able to dispose of the case without detailed consideration of that matter. Although we accept that there will always be a degree of subjectivity in such assessments, future disputes of this kind might be avoided by provision of further guidance on the relevant practical factors which fall to be considered in those assessments.
 Finally, it has not been necessary for us to deal with Mr Cameron’s submission that the decision made by the respondents in this case involved an exercise of discretion by a public authority, but we are not to be taken as agreeing with it. We leave this issue over for fuller submissions if it is thought to be critical in another case. At present we see no very obvious support for it in the relevant legislative material. Because exact measurement of the eligible, or ineligible, areas is, in practical terms, impossible there will be a range of results none of which is demonstrably wrong. Conflicts between bodies of reliable evidence may fall to be determined by reference to the concept of onus rather than to that of the scope of administrative discretion. But we express no concluded view.
 Although Dr Banks did address us on the question of expenses, we have followed our usual practice of inviting written motions and submissions on expenses.
For the appellant: Dr Gerald P Banks
For the respondents: Donald Cameron, Advocate; Scottish Government Legal Directorate