(Lord McGhie, Mr D J Houston)
(Application RN SLC 164/10 – Order of 10 Aug 2011)
AGRICULTURE – APPEAL – SINGLE FARM PAYMENT – ENTITLEMENTS – SHORTFALL IN AREA – METHODS OF MEASUREMENT – TWO-DIMENSIONAL PLANS – SURFACE AREA – GPS – MIDAS SHEETS – TECHNICAL ASSESSMENT – AERIAL PHOTOGRAPHS – GAEC – GOOD AGRICULTURAL AND ENVIRONMENTAL CONDITION – PERMANENT PASTURE – IMPENETRABLE GORSE – WATERCOURSES AND DITCHES – COMMISSION REGULATION (EC) NO 796/2004 – COUNCIL REGULATION (EC) NO 73/2009 – RURAL PAYMENTS (APPEALS) (SCOTLAND) REGULATIONS 2009
The appellant appealed against a decision of the Scottish Ministers finding that he had failed to hold sufficient land meeting the criteria for “permanent pasture” to cover the entitlements to single farm payment he held and claimed payment on in 2009. Much of the dispute related to the proper assessment of areas of thick gorse but various issues of measurement arose.
HELD (1) that it was proper to measure on a two-dimensional plan basis; (2) that the proper approach was to determine whether the land was properly described as “permanent pasture” rather whether the gorse was “impenetrable”; (3) that a discriminating approach was required to assessment of watercourses rather than use of an average over the whole length within a particular parcel; (4) that where a change of boundary led to reduction in area of one parcel, it was proper to add the land to the enlarged parcel rather than ignore it as falling within a tolerance; and (5) the total area of eligible land found had to be adjusted. It was short of the claimed entitlements but that shortfall was less than 2 ha and accordingly no penalty fell to be imposed. OBSERVATIONS were made about the technical detail of the GPS measuring system and about whether the need to comply with “GAEC” – “good agricultural and environmental condition” – required land to be in both good agricultural and good environmental condition but the Court found it unnecessary to take a decision on the point.
The Note appended to the Court’s order is as follows:
 Thomas J Brown appealed against the decision of the Scottish Ministers conveyed in their letter dated 25 August 2010. The key issue is one of measurement or assessment of area and relates to the question of whether the appellant held sufficient land meeting the criteria for “permanent pasture”, to cover the entitlements to single farm payment he held and claimed payment on in 2009.
 We heard evidence on 8 April and 23 May and evidence and submissions on 21 June, all 2011. Dr Banks, an Agricultural Consultant, was permitted to appear on behalf of the appellant. He was the only witness for the appellant. For the respondents, Mr Donald Cameron, advocate, appeared and led evidence from Ms Lynne Todd, Agricultural Officer and Mr John Reid, Senior Agricultural Officer, both employed by the Scottish Ministers. At the end of the second day we had to adjourn before cross examination of Ms Todd was concluded. Due to illness, she was unable to return on the third day fixed for hearing and the parties agreed to proceed without further evidence from her. We understood that it had been expected that Mr Reid would be able to cover some matters left unresolved by her evidence. In the event, it was not clear that he had been able to do so. Our decision is based on the evidence which was led.
Commission Regulation (EC) No 796/2004
Council Regulation (EC) No 73/2009
 No relevant decisions were cited and it was accepted that there were no relevant domestic regulations. We had reference to various publications by Scottish Ministers giving guidance on aspects of the application of the European Regulations but it was not suggested that any of this material was binding in the present context. It was not disputed that our decision should turn on the proper application of the European Regulations and not on a construction of guidance material provided by Scottish Ministers.
 In May 2009 the applicant duly submitted a Scottish 2009 Single Application Form (“SAF”). This included an application for Single Farm Payment (“SFP”). He declared and claimed 254.33 entitlements on a total area of 255.34 ha which was said to meet the eligibility criteria. The land had been inspected in 2006. At that time, some ineligible areas had been found and some mapping inconsistencies had been corrected. His 2009 claim was consistent with the areas of permanent pasture accepted in 2006.
 Prior to 2009 there was little guidance as to what might be accepted as permanent pasture. Paragraph 182 of the Cross Compliance Notes for Guidance published in December 2005 provided: “You must avoid the encroachment of unwanted vegetation which degrades the agricultural and environmental value of the land to the extent that the land is not capable of returning to agricultural production by any time during the growing season in the following calendar year”. However the present case does not relate to issues of cross compliance and there was no other guidance relevant to the impact of gorse, broom or other unwanted vegetation on eligibility of permanent pasture.
 In 2009 the Notes for Completion IACS (1a) 2009 stated, at page 6: “There may be more areas within the parcel which are ineligible and you must ensure that you deduct any ineligible land (original emphasis) when you claim. … Examples of ineligible areas include rock, scree, ponds, watercourses, forestry, buildings, hard stands, roads, impenetrable gorse and stands of only bracken”.
 “Spot-check” inspection of various parcels - in other words, fields - included in the claim was carried out by Ms Todd on behalf of Scottish Ministers on 17 and 18 November 2009 at one of the appellant’s holdings, Little Dens, and on 25 November at Craibstone. On the last date she was accompanied by James Daniel an experienced colleague of similar rank. When making her assessments of eligible and ineligible areas covered by gorse at Craibstone she did so in discussion with Mr Daniel. Ms Todd advised Dr Banks of her findings. The findings were that the claimed areas of certain parcels, totalling 221.92 ha should be reduced by 3.06 ha. The original area claimed had included a surplus over entitlement of 1.01 ha. The total found to be eligible was, accordingly, 2.05 ha short of the claimed entitlement. Applying the provisions of Arts 50 and 51 of Reg 796/2004, this meant not only that the applicant would not be paid on his full entitlement but that he would face a penalty of double the shortfall. Put shortly, the effect of Art 51 is that a shortfall of under 3% does not attract a penalty unless it is in excess of 2 ha. We note, by way of comment only, that while this provision can be said to be “proportionate” in the strict sense that as a farm gets bigger the permitted margin of error is progressively smaller, it is not self evident that such effect is proportionate to the aim of the provision. However, there was no challenge to the provisions as such. Dr Banks made it clear from the outset that he disputed only that any shortfall exceeded 2 ha. In other words the dispute was triggered by the excess 0.05 ha or an alleged discrepancy of 0.02% of the total claimed. The inspector’s findings were sent formally to the appellant in an initial decision dated 18 January 2010.
 An application for review was then made. On 16 February 2010 a second Inspection Report was produced by Ms Todd. This followed a review of the evidence carried out as a desk exercise with the assistance of the equipment used at inspection as described below. This review led to a reduction of the figure of eligible pasture land found to 252.07 ha and, accordingly, the relevant shortfall was then said to be 2.26 ha. Discussion and exchange of correspondence between Dr Banks and officials continued thereafter. Much of Dr Banks’ concern, stressed in meetings with officials, related to the deduction of 1 ha of gorse in each of fields 5 and 8/9, discussed further below. It was contended that there had been no significant change in the nature of the gorse since 2006. On 5 August 2010, Jenni Kinnaird, a Principal Agricultural Officer and John Reid, then a Senior Agricultural Officer based at Thainstone, inspected the gorse in these two fields. They concluded that the area of ineligible gorse in field 5 was 0.91 ha rather than 1.0 ha and in field 8/9 was 1.07 ha. Their inspection was restricted to assessment of gorse in these fields and they did not make any assessment of other parcels or of other matters relied on by Ms Todd in making her assessment.
 The formal Review Meeting Report was issued on 25 August 2010. It included the narrative:
“The appellant holds 254.33 entitlements. Area found comes to 252.07 ha so 2.26 ha difference. The area eligible is then 252.07 ha - 4.52 ha and 247.55 ha eligible for payment.”
In relation to the gorse it provided the further explanation:
“Mr John Reid and I visited both fields on 5/8/2010. In NK/07322/44460 we found, by assessment, 1.07 ha ineligible which given the time since the initial inspection and the potential growth dynamics of the site is being left at the 1 ha found. For NJ/49242/59298, 0.91 ha was found in comparison to 1 ha. Given there was no evidence of control the area has been updated to 0.91 ha as it would not have contract(ed) in the intervening timescale unless mechanical control had been done.”
 The appeal was nominally in respect of the decision in that report and the respondents’ pleadings appeared to be based on it. However, despite its date, the report purported to reflect matters discussed on 30 June 2010. In light of the inspection on 5 August, Ms Kinnaird had written to the appellant on 23 August 2010, reporting the details of the field inspection findings and setting out in a table as an annex; the eligible areas claimed in each parcel, the areas found by the inspectors in relation to each parcel, and what were described as the “tolerated” areas in respect of each. For reasons which were not clear to us, parcels totalling 3.56 ha were omitted from the lists in the detailed table but it was not disputed that allowance had been made for such parcels in all the relevant totals. This report showed the area “found” to be 252.16 ha. This change, from the 252.07 ha in the Review Meeting Report, reflected the revision of ineligible gorse in field 5 from 1.0 ha to 0.91 ha. The schedule prepared before the hearing showed the area accepted by the respondents as 252.16 ha and the hearing was conducted on the basis that the dispute was over that figure.
 We return to the detail of the findings in relation to individual fields on a field by field basis below. It is sufficient, at this point, simply to note that the effect of the overall review process was to increase the apparent shortfall from 2.05 to 2.17 ha and apply a penalty which had the effect of deducting double that figure when calculating the entitlements allowed.
 Prior to the hearing we asked parties to attempt to prepare a schedule setting out their detailed contentions in relation to each relevant field or parcel of land. In their response, the Scottish Ministers said that the issues were all questions of fact. As will be seen, we think that the main issues are factual but there are some issues of law or principle which have a general bearing and it is convenient to discuss them first.
 We deal separately below with Regulations relating to the issue of measurement. At this point we consider the provisions relating to claims, like the present, based on entitlements to single farm payment held.
 Article 34 of Reg 73/2009 provides:
“1. Support under the single payment scheme shall be granted to farmers upon activation of a payment entitlement per eligible hectare. Activated payment entitlements shall give a right to the payment of the amounts fixed therein.
2. For the purposes of this Title, ‘eligible hectare’ shall mean:
(a) any agricultural area of the holding, and any area planted with short rotation coppice (CN code ex 0602 90 41) that is used for an agricultural activity or, where the area is used as well for non-agricultural activities, predominantly used for agricultural activities …”.
 It may be noted that although we expressed a doubt as to construction of this provision and Mr Cameron acknowledged a possible ambiguity, he said that in the context of the present case no relevant doubt arose. Dr Banks did not make any comment.
 Article 2 of that Regulation provides certain definitions including:
“(h) ‘agricultural area’ means any area taken up by arable land, permanent pasture or permanent crops”.
 The effect of these provisions in their application to the present case was that for each entitlement claimed, the appellant had to hold one hectare of permanent pasture.
 It may be noted that Article 2(c) defines ‘agricultural activity’ to mean “the production, rearing or growing of agricultural products including harvesting, milking, breeding animals and keeping animals for farming purposes, or maintaining the land in good agricultural and environmental condition as established in Article 6 ”. It was agreed that the latter phrase was the relevant part of the definition for the purposes of the present case. Article 6 places obligations on Member States to ensure “all agricultural land especially land which is no longer used for agricultural purposes, is maintained in good agricultural and environmental condition”. It provides that “Member States shall define, at national or regional level, the minimum requirements for good agricultural and environmental condition …” The duty under this provision rests on Scottish Ministers. They might have been expected to implement it by use of domestic Regulations. However, we do not think anything turns on this in the present case. We discuss below the relationship of good agricultural and good environmental condition.
 The penalty provisions are not in dispute although Dr Banks appeared at one stage to suggest that he understood them to have the effect of tripling the shortfall in applying the deduction. The main issue in the case is how the actual area of “permanent pasture” is to be determined. Article 2(2) of Reg 796/2004 contains various pertinent definitions. For present purposes the most important is the definition of “Permanent pasture” which is “land used to grow grasses or other herbaceous forage naturally (self-seeded) or through cultivation (sown) and that is not included in the crop rotation of the holding for five years or longer”.
 Article 2(15) provides that “ ‘use’ shall mean the use of area in terms of the type of crop or ground cover or absence of a crop”
 It may be noted that there is a definition at (22) of “area determined”. That is to mean “the area for which all conditions laid down in the rules for granting the aid have been met; in the case of the single payment scheme, the area declared may be deemed as being determined only if it is actually accompanied by a corresponding number of payment entitlements.” However, nothing turns on that in the present case.
 Art 50(2) of 796/2004 provides that “if there is a discrepancy between the payment entitlements declared and the area declared, payment shall be based on the lower size.”
 Although we heard no submission about the effect of this, we note that the number of entitlements held by a producer is expressed as a total and is not based on any identifiable individual parcels of land. Accordingly, when the claim falls to be assessed by reference to entitlements claimed - as an alternative to areas declared - we are not persuaded that the producer is limited by reference to what he declared on a field by field basis. On the face of it, the question is simply whether the total area found supported the entitlements. It would seem to follow that provisions about tolerance - which relate to discrepancies between areas claimed and areas found - have no bearing when the claim falls to be assessed by reference to entitlements. The number of entitlements is an issue of fact. The assessment of area to back that total number, might also be expected to be a question of fact unconstrained by the claimed measurements.
 In the event, we did not consider that the question of the respondents’ approach to tolerances arose sharply in this case and it is unnecessary to express a concluded view. We touch on the matter when dealing with Field 6, below, but, as will be seen, that related to a situation where land was clearly identified as having moved from one field to another rather than as an issue of inaccurate measurement. For interest it may be noted that Article 30(1) provides for determination of “parcel areas” and allows measurement tolerances. It was not contended that any of this was directly relevant to the issues in this case and we simply observe that in relation to individual parcels a tolerance of 5% and a maximum of 1.0 ha is allowed. This contrasts with the allowance calculated by reference to totals. As we have seen, the initial dispute in the present case arose from an alleged discrepancy of about 0.02% of the appellant’s total entitlements. That figure increased slightly after further investigation but it was a discrepancy of that order which triggered the whole appeal procedure.
 A question arose in relation to the watercourse in Field 5. Art 30(2) provides that “The total area of the parcel may be taken into account provided it is fully utilized in accordance with the customary standards of the Member State or region concerned. In other cases the area actually utilized shall be taken into account.
In respect of the regions where certain features [such as ditches] are traditionally part of good agricultural cropping or utilisation practices the Member States may [accept them as part of pasture] on condition that it does not exceed a total width to be determined by the Member State. That width must correspond to a traditional width in the region in question and shall not exceed 2 metres.”
 The Regulation appears to provide that it is simply left to the Member State to decide what is to be accepted. However, we cannot regard that as being equivalent to leaving matters to the decision of individual inspectors. There was no suggestion that Scottish Ministers had determined to deal with the issue in that way and no suggestion that they had simply decided to exclude watercourses in their entirety. Mr Cameron appeared to accept that it was necessary to determine, as a matter of fact, whether a feature exceeded 2 metres. As will be seen, that raises the question of whether it is necessary to view the watercourse as a whole. That appears to turn on the effect of the Regulation. We are satisfied, as discussed below, that the Regulation allows a discriminating approach
 It is clear that payment is dependant on “eligible hectares”. As we have seen, this is defined in Art 34(2)(a) of Reg 73/2009. It should be noted that in his final submission in response, Dr Banks made a tentative reference to Art 34(2)(b) which provides for “any area which gave a right to payments under the single payment scheme or the single area payment scheme in 2008 and which: (i) no longer complies with the definition of ‘eligible’ as a result of the implementation of [certain specified directives relating to wild birds and natural habits and of wild flora and fauna]”. However, there was no attempt to establish positively that any relevant area which had “given right to payment” in 2008 had changed by 2009 so as to have ceased to comply, nor that such change as there might have been was attributable to any of the specified directives. In the circumstances, we have taken the view that this provision cannot be considered for the purposes of our determination of the present appeal.
 We deal further below with the question of whether measurement for the purposes of the Regulations should have regard to surface area or to the two dimension system used for plans and maps. However, it is convenient first to give a brief outline of our, comparatively limited, understanding of the technical detail of the respondents’ methods of use of the GPS system. Although we heard a good deal about this, neither party had initially lodged or attempted to lodge the equipment the inspectors used. On day two, in course of Ms Todd’s evidence, there was an attempt to do so. We were reluctant to risk the delay which might have been involved at that stage in attempting to deal with the equipment as a production and in getting a full formal description of what it was and how it worked. It was not thought that a complete understanding of the technical detail was directly relevant in relation to the matters in dispute. We allowed Ms Todd to use the “kit” to answer certain questions relating to calculation and to demonstrate aspects of how it worked. We shall attempt to summarise the more relevant aspects.
 The “kit” is a portable computer which can be linked to a GPS system. It can calculate lengths and determine areas. The process of determining figures by use of the computer is referred to as “interrogating the kit”. This can be done by reference to OS maps loaded onto the computer or by reference to the physical position of the computer itself. At inspection the inspectors will take steps to satisfy themselves that the OS map pre-loaded onto the computer appears accurately to reflect what is seen on the ground. For example, the inspector might position herself, or himself, at an identifiable point in a field, such as a corner, and check that the reading given by reference to the OS map reflects the reading given as to the physical position of the computer when it is at that corner. She might then check that the lines of fencing from that point accorded with the directions shown by the map on the computer. If so, the OS map would be accepted as accurate on these matters. The nature of further detailed checks of the map against the “real world” position of the computer would depend on the nature of the topography. If a dyke or fence appeared to be accurately mapped, the inspector would not see a need for repeated checking against the computer. Much more checking might be required in some circumstances.
 Areas bounded by features identifiable on the map, such as a fence, could be determined by interrogating the kit which would provide a figure for the area shown on the map as being enclosed by such fence. Where areas had no complete boundary the inspector could delineate the boundary with a pointer on the map and the kit would be able to provide a figure for the area thus defined. An alternative would be for the inspector to walk round any area asking the computer to use its GPS facility to track the boundary and calculate the area so enclosed. But inspectors might also assess small or irregular areas by judgment. If for example, they thought a number of small areas had to be excluded for any reason they might do so on the basis of their own assessment. They would note the figure they assessed for each in a notebook and then register a total for the exclusion in the computer and mark the screen with an appropriate icon.
 The icons as they appeared on the printouts had broad labels and did not set out any detail as to what they showed. However, the kit could be interrogated to disclose the detailed information underlying the icon. The printouts were known as “Midas sheets” and we understand “Midas GIS” to be a trademark name for the software provided by Midas Technologies Private Ltd and used to create them.
 We record, for completeness, that there was some apparent confusion as to the significance of yellow symbols labelled “GPS” on the Midas sheets. As a result of his understanding of evidence given in a previous case, Dr Banks had an impression that any attempt by an operator to use the kit to determine a position would be shown on the sheet by a yellow icon. His cross examination was confusing in parts because of this. We had no difficulty in accepting Ms Todd’s evidence that this was not so. Where she had used the kit to determine a precise location this was shown on the sheet by a small black unlabelled square. Typically there would be many such squares on the sheet and only one or two yellow icons. The latter were simply to record that she had inspected the field in question at a particular time and the kit would be able to show details relating to that icon including detail of where she was when the icon was logged.
 To help put the dispute in context, Dr Banks made the point that the kit used by Scottish Ministers only purported to be accurate in measuring lengths to within a metre. More accurate equipment was readily available but was more expensive. We were referred to reports of validation tests of the equipment. These showed that different operators produced different figures even in a formal, regular shaped, test setting. The range of error was small but of a scale which, in relation to the claimed area as a whole, could have accounted for the whole discrepancy in the present case. In the event, however, this was not said to be relevant. Most parcels were agreed and the dispute related to the matters dealt with field by field below.
 The main dispute over measurement techniques related to measurement or assessment of areas populated by ineligible gorse. We shall return to consider the detail of evidence bearing on the gorse in fields 5 and 8/9, but it is appropriate first to look at the wider issues underlying this dispute.
 The appellant required to have eligible hectares to match his entitlements. In the context of his particular farming operations in Scotland this meant hectares properly described as permanent pasture. It was not disputed that areas which might, in the past, have been accepted as permanent pasture had included ineligible features where they lay within areas of unmistakeable pasture land. Most areas of gorse had simply been accepted as part of the pasture, possibly on the basis that animals would use gorse areas for shelter even where there was little underlying grazeable vegetation. However, as the exclusion of roads or areas of hard standing plainly indicates, the question is not one of use by stock but of proper description as pasture. Stock may well use buildings, areas of rock or hard standing or ponds. That does not mean that such items can properly be described as permanent pasture.
 Dr Banks made much of the difficulty he faced in understanding what was meant by “dense” or “impenetrable” gorse. He suggested that guidance on such terms was “obtuse” and “unfathomable”. He said that some more effective guidance was needed for both Scottish Government staff and farmers. However, Ms Todd and Mr Reid explained that in recent times Scottish Government staff had been given more guidance as to what to do about thick gorse. The problem had been addressed by a process of instruction and test exercises. Ms Todd had been trained by instruction and by having to make assessments under the supervision of a more senior colleague who would then review her findings and discuss them with her. Inspectors would discuss their assessments with each other. She thought this led to a fairly consistent standard among staff although assessments would vary.
 Inspectors had to make an assessment as to whether areas of gorse could properly be regarded as pasture, having regard to the amount of herbaceous foliage growing beneath and whether or not such foliage could be readily accessed by stock. It seemed plain that their assessments gave most weight to availability of such foliage. They took the view that if it was there, stock would probably get to it unless it was in a really dense clump. Where there was a thick stand of gorse with parts regarded as dense or impenetrable, it was not possible, in practical terms, to measure accurately how much could be taken as eligible. Inspectors would determine the overall area of the stand either by interrogation of their kit or, possibly, by estimation, and then use their training and experience to make a “technical assessment”. This would be assessed in terms of an ineligible percentage of the measured area and then expressed in terms of a specific area being deducted as ineligible. Where there were several clumps of thick gorse within a single parcel an inspector might make such an assessment of each, noting the figures in a notebook and then deducting a single total figure from the area of the overall parcel. Ms Todd freely accepted that a technical assessment could only identify a range. It could easily be 10% on either side of the figure selected. The use of the term “technical assessment” might convey an impression of accuracy but it was no more, and no less, than an exercise of professional judgment.
 Dr Banks was eloquent in support of his contention that his professional assessment of ineligible gorse was to be preferred to that of the respondents’ witnesses. He stressed his experience. He pointed out that they had provided little or no supporting evidence. They had provided very few photographs. He contended that such photographic evidence as they did provide, in particular the aerial photographs, strongly supported his assessments. He, himself, provided a large number of photographs. We accept his evidence that they had been taken at fairly regular intervals in the course of his walking round the subjects to present a full picture and not a selective one. He suggested that they strongly supported his assessment.
 However, in making our assessment of the weight to be given to his evidence we think it relevant to have in mind that his approach started on the back foot, so to speak. He made much of the fact that the appellant’s application was based on the figures accepted in 2006. He drew attention to the “Notes for Guidance IACS (1) 2009 ” issued by the respondents to assist farmers when completing their 2009 form. He made the point that this document did not draw specific attention to the gorse. However, in cross-examination he had to accept that farmers had also been provided with “Notes for Completion (1a) 2009 ” which contained the explicit reference to gorse quoted above. We took Dr Banks’ evidence on this issue to be equivalent to an assertion that whoever had completed the 2009 SAF on behalf of the appellant had not considered the content of the latter guidance. In short, the appellant had approached his 2009 claim without making any assessment of the eligibility or otherwise of the gorse. The need for a precise distinction has arisen because completion of the application without due attention to the guidance has given rise to a potential over-claim. Dr Banks has had to set himself the task of trying to persuade first the inspectors, and then the Court, that much of the gorse was eligible. He did so by concentrating on the question of whether the gorse was “impenetrable”. But he professed himself incapable of understanding what was meant by this term. Indeed, he asked rhetorically whether this was supposed to be impenetrable to him as a man recovering from a hip operation or to “a cow, a sheep, a pig, a rabbit, a badger or a chicken?” In these circumstances we could attach no particular weight to his evidence in assessing whether particular stands of gorse were, or were not, to be accepted as permanent pasture. He was critical of the respondents’ failure to provide a better description of the test but plainly he was unable, himself, to do so.
 Dr Banks’s own inability to suggest any more precise verbal test tends to support the approach taken by the Ministers of relying on training and experience of working together rather than attempting to refine the concept further in words. Broadly speaking we think it clear that a thick stand of gorse would not normally be described as permanent pasture, even if it was possible to struggle through it on foot. On the other hand we accept that the presence of small clumps of gorse does not prevent grassland areas from being viewed as pasture. Clearly there is a need to distinguish the two categories. But we think that it is misleading to see the point of distinction as dependant on a fine line between what is penetrable and what is not. The essential question is the broader one. Is the land properly to be described as permanent pasture? The Ministers have attempted to give guidance to their staff indicating that where there is edible foliage under gorse they will tend to accept that as pasture unless it is hard to get at. We accept that approach as reasonable and we have followed it in making our assessment in the present case. But we think that a farmer trying to measure his pasture for the purpose of completing a claim form would be better advised to take a common sense view of what is pasture rather than attempt a fine assessment of whether gorse is “dense” and “impenetrable”. Although these terms do provide a clear enough definition of what is clearly not pasture we do not think that penetrability in itself provides a test of whether land is pasture. The dominant consideration must be a realistic assessment of whether the land is used to grow grasses or other herbaceous forage: see Article 2(2) of Reg 796/2004, above.
 Further, Dr Banks’s photographic evidence was of conditions in the spring of 2011. He had visited in June 2010. He had tried to take photographs but his batteries had failed. He thought that there had been little difference in the state of the gorse between the visits. But there had been two hard winters by the time the photographs were taken. He accepted that some gorse might have been killed in the winters but thought that the overall area under gorse would not have changed in this time. Although we heard a good deal of evidence about the development of gorse and readily accept that, in absence of positive measures of control, the overall area of a stand of gorse would tend to increase rather than shrink, we are not persuaded that it will necessarily maintain the same density over the years. There is no doubt that older gorse can become tall and straggly. Indeed, Dr Banks spoke to the difficulty which the appellant had had in his recent attempts to burn off the gorse at field 8/9. Some had become too straggly to burn effectively. We do not know when it reached that state. Our concern is with the state of things in 2009. The evidence of Ms Todd as to her findings in November 2009 is likely to be more reliable than the evidence of what was found in 2011.
 We consider that, in broad terms, the evidence of the respondents’ witnesses on this matter is to be preferred. We are satisfied that they were attempting to address the question of whether particular areas were eligible as permanent pasture whereas Dr Banks was trying primarily to determine whether particular areas of gorse were properly to be described as impenetrable. His ability to carry out that exercise was not assisted by his professed inability to understand the meaning of that word as applied in the present context. We consider that the approach of placing emphasis on the presence of herbaceous foliage was sound. We accept that photographs can be misleading and we note that Dr Banks’ photographs all appear to be taken from the periphery of the gorse areas and not in the centre of such areas.
 We think it important that the inspectors were working to try to achieve a consistent standard. The approach taken of identifying an overall stand or section of gorse and determining first its area with a fair degree of precision by interrogating the kit and then making a “technical assessment” of the percentage to be allowed as eligible provides a sensible, practical approach to calculation of eligible area. Both farmer and inspector must take a practical approach and if a stand of gorse is claimed as part of the permanent pasture the farmer can be expected to err on the side of caution. It is obvious that no precise measure of eligibility can in practice be achieved and a fair assessment made by trained inspectors can normally be expected to provide a workable way of dealing with the matter.
 Our broad approval of the Scottish Ministers’ approach is, of course, subject to evidence of how well any particular assessment was made. We shall return to look at such details on a field by field basis. It will also be important to keep firmly in mind that the process of making a technical assessment does not produce a definitively “correct” figure. It is entirely reasonable, in the context of the exercise which Ms Todd was carrying out, to recognise that there is a band of possible figures and to select a figure in the middle of the band. On balance the mid figure is more likely to be correct than the figures at each end of the band. But, where a particular figure, chosen from a reasonable band, is the basis for a penalty, other issues may arise. If the claimant asserts a figure at the low end of the inspector’s assessed range it might be quite proper to accept that figure on the basis that the inspector cannot confidently say it is wrong. In the event, however, our decision in the present case has not required to depend on that type of issue and it is unnecessary to say more.
 It was not disputed that much of the relevant land was steep and that the actual surface area accordingly exceeded the area as measured on a two dimensional (“2-D”) system - such as by reference to a map or photographic image. Although the Scottish Ministers did not explicitly agree the relevant figures, we understood it to be accepted that if surface areas were to be used, the total area found would have been sufficient to cover the claimed entitlements. There would have been no shortfall or, at least, no penalty.
 For the appellant it was argued that “area” meant surface area. It was accepted that the EC Regulations referred to geodetic systems which were two dimensional but it was contended that these references related only to identification, as opposed to measurement, of parcels of land. It was pointed out that the equipment used – “the kit” – was able to measure heights. It was also pointed out that the OS maps included contours. It appeared to be contended that calculation of surface area would be possible using the formulae of Pythagoras but there was no evidence that any equipment was presently available to allow accurate measurement of surface area of farms on a national scale.
 For the respondents it was contended that in terms of the European Regulations, the Scottish Ministers were obliged to use a national geodetic system: Reg 796/2004 Art 6. The OS system met the criteria of Art 2(28). Both the OS and GPS system were two dimensional. They met the requirements and produced consistent and comparable figures.
 Before we examine the detail of the regulations it is worth adding that, in course of his evidence, Dr Banks accepted that, in agricultural terms, a sloping surface was not directly comparable to a level surface: plants tended to grow upwards rather than at right angles to the land. We think it important to keep this in mind. Although it might seem that, without further definition, “area” should be taken to mean the area of the surface, words must be understood in context and we do not accept that, in the present context, there is any “commonsense” approach to construction. The matter must be resolved by considering use of the references to area in the whole context of the Regulation.
 Article 6 of Reg 796/2004 provides:
“1. The identification system for agricultural parcels … shall operate at a reference parcel level such as cadastral parcel, or production block which shall ensure unique identification of each reference parcel.
Member States shall, moreover, ensure that agricultural parcels are reliably identified and shall in particular require the single application to be furnished with particulars or accompanied by documents specified by the competent authority that enable each agricultural parcel to be located and measured. The GIS shall operate on the basis of a national geodetic system.”
 In terms of art 2(28): “ ‘National geodetic system’: shall mean a coordinate reference system which permits standardised measurement and unique identification of agricultural parcels throughout the Member State concerned; where different coordinate systems are used, they shall be compatible within each Member State;”
 We have no doubt that, on a proper construction of these provisions, the Scottish Ministers are correct to contend that a 2-D system is envisaged for the purposes not only of identification but of locating and measuring. The concept of a geodetic system is one operating on a global scale. While it allows for the curvature of the earth it plainly does not envisage measuring and calculation of the surface area of every hump and howe. A purposive approach is required. We are satisfied that the concept of consistency and a standard approach to producers throughout Europe is fundamental to the operation of the whole system. That can be achieved by a 2-D, map based, approach. It would be extremely difficult to operate a system of surface assessment on that scale.
 This view of the two provisions relied on by Mr Cameron is supported by consideration of other provisions of the same Regulation. For example, paragraph (37) of the Preamble shows that one aim of the substantive articles is to provide “detailed rules regarding determination of areas and the measurement methods to be used”. We find no mention of measurement methods for parcel areas other than those referred to in the context of identification and in art 2(28). We note references in art 2(25) to geographical information systems and in (27) to graphical material including maps. The draftsmen must have assumed that these various provisions provided sufficient detail as to how area was to be measured. All specific references are clearly to 2-D measuring systems.
 Paragraph (36) narrates: “On-the-spot checks of areas, as a general rule, consist of two parts, the first of which relates to verifications and measurements of declared agricultural parcels on the basis of graphic material, aerial photography and so forth. The second part consists of a physical inspection of the parcels to verify the actual size of the agricultural parcels declared and, depending on the aid scheme in question, the declared crop and its quality. Where necessary, measurement should be carried out. The physical inspection in the field may be carried out on the basis of a sample.”
 The purpose of the second stage is to “verify” the first, not to replace it by a quite different method. This provision is supportive of the view that the Regulation proceeds on the basis of an understanding that a 2-D system of measurement will be used.
 Article 32(5) makes provision for remote sensing as an alternative method of “on-the-spot” checking:
“Where a Member State applies the possibility to carry out on-the-spot checks by remote sensing, it shall:
(a) perform photo interpretation of satellite images or aerial photographs of all agricultural parcels per application to be checked in accordance with paragraph 4 with a view to recognising the ground cover and measuring the area:
(b) carry out physical inspections in the field of all agricultural parcels for which photo interpretation does not make it possible to verify the accuracy of the declaration to the satisfaction of the competent authority.”
A similar provision can be found in art 20(2) of Reg 73/2009. It is plain that the specified methods of remote checking are based on a 2-D system.
 The parties did not address us on the scope and intent of the word “identification” in relation to agricultural parcels. However, for completeness we simply note that in the context of the Regulations, the concept appears to encompass more than merely showing where each labelled parcel is to be found. Article 17 of 73/2009 provides:
“Identification system for agricultural parcels
The identification system for agricultural parcels shall be established on the basis of maps or land registry documents or other cartographic references. Use shall be made of computerised geographical information system techniques, including preferably aerial or spatial orthoimagery, with a homogenous standard guaranteeing accuracy at least equivalent to cartography at a scale of 1:10000.”
 In so far as area is viewed as an aspect of identification there is little doubt that the two Regulations each envisage use of a 2-D system.
 We heard conflicting submissions as to the proper approach to land which was being managed under an environmental scheme. The Scottish Ministers’ position can be simply stated. They contended that the conditions for entitlement to grant under such schemes were quite distinct from the conditions for SFP. The fact that an area of land was being properly managed under a Rural Stewardship Scheme (“RSS”) or Rural Priorities (“RP”) scheme had no bearing on the question of whether it could properly be described as permanent pasture – that being the only disputed condition of the SFP scheme relevant in the circumstances of the present case. Mr Cameron stressed that the term “GAEC” referred to good agricultural and environmental condition. There was no justification for reading it disjunctively. Compliance with what was known as the “GAEC requirements” was an additional requirement for payment of SFP. It did not have any bearing on the primary issue of whether particular land was eligible or not.
 Dr Banks argued that management for approved environmental purposes under an approved scheme was equivalent to keeping the land in good agricultural condition. He made a variety of assertions in evidence and in his submissions relating to environmental schemes and the typical conditions. Some expressly required exclusion of stock. Some required stock to be managed. The basis of payment was, he submitted, that the farmer was being compensated for “income foregone”. It followed, he said, that payment under such schemes assumed that the land was agricultural land. Under GAEC, the farmer had an option of using it for agricultural or environmental purposes.
 Mr Cameron disputed the contention that all environmental schemes were based on the idea of “income foregone”. It, accordingly, became apparent that Dr Banks’ submissions on this matter might come to depend on the precise terms of particular schemes. We had no direct evidence of the schemes which applied in relation to any of the parcels in dispute in the present case.
 As a matter of general principle, it seems to us that the argument based on the concept of “income foregone” assumes that the land in question was initially used productively for agriculture and that something has happened to change it. We heard no evidence that any of the disputed areas had been in any specific condition at any identified prior date. The acceptance of land as permanent pasture in 2006 was based on the standards of assessment used at that time, not on the different state of the land. We think that to be effective in the context of the present case, Dr Banks’s submission must be taken to turn on the proposition that if the land was being managed for environmental purposes, this, in itself, meant that it was entitled to be described as permanent pasture without regard to its condition at any previous date. We are not persuaded that this is a sound proposition.
 On the other hand, we have had some difficulty with the simple submission that “good agricultural and environmental” condition are necessarily to be read as cumulative. One consequence would be that it would not be sufficient simply to maintain the land in good agricultural condition. The farmer would also have to be confident that he met some relevant environmental standard. It might, of course, have been assumed that all land in good agricultural condition would necessarily be in good environmental condition. But, if so, there would be no need for the two to be specified.
 In the event, although this is plainly an important issue we have not been persuaded on the basis of the evidence led, that it is an issue which must be determined for the purposes of the present case. We are satisfied that on any view of the matter, including the concept of “income foregone”, it is necessary to show that the land in question was at some relevant date permanent pasture and that, if it does not currently have that status, it lost it specifically in order to comply with an identified requirement of a specific environmental scheme. We heard no proper evidence of this. It was not disputed that various RSS or RP schemes were in place at the relevant time and payment schedules lodged provided outline detail of schemes applying to certain parcels. But we had insufficient material to allow us to say whether the land used for the schemes had changed character as a result of the schemes.
 Accordingly, we have proceeded on the view that to be eligible in 2009 the land in question required to qualify as permanent pasture at that time. It may be added that it was not suggested that the question of eligibility of land as pasture turned in any way on actual use at the time of the land by stock.
 At the hearing Dr Banks raised an entirely new point. He contended that it could be seen from the overlay of the linear boundary of certain parcels on aerial photographs, that the areas calculated by the respondents were too small. This contention proceeded on an assumption that the photographs accurately showed the real world and that the blue or yellow lines accurately showed the boundaries derived from an OS map. Accordingly, he said, where there was no dispute that a particular boundary ran alongside a watercourse and it could be seen on the photograph that the line was several metres away from that watercourse, this showed that the boundary shown on the map was inaccurate. We understood the contention to be that the line of the watercourse on the photo should be adopted in preference to the blue or yellow line. If that was to be done in relation to field 8/9 it would effectively add a strip of perhaps 6 metres wide down almost the entire length of the boundary and this could easily add something of the order of 0.30 ha. to the area of the field.
 That submission had some attraction. It is within the knowledge of the Court that OS maps in country areas are not necessarily accurate in all respects and it is not difficult to accept that an OS map would not necessarily reflect the current position of a burn running through low-lying ground. It would not be hard to accept that the boundary shown on the photograph and apparently derived from the edge of a stream as shown on an OS map, was indeed inaccurate. However, the argument was also supported by reference to the position of the blue line on the west side of field 17 at Craibstone, compared with the position of the road visible in the photograph of that field. It is plain that the line is well into the field. But we are not satisfied that this can be explained by an assumption that the OS map has the road in the wrong place. That would be much less likely than that the position of a burn had moved. There may be an explanation for the blue lines being wrong relative to the photograph. Ms Todd did not accept the contention that the area measured by reference to the blue lines was necessarily wrong. She seemed to think there was another possible explanation. She said that the lines could simply “slip”. It might be that on more detailed questioning this would not have proved a satisfactory explanation. But, the opportunity for further questioning was forgone and we must proceed on the evidence given.
 We do not know what distortion there might be in a photograph or what technical explanation there might be for the apparent mismatch between the photo and what were said to be OS based lines. Plainly the effect of Ms Todd’s evidence was that she understood there was some sort of explanation. We are anxious not to let technicalities of pleading become a barrier to proper investigation of real issues. We recognise that it was a line of argument which only came to the attention of Dr Banks at a late stage. We might well have been prepared to overlook the late emergence of this point if we had been persuaded that the evidence bearing on it was sound and complete. But in the whole circumstances, when it was an issue not foreshadowed in any way by pleadings or written submissions and Ms Todd did not have an opportunity to explain fully her position, we do not consider it appropriate to make any adjustment to area based on this issue.
 This is a long narrow field sloping down to a watercourse. There are various areas of gorse. The branches of an adjacent plantation of Sitka extend over a substantial length of the south boundary at the west end. It was not disputed that the total area lying within the boundaries of the field was 3.67 ha. However, Scottish Ministers had determined that, of this, a total of 1.08 ha was ineligible. Dr Banks contended that there was no clear evidence to support such deduction. He accepted that 0.03ha be deemed as ineligible land due to a pond at the east end. He expressed doubt as to where the inspectors had found the remaining ineligible areas. Although we did not understand him to concede any areas apart from the pond, he cross-examined both Ms Todd and Mr Reid on the basis that certain other areas might fall to be deducted. He pointed to approximately 0.14 ha shaded by the Sitka plantation but falling within a managed water margin. It was currently bare land with little vegetative growth. He identified two areas of gorse; one which he labelled “(a)” covering more or less the whole width of the field at its west end and extending in total to no more than 0.26 ha and another area, “(b)”, consisting of a previously fenced parcel of ground which could be measured as no more than 0.1 ha; a total of approximately 0.36 ha. However, he contended that this was not all impenetrable. He contended that no other areas of impenetrable gorse could be identified in the field.
 He took us through his photographic tour of the field with a set of photographs taken at intervals round the perimeter and in or near the watercourse to show the width of the watercourse. His ultimate contention was that he accepted the deduction of the pond at 0.03 ha. He contended that there should be no deduction for the 0.14 ha shaded by the Sitka; that the watercourse was under 2 metres for most of its length and ought not to have been excluded. He contended that none of the gorse was impenetrable although much of it was “thick”. Much of the thickest part was within the managed water margin and provided a habitat for linnets and yellowhammers but, on any view, the maximum area which could possibly be taken to be ineligible gorse could not exceed 0.36 metre.
 We found the evidence led for the respondents in relation to this field to be confused and inconsistent. Their formal position was that they took off 0.91 ha of impenetrable gorse and 0.17 ha for the watercourse. Neither Ms Todd nor Mr Reid mentioned the pond in their initial attempts to explain their calculations and, accordingly, it may reasonably be assumed that it was included as part of the 0.17 ha. Mr Reid’s examination had been limited to the gorse. He attempted to explain the whole deduction as attributable to the gorse and the watercourse. He had not made any deduction for the land overhung by Sitka but Ms Todd said she had. He had not made any deduction for any gorse in the eastern half of the field. But Ms Todd told us there were ineligible patches throughout.
 Ms Todd’s evidence was particularly difficult to assess because she started by describing things as seen from one point in the field and then had to accept that, in fact, she was talking about a quite different place. She had started off by explaining that she had gone to the field through the Sitka plantation and had entered it at a point which she identified clearly, and without any apparent doubt, by reference to numbers on one of the appellant’s productions (namely the handwritten number 29 shown on production 37). She said it was near the eastmost point of a triangle of two types of vegetation of some sort which could be seen on the Scottish Ministers’ aerial photograph of the field. That appeared to be an area bounded on the north by the line of an open ditch or drain and on the west by the boundary of the field. The south side of the triangle was hidden in the photograph by the dark shadow of the Sitka trees. Ms Todd initially said she had made a technical assessment of that whole triangle of vegetation and that it lay to the west of where she had come in to the field. She had properly been subject to detailed cross examination as to how much she could see from there and whether such a technical assessment was reliable in relation to such a large area. However, she was later persuaded by the cross-examination to change her evidence and confirm that in fact her assessment had been made from a point somewhere within the square shown as a yellow icon on the Midas sheet relating to that field in production 49. This was a point in the middle of the said triangle and at a place where the respondents’ aerial photograph shows a distinct division in the type of plant material. The effect of her evidence was that she was making an assessment of gorse to the west of her vantage point. That was admittedly gorse. We heard no detail of any assessment by her of the brownish coloured area shown in the aerial photograph and forming the east section of the triangle. It is unfortunate that Ms Todd was not able to clarify matters because of her illness and this may lead to an appearance of unfairness to her in our various observations. But, we mean no criticism of her personally. Our task is to reach a decision based on the evidence before us. We must simply make the best we can of it.
 We think that it can be inferred from this passage of evidence that Ms Todd may not presently have an instinctive “feel” for areas. For the most part this would not be of critical importance for her work. She would be able to rely on the kit for assessment of overall areas and her main task in the present context would be to assess the nature of the gorse. However, she spoke initially of her findings in relation to an area of land shown on the aerial photograph without apparently realising that the land shown was, in length and effective area, almost twice that of the piece of land she thought she was describing. Of more importance is the fact that, if her evidence in relation to the gorse at the west end of the field is to be taken to relate to the land to the west of her revised vantage point, it appeared to cover only the area identified by Dr Banks as (a). Although she said that her total figure of 1.0 ha was made up not only of the areas labelled by Dr Banks as (a) and (b) but also of the sum of other ineligible areas within the field, she did not give persuasive evidence as to where the latter lay - except for the land shadowed by Sitka. Mr Reid’s evidence provided no support for that and Dr Banks’s own evidence was that that there was no significant areas of gorse other than the areas he marked (a) and (b). Mr Reid said expressly that there was no ineligible gorse in the eastern half of the field.
 We accept the difficulties which inspectors have in giving evidence. They carry out many inspections and must tend to be dependant on their records of the assessments made to the best of their ability at the time. A further difficulty in the present case was that Mr Reid explained that his memory at present was adversely affected by a medical condition. He was unable to engage in much discussion of detail. He repeatedly said that he could only go by what he had measured on the day. However, he did say that he and Ms Kinnaird had not used GPS. He said “the leg work had already been done”. We understood that to mean that they were not engaged in a measuring exercise but simply in evaluation of the percentage eligibility of areas which had been measured. He apparently thought that Ms Todd’s figures related wholly to the gorse at the west end.
 However, where an inspector’s figures are challenged by proceedings in this Court it is necessary for us to reach our decision on the basis of the evidence presented to us. Dr Banks provided photographs to support his proposition that there were no other ineligible areas of gorse in field 5. Ms Todd was unable to describe positively any such areas. Dr Banks was broadly supported on this point by Mr Reid’s evidence.
 On the evidence we accept that there were two areas of gorse as spoken to by Dr Banks. We recognise that all three of the officials might be thought to support each other in their conclusion that the ineligible area in field 5 was something of the order of 1.0 ha. But where they were not consistent in explaining the basis upon which that figure was reached, we do not think that the similarity of figures can properly be regarded as providing mutual support. The fact is that Dr Banks was clear in defining the areas of gorse in question. The evidence for the respondents was confused and inconsistent. In the circumstances, we have, accepted Dr Banks’s assessment of the total of the areas affected by gorse. We accept it at 0.36 ha. He did not have the assistance of the “kit” or technology in assessing the area but he explained his broad approach. His figure is in line with our approximate assessments as scaled from the aerial photograph and our planimeter measurements. Indeed we understood Ms Todd to accept that such a figure would cover the patches of gorse he described. She thought there was other gorse.
 We have had greater difficulty in making an assessment of the eligibility of the areas. Although we have preferred Dr Banks’s evidence as to the overall extent of thick gorse, we have not found his assessments of eligibility of gorse to be reliable for the reasons discussed above. There was nothing in his evidence which would have allowed us to take the view that any significant parts of the two areas should be found to be eligible as permanent pasture. We have concluded that all of areas (a) and (b) should be deducted. They are areas of thick gorse and we have no reliable basis for making a finding that any parts were eligible.
 We have also had difficulty with the water. Neither of the respondents’ witnesses spoke expressly to exclusion of the pond and Mr Reid purported to support the deduction without reference to it. It is possible he simply assumed it to have been treated as part of the watercourse. Ms Todd said she had deducted the whole of the watercourse. This was on the basis that most of it was shown, by calculation from the OS based data, to be over two metres wide. However, we accept the direct evidence of Dr Banks that for at least half its length it was well under 2 metres. This was demonstrated by his photographs which included an A4 binder to provide a rough scale. Mr Reid made some attempt to suggest that the photographs might be misleading if the water undercut the banks but, of course, the aim was to identify pasture. The banks would not cease to be pasture simply because water could be identified as flowing underneath them. There was a faint suggestion that the flow in the watercourse might have been different at the time of Dr Banks’s photographs but Mr Cameron did not seek to challenge his evidence on that basis and we do not think it likely that this would have produced an explanation for the apparent discrepancy. It is fairly clear that towards the eastern end the watercourse became wider. One specific ditch was identified as running into it. There might have been other, less defined, contributory sources. There is a tendency for watercourses to increase in width downstream although over short distances much depends on slopes and the nature of the land. A question arises as to whether the proper approach is to take an average or to attempt to identify sections which are under two metres.
 We think a broad approach is necessary. We have little doubt that, in theory, it is appropriate to take a discriminative approach. There may well be occasions where a watercourse is so irregular in width as to make it impossible to suggest that any significant length should be accepted as under two metres. There may also be occasions where a watercourse is so predominantly in excess of two metres that it would be unrealistic to attempt to isolate some short narrow stretches. However, at least where there is a gradual progression as the water flows downstream, we have no doubt that the provisions of art 30(2) of Reg 796/2004 allow the Scottish Ministers to identify a point beyond which the watercourse is ineligible and accept the section under 2 metres as eligible.
 In the present case, it is difficult to make an accurate re-calculation because we heard that the overall deduction was based on a measurement provided by the kit based on OS data which appeared to show the burn as over 2 metres throughout its whole passage through the field. If we deduct the length which we accept as being under 2 metres in width and attempt to adjust the calculation on the assumption that it had originally been assessed at precisely 2 metres, there will be an underestimate. We cannot compensate for the underestimate because we have no evidence bearing on the actual width assumed by the OS data over that stretch. Dr Banks said there were few stretches which he could not step over but his positive assertion to the witness was simply that there was a stretch of at least 350 metres which were less than 2 metres wide. There was no satisfactory rebuttal. There is some difficulty in making our own assessment of the relevant length because the burn meanders to some extent and in places passes into the adjoining plantation. However, we consider that Dr Banks’s assessment can be accepted. That stretch should not have been deducted. Thus, a correction of at least 700 square metres or 0.07 ha is required. We consider that some deduction for water including the pond and part of the burn required to be made. We cannot assess this precisely but on balance conclude that the appropriate deduction for the watercourse should have been no more than 0.1 ha.
 In the result, therefore, we find that a total of 0.36 ha falls to be deducted as ineligible gorse, together with 0.14 ha in respect of the land affected by the Sitka, a total of 0.5 ha, instead of 0.91 ha, and 0.1 ha of water, instead of 0.17 ha. This is a deduction of 0.6 ha instead of 1.08 ha made by the respondents
 Although this parcel was not referred to explicitly at earlier stages of the appeal process, we accepted that it was appropriate to deal with it. The point raised was a short one and Scottish Ministers had ample notice of it.
 There was no dispute as to the facts on the ground although there was, in the evidence, much dispute as to how matters had been dealt with and how they should have been dealt with. The area claimed was 8.83 ha. However, it was agreed that the fence with Field 5 was moved whilst fencing in connection with a RP 2009 scheme. The findings in relation to Field 5 showed a reduction in the claimed area to reflect the change in the fence. This removed 0.16 ha from that field. There was no dispute that this was appropriate. The physical area of Field 6 area was increased by that 0.16 ha but in the course of the evidence there appeared to be a dispute as to whether or not the additional area had been accepted. This was not fully clarified until the submissions when Mr Cameron made it clear that this had been treated as an issue of tolerance. The Scottish Ministers took the view that if the found area was within 2% of the claimed area, the claimed area had to be used. So although the excambed portion had been taken from field 5 it was too small to be added to field 6.
 We are not satisfied that the “tolerance” provisions are, or were, applicable where there has been an identified change of boundary which has reduced the area of one parcel and increased the area of the adjacent one. This not an issue of tolerance of measurement but of a clearly identified change. We consider that the proper approach was that set out by Ms Kinnaird in her email of 12 October 2010 in response to the explicit question raised in Dr Banks’s message of 18 September. Accordingly, we are satisfied that, having adjusted the area of Field 5 to remove the land transferred to Field 6 by a new fence, that area fell to be added to Field 6. The area eligible for SFP purposes for Field 6 should be 8.99 ha. That is, an addition of 0.16 ha to the eligible figure.
 The claimed areas of fields 17 and 14 at Craibstone were 2.78 ha and 3.31 ha respectively. However the boundary between these fields was moved by GPS mapping and Field 14 was increased to 3.50 ha. There is no dispute over that change.
 The area of Field 17 was reduced by deduction of the watercourse (0.25 ha) to 2.53 ha. The appellant did not dispute this deduction. However, a further 0.1 ha was removed. Ms Todd explained that this was because of elements within the water margin which was managed under RP. She described specifically an outcrop of rock and stones as being the dominant feature to be deducted. Although she said there might have been other elements in the water margin which had to be deducted she could not give positive evidence identifying such other elements. In any event, as we understood her evidence the outcrop was said to form at least the vast bulk of the 0.1 ha. She did not identify any other significant feature and Dr Banks gave evidence, supported by his photographs, challenging the existence of any other significant feature.
 We accordingly accept that there was an area of stone to be deducted from the area of the water margin. Ms Todd said that was 0.1 ha but when asked to describe the area by reference to the size of the courtroom she described it as extending to no more than about three quarters of the courtroom. That, in fact, is an area of less than 60 square metres, in other words, less than 0.006 ha. Although we place no great weight on the specific area indicated by her, we consider that, in broad terms, this direct evidence must be preferred to the evidence of her recorded figure. On the face of it there has been an error of measurement or of recording the measurement suggestive of a simple error over a decimal point at some stage in the assessment process. We are satisfied that her deduction of 0.1 should be revised to 0.01 ha.
 We find on this evidence, that the proper area of Field 17 for present purposes should be 2.52 ha. That is, an addition of 0.09 ha
 This field was formerly part of the parcel known as field 23 (NJ/49996/58924) which was said to have had an area of about 3.05 ha. At inspection this parcel was split into 2 parts, Fields 23 (NJ/49976/58931) and 24 (NJ/50039/58950) because a fence had been erected apparently to protect land subject to environmental schemes. Following inspection the area of Field 23 was determined as 2.57 ha. No parts of the other field, Field 24, were accepted as eligible for SFP. It was not disputed that the bounded area of the field was 0.55 ha.
 The field is narrow and very steep. Some 0.1 ha was being managed under an RP scheme. It was also apparent from aerial photographs that there was a distinct area of open grassland. This was said to be very steep. However, it was not contended by Scottish Ministers that the steepness made it inaccessible to stock. Ms Todd said that she disallowed the whole field because it appeared to be fenced off and accordingly inaccessible.
 We are satisfied that eligibility of land for SFP purposes depends on the characteristics of the land as permanent pasture, not on the use which happens to be made of it from time to time. Where land is being managed under environmental protection or enhancement schemes a question may arise as to whether it might lose its character as pasture by virtue of physical changes following on compliance with a relevant environmental scheme. For reasons discussed above we have not felt it necessary to reach a concluded view on that issue. However, we are satisfied that, where land has the characteristics of pasture, it does not lose these characteristics merely because it is not being actively used by farm stock.
 In relation to this parcel, we accept the evidence of Dr Banks that access could be taken at the east end of the parcel and that stock could find their way to many parts of it. In any event the fence could be removed. We do not think that Ms Todd’s basis for the decision of Scottish Ministers in relation to Field 24 can stand.
 We have had some difficult in determining what figure should be used. We accept the evidence of Dr Banks that at least 0.1 ha should be allowed. That appears to us to be a conservative assessment of the apparent grass areas on the photograph. But we recognise that it is not possible to make an accurate assessment from photographic evidence and we cannot confidently identify any area beyond his estimated figure. We think it reasonable to hold that 0.1 ha can be taken to be pasture. That figure must accordingly be added to the total found figure.
 Dr Banks challenged the accuracy of the OS based assessment arising from the fact that the blue line on the aerial photograph was some distance away from the watercourse which was the correct boundary. We have dealt with that issue above. We proceed on the basis that the external boundary was agreed to encompass an area of 5.07 ha. The Scottish Ministers deducted 1.10 ha because of rock and dense gorse. The appellant accepted that approximately 0.10 ha be deemed as ineligible land due to bare rock and field gatherings and that some deduction fell to be made in respect of gorse. He accepted that stands of gorse covered some 1.00 ha but contended that this was not dense or impenetrable gorse. He said that no more than 50% was dense and the deduction for gorse should not have exceeded 0.50 ha, a total deduction of only 0.60 ha.
 For the reasons discussed above, we consider that the Scottish Ministers’ evidence as to eligibility of stands of gorse is to be preferred to that of Dr Banks. The areas of gorse are not in dispute and we have no sound basis for interfering with their assessment of the nature of the gorse they found. We accept that a combination of thick gorse and steepness may preclude land from being properly described as pasture. We consider their deduction of 1.10 ha must stand.
 The effect of our various findings is that the total area of eligible land found must be adjusted. To the original figure of 252.16, we add a total of 0.83 ha being 0.48 ha in respect of Field 5, 0.16 ha in respect of field 6; 0.09 ha in respect of field 17; and 0.1 ha in respect of field 24. We make no adjustment in respect of field 8/9. The revised found total is accordingly 252.99 ha. This is still 1.34 ha short of the 254.33 claimed entitlements but the shortfall is less than 2 ha and no penalty falls to be imposed.
For the appellant: Dr G P Banks, Peterhead
For the respondents: Donald Cameron, Advocate; Scottish Government Legal Directorate