No to ILI Cato DC

Objection to Planning Application 26/01243/PPP

  • To: Scott Simpson, Planning Service, Fife Council, Fife House, North Street, Glenrothes KY7 5LT
  • Application: 26/01243/PPP, Planning Permission in Principle
  • Proposal: Data centre (Class 6) and associated development ("Cato")
  • Site: Land to the north-west of Auchtertool / Camilla Farm, KY2 5XW
  • Applicant: ILI Cato Ltd (agent: LoganPM Ltd)

This is the text of a planning objection to Fife Council on application 26/01243/PPP. Personal contact details have been removed for publication.

I object to this application. I am a Fife resident in Kirkcaldy, downstream of the application site on the watercourse that drains it (the Dronachy Burn, which feeds the Tiel Burn and Raith Lake before reaching the Firth of Forth at the Kirkcaldy coast), and I have a direct interest in the water environment, the designated landscape and the safety case for a 600MW facility of this scale in this location.

My objection is not a general dislike of data centres. It is that this particular application, on this site and in this form, is not in accordance with the development plan, has been screened out of Environmental Impact Assessment on premises the applicant's own documents contradict, and defers almost every determinative environmental question to a later stage. I set out my grounds below. Each is tied to the relevant policy and, wherever possible, evidenced from the applicant's own submitted documents rather than from assertion.

The grounds, in summary, are: (1) the development should have been subject to EIA and the screening-out is unsound; (2) the application defers the determinative environmental assessments and so cannot be properly determined; (3) it puts at risk the Camilla Loch Site of Special Scientific Interest; (4) it demolishes a castle against the advice of the Council's own archaeologist; (5) its energy, water, flood-risk, major-accident, landscape, noise and cumulative effects are either unjustified, unassessed or contradicted within the application; and (6) the evidence base is internally inconsistent to a degree that undermines the weight it can carry.

A. The application should be subject to Environmental Impact Assessment

1. The EIA screening-out is unsound and should be reconsidered. The screening opinion (reference 25/03079/SCR, issued 1 April 2026) screened this development out of EIA. It rests on three factual premises that the applicant's own later application documents contradict:

  • The opinion screens Camilla Loch SSSI out on the basis (attributed to NatureScot) that there is "no hydrological connection" between the site and the SSSI. The applicant's own Ecological Impact Assessment, however, records the SSSI connectivity as "yes", concedes a potential significant effect through alteration of ground and surface water chemistry and availability in the loch's western catchment, and states that a hydrogeological assessment is required to confirm the connectivity and the level of effect. It is not clear whether NatureScot's no-connection advice considered groundwater or only surface water; if it considered only surface water, the groundwater pathway the Ecological Impact Assessment identifies was never assessed at screening, and if it considered both, its conclusion is difficult to reconcile with the applicant's own later ecological work. On either reading, the groundwater pathway to the SSSI remains unresolved and is deferred, which is precisely the matter screening should have determined.
  • The opinion states that no abstraction of groundwater is proposed; the Ecological Impact Assessment concedes "significant abstraction of groundwaters".
  • The screening request told the Council the applicant had committed to "no diesel or gas and no back-up generators on site", and on that basis combustion, air quality, accident risk and pollution were all recorded as not significant. The Air Quality assessment then describes "high-capacity standby generators" that are "tested regularly"; the Energy statement concedes that a number of generators are included; and the Noise assessment models around 25 generators on the campus. Substituting a renewable fuel for diesel is not the same as having no generators: the combustion, air-quality, bulk-fuel-storage and major-accident dimensions were screened out on a premise the application does not meet, so they were never assessed at all.

Beyond these contradictions, the screening opinion repeatedly defers individual receptors to "any future application for planning permission". A conclusion of no likely significant effect cannot be reached by relying on assessments that have not been carried out, because deciding whether those assessments must be done within an EIA is the entire purpose of screening. The opinion also assessed heritage only through the setting of Auchtertool House at 687 metres and never registered that the application totally demolishes a castle on the site; and it screened landscape and visual effects out even though the applicant's own Landscape and Visual Appraisal reports "Major" and "Major / moderate adverse" effects, assessed under a framework the consultant has expressly chosen to avoid using the word "significant" because a significance finding "could trigger the requirement for a formal EIA". The Council's own significance checklist records the effect as long-term, permanent, and a low-probability but potentially highly significant effect, yet still concludes no EIA is required; those are Schedule 3 criteria pointing toward EIA, not away from it.

The screening request also misstated the baseline, describing no statutory ecological designations within one kilometre when the Camilla Loch SSSI lies a little over 200 metres away, and described the workforce as 50 full-time staff when the application itself is for 120, so the development that was screened out of EIA was presented with under half its applied-for workforce.

A request to reconsider the screening opinion has already been made to the Council (by Gemma Easton, 12 May 2026). I support that request and ask the Council to adopt a fresh screening opinion in light of the contradictions above, and I note that the Scottish Ministers retain the power to issue a screening direction before the application is determined. As part of that reconsideration, I ask the Council to re-consult NatureScot specifically on the Ecological Impact Assessment that records connectivity to the SSSI and a potential significant effect on it, since NatureScot's earlier "no hydrological connection" advice appears to predate that assessment and it is not evident that NatureScot has seen it; and to seek SEPA's position both on the deferred hydrogeological assessment for the SSSI and on the undisclosed bulk-fuel storage and its compliance with the Controlled Activities Regulations. Relevant framework: the Town and Country Planning (EIA) (Scotland) Regulations 2017, Circular 1/2017, and the principle in R (Finch) v Surrey County Council [2024] UKSC that reasonably foreseeable effects (here, generator combustion) cannot be left out of account.

2. The application defers the determinative assessments and cannot properly be determined. This is an application for permission in principle, but it goes further than leaving design detail to a later stage: it defers the very assessments that determine whether the principle is acceptable. The hydrogeological assessment that would establish whether the SSSI is harmed is deferred to pre-construction; the generator count, size, fuel type, fuel volume and run regime are deferred to conditions; the construction traffic management plan is to be "finalised in advance of the commencement of site works"; the contaminated-land ground investigation is deferred; the noise mitigation depends on a chiller procurement limit that does not yet exist; and the landscape mitigation rests on an indicative planting scheme that is not secured. The Council is being asked to accept the principle of the development while the evidence needed to judge that principle is postponed. Where a deferred matter could itself determine acceptability (the SSSI hydrogeology being the clearest example), deferral is not a lawful substitute for assessment. This is a matter of prematurity and of NPF4's requirement that decisions be based on sufficient information. The point is sharpest for the statutory designation: where the applicant's own ecologist has identified a potential significant effect on the Camilla Loch SSSI and the assessment needed to resolve it has not been carried out, the precautionary approach requires that the uncertainty be resolved before consent is granted, not afterwards by condition.

3. Alternatives and the case for this location are inadequately made. The applicant presents a nationwide site search but does not demonstrate that the environmental harms of this particular site (the SSSI, the castle, the on-site landfill, the flood-zone land, the designated landscape) were weighed against alternative sites without those constraints. For a development of this scale and impact, the absence of a genuine alternatives assessment is a material shortcoming.

B. Principle of development

4. The site's allocation is tied to the Mossmorran complex and that basis has not been extinguished. The land's industrial standing in the development plan is associated with the Mossmorran complex. The Shell Fife Natural Gas Liquids plant remains operational, so the premise on which industrial use here was contemplated has not fallen away in the manner the applicant's narrative implies. The principle of a 600MW hyperscale data centre is a materially different proposition from the allocation's original intent and should be assessed as such.

5. The employment benefit is thin for the scale of impact. The applicant relies heavily on economic benefit, but a 72.9-hectare site delivering on the order of 120 operational jobs is a very low employment density, and the headline jobs figure is itself inconsistent across the application (see ground 24). The benefit claimed should be tested against that low density and against the inconsistency in the figures, not accepted at the higher number.

C. Climate and energy

6. The climate case does not meet NPF4 Policy 1. The application presents the development as supporting net zero, but the substance does not bear this out. Power purchase agreements do not by themselves add renewable capacity to the grid (they are not additionality); the standby generators combust fuel whatever its source, and "non-fossil" fuel such as HVO still produces combustion emissions; and the embodied carbon of six buildings up to 35 metres, a three-year construction programme and the associated plant is not assessed at all. A continuous load rising to around 4,000 GWh per year is a very large new electricity demand, and the climate balance NPF4 Policy 1 requires has not been properly struck.

7. The district-heating and low-carbon commitments are intentions, not commitments. The heat-offtake proposal relies on low-grade waste heat with a community-fund fallback if it cannot be delivered, and the energy-efficiency commitments are expressed as a statement of intention "or equivalent" rather than as binding parameters. Against NPF4 Policy 11 and FIFEplan Policy 11 (Low Carbon Fife), intentions of this kind carry little weight and are not secured.

D. The Camilla Loch Site of Special Scientific Interest

8. The development puts the Camilla Loch SSSI at risk, on the applicant's own evidence. Camilla Loch is a statutory Site of Special Scientific Interest (NatureScot site 307), notified for water-dependent features including a meso-eutrophic loch and an open-water transition fen described as the best example of its type in West Fife, at the loch's western end. The applicant's own Ecological Impact Assessment concedes a potential significant effect on the SSSI through alteration of ground and surface water chemistry and availability in the loch's western catchment, and concedes "significant abstraction of groundwaters". Having identified that pathway, the assessment then defers the hydrogeological work that would determine the effect to a pre-construction stage and proceeds as if the conclusion were already favourable. That is the wrong way round: where the applicant's own ecologist has identified a potential significant effect on a statutory designation, the assessment to resolve it belongs in the determination of this application, not after it. The earlier screening-out relied on a "no hydrological connection" conclusion; whether or not that advice extended to groundwater, the groundwater pathway the Ecological Impact Assessment identifies was not resolved at screening and remains deferred. In addition, nitrogen deposition on the SSSI from the generators, the wider plant and traffic, and cumulatively with Mossmorran, is not assessed by any document. This engages NPF4 Policy 4 (natural places) and the Council's duty to further the conservation of the SSSI's features. I accept that the survey effort in the Ecological Impact Assessment appears competent and that the screening of the Firth of Forth designations for geese appears reasonable; the objection is specifically the deferred groundwater question and the unassessed nitrogen deposition.

9. Groundwater-dependent terrestrial ecosystems and biodiversity (NPF4 Policy 3). The Ecological Impact Assessment confirms groundwater-dependent terrestrial ecosystems on the site (wet grassland described as moderately groundwater dependent), which directly contradicts the contamination assessment's conceptual model statement that there is no such ecosystem within 250 metres. The proposed translocation of wet grassland to the floodplain is uncertain, and biodiversity net gain is claimed in accordance with NPF4 but with no metric to support it. The Auchtertool Linn local wildlife site lies partly within the site.

10. Woodland and ancient woodland connectivity (NPF4 Policy 6). The Dronachy Burn is a wooded riparian corridor connecting on-site woodland to downstream plantations in and around the Raith designed landscape. The Ecological Impact Assessment treats trees only as habitat and never addresses ancient or long-established woodland status or downstream connectivity as a Policy 6 question. The applicant's own heritage site photography is captioned as showing an "area of Ancient Woodland" beyond the site to the south-west, which confirms ancient woodland in proximity that no document assesses for impact or connectivity. The Ancient Woodland Inventory status of the nearby plantations should be checked and the connectivity question answered before any decision.

E. Water environment and flooding

11. The development fails the NPF4 Policy 22 avoidance test. The Flood Risk Assessment gets a flood-risk site over the line by classifying the data centre as "Essential Infrastructure" (in its own tentative phrasing) to invoke the Policy 22 exception, but that exception also requires that the location is required for operational reasons. No operational locational need is demonstrated; the drivers are grid connection, land availability and commercial preference. The assessment's own compliance table undercuts the classification by arguing the facility can be operated remotely in emergencies. It also leans on flexibility intended for previously developed sites in built-up areas while describing the site throughout as greenfield grassland. On greenfield land, the first principle of Policy 22 is avoidance, and avoidance was achievable here.

12. Downstream water quality is not properly assessed. The Drainage Impact Assessment treats site runoff as generally uncontaminated using an office-car-park category and places hydrocarbon interceptors at the car parks rather than at the generator and fuel compound, which is the highest hydrocarbon hazard on the site. The receiving watercourse runs through the Raith designed landscape and on to the Firth of Forth at Kirkcaldy; the Firth of Forth carries SSSI, Special Protection Area and Ramsar designations, and the potential downstream pathway to those designated waters was treated only as a bird-feeding matter, not as a water-quality receptor, and was never screened as a Habitats Regulations question.

13. Downstream flood receptors are not assessed. The assessment addresses flood risk to and from the site but does not assess the effect of altered runoff and discharge on downstream receptors along the Dronachy and Tiel Burn corridor toward Kirkcaldy.

14. Standby generation and bulk fuel storage are an unassessed pollution and hazard source. A facility of this scale requires standby generation broadly matched to its critical load. On the applicant's own cited industry standard of 72 hours of fuel autonomy, a critical load of this order implies on-site fuel storage on the scale of several million litres (roughly five to seven thousand tonnes). The application gives no figure, so I put the burden on the applicant to disclose it, but at that scale the store is a major hydrocarbon pollution source (through tank and pipework failure, overfill, refuelling and delivery spills, transformer oil and routine generator testing) and, as set out below, may cross the threshold for the major-accident regime. The Noise assessment's disclosure of around 25 generators is consistent with a substantial fuel inventory. Bulk fuel storage over 2,500 litres is regulated by SEPA under the Controlled Activities Regulations 2011, General Binding Rules 26 to 28 (secondary containment, impermeable bunding, drip trays); the application should specify and demonstrate compliance and does not. In a flood or drainage exceedance, flows bypass the sustainable drainage train and run overland to the burn, so a fuel release coinciding with flooding is the realistic worst case and is not assessed anywhere.

F. Major-accident hazard

15. The major-accident and societal-risk dimension is wholly unassessed. The application does not address major-accident hazard anywhere. There are three layers to this. First, the Mossmorran complex (the operational Shell Fife Natural Gas Liquids plant) is a COMAH establishment regulated jointly by SEPA and the Health and Safety Executive, with an off-site emergency plan and a public consultation area; siting a 120-person operational workforce, a multi-year construction workforce and occupied buildings in its vicinity is a land-use-planning matter on which the Executive is a statutory consultee, and the application does not establish whether the site lies within Mossmorran's consultation distance or whether the Executive was consulted. Second, on the fuel volumes implied above, the development may itself be a lower-tier COMAH establishment and may require hazardous substances consent, creating its own consultation zone over Auchtertool and the SSSI; this is undisclosed and unassessed. Third, the co-location of a large new fuel store, the existing grid-scale battery storage facility in the north-east of the site (with its thermal-runaway fire and toxic-gas risk), the data centre's own batteries, and the adjacent gas plant raises the question of interacting hazards (domino effects), which COMAH requires to be considered and which no document addresses, and no battery fire-safety assessment appears. I make no allegation here; the point is simply that none of this has been assessed, and on the no-generators premise of the screening it never could have been. This engages NPF4 Policy 23, the EIA Regulations' requirement to consider risks from and vulnerability to major accidents, the hazardous substances consent regime, and COMAH 2015.

G. Heritage

16. The development demolishes a castle against the advice of the Council's own archaeologist. The application proposes, in the applicant's own words, to "demolish and fully remove the remains of this castle": the total loss of Camilla, formerly Hallyards, Castle, assessed by the applicant's own consultant as a "High Adverse" effect. The Council's records describe the castle as possibly originating in the eleventh century, documented from before 1539, with royal connections. The Supporting Statement described these remains as "a few sections of wall" to be "recorded and removed", but that phrase was the one Fife Council's Archaeological Unit applied to the lesser nineteenth-century cottage row beside the castle, not to the castle itself, so the headline document understated what is being lost. Critically, the Archaeological Unit advised (22 September 2025) that the applicant should, "if at all viable", design a layout incorporating the castle into untouched greenspace, calling that the best outcome for the asset and very probably in the applicant's financial interest, with the default position being preservation in situ. The applicant instead chose the option its own heritage evaluation describes as the greatest loss to the heritage resource. NPF4 Policy 7 requires non-designated assets to be preserved in situ wherever feasible, and FIFEplan Policy 14 permits loss only where there is no reasonable alternative means of meeting the development need. On a 72.9-hectare site, with a castle footprint of around a tenth of a hectare, an indicative and unfixed layout, and the Council's own archaeologist advising that retention is viable, the applicant has not demonstrated that there is no reasonable alternative. "Preservation by record" (excavation, recording, publication and then destruction) is the policy fallback once in-situ retention has been shown to be impracticable; the applicant has gone straight to the fallback without discharging that precondition, and even the recording is deferred to a condition. The applicant's own landscape appraisal confirms the castle walling is a prominent feature in the existing view that the proposed buildings would screen entirely, so the asset is both physically demolished and visually obliterated. I accept that the assessment of the designated heritage assets (the Raith and Beveridge Park designed landscape and the listed buildings) is competent and finds only negligible to minor effects; the objection is specifically the unjustified total demolition of the castle.

H. Landscape, amenity, noise, transport and soils

17. Harm to a designated landscape (NPF4 Policy 4). The applicant's own Landscape and Visual Appraisal finds, even at year 10 with mitigation planting matured, residual Moderate to Major/moderate adverse effects on the Cullaloe Hills and Coast Local Landscape Area (a Fife local landscape designation whose western perimeter defines the site boundary), on local landscape character, and on the valued recreational routes and viewpoints (the core path that crosses the site, the path at Camilla Loch, and viewpoints near the site). During construction and at year 1, the effects are Major or Major/moderate across nearly all receptors. Six buildings up to 35 metres tall, with a combined footprint of around 14 hectares (close to the area of the whole village), in open farmland, are theoretically visible from around 37 per cent of the study area even after screening by woodland and buildings is taken into account; and the appraisal concedes that the buildings rise above the tree canopies to form a new horizon line, so the planting screens only their base, not their bulk. The more palatable year-10 figures depend entirely on indicative planting maturing as hoped, on an application where the layout, heights, cladding and planting scheme are all reserved and unsecured. As noted under ground 1, the appraisal is also deliberately framed to avoid the word "significant". I accept the methodology is competent, the existing industrial context is real, the village core is largely screened, and the planting and wetland have genuine biodiversity and flood value; the objection is the persisting adverse harm to a designated landscape and to the valued scenic receptors.

18. Noise and residential amenity (NPF4 Policy 23). The applicant's own Noise Impact Assessment predicts a "Significant Adverse Impact" from operational plant at the nearest homes (the Camilla Road properties, recorded as 50 metres from the site boundary), day and night, before mitigation. Even after a substantial mitigation scheme (rooftop and ground-level acoustic barriers and façade linings), the night-time effect at those homes remains adverse, and the report concedes the requirements would not be met there unless an additional chiller procurement noise limit is achieved, a limit that does not yet exist. The rural night-time background is as low as 34 decibels, and the assessment applies a tonal "hum" correction, acknowledging a continuous mechanical hum at quiet rural homes. The assessment excludes the generators from its routine assessment on the basis that they are emergency-only and rarely used, but the Air Quality assessment says they are tested regularly, so the recurring noise of testing around 25 generators is unassessed.

19. Transport. I accept the operational traffic case is modest and competently presented (around 300 movements a day, a roughly 14 per cent increase on the B925). The objection is the construction phase, which is deferred: the outline construction traffic management plan contemplates up to 100 heavy goods vehicle movements a day over a 36-month programme, plus abnormal loads, with no junction or network capacity analysis and with the plan expressly to be finalised before works begin. There is no cumulative construction-traffic assessment with the concurrent consented solar and battery development to the east, and the applicant's own accident data shows a higher-than-average single-vehicle accident rate on the B925 bends, which three years of heavy and abnormal-load traffic would aggravate.

20. Core paths. Core path R496 currently crosses the site and would be rerouted along the Dronachy Burn. The applicant's landscape appraisal assesses the rerouted path as the worst-affected route in the study area (Major adverse during construction and at year 1). The long-term betterment is not denied, but the construction-period amenity and the deliverability of the reroute (along the burn corridor near the on-site landfill and the castle) depend on matters still deferred.

21. Contaminated land and ground gas. There is an on-site historic landfill (Dronachy Den) with a contaminant suite including asbestos, hydrocarbons and metals, and the assessment records high ground-gas potential. The detailed ground investigation is deferred, which is a prematurity concern given the landfill's interaction with the watercourse and the flood-risk land. I accept that the coal-mining legacy risk is assessed as negligible and the radon protection requirement is acknowledged.

22. Prime agricultural land. I do not pursue an objection on the loss of prime agricultural land: the land capability classification is confirmed as non-prime, and I accept that point.

I. Cumulative effects and the adjacent battery storage facility

23. Cumulative effects and the energy-infrastructure cluster. The site sits within an intensifying cluster of energy infrastructure: the operational Mossmorran complex, the Little Raith wind farm, the existing Camilla battery storage facility in the north-east of the site, two substations, pylon lines, and a consented but not-yet-built 39MW solar farm with 10MW battery storage directly to the east (reference 23/02598/FULL). The application assesses none of the cumulative effects of stacking a 600MW data centre, its generators and fuel store, its grid connection and the surrounding infrastructure as a single cluster. The landscape appraisal expressly declines to assess cumulative landscape effect on the basis that no other data centre is nearby, which is too narrow a reading of cumulative impact. Cumulative nitrogen deposition on the SSSI, cumulative major-accident hazard, cumulative water effects and cumulative construction traffic are all unassessed, and the screening opinion dismissed cumulative effects in a single asserted sentence.

The existing Camilla battery storage facility was consented and built as a standalone grid asset, justified on renewable integration, grid balancing and relieving constraints on grid capacity. It was developed by the same group (ILI) that is now promoting the adjacent data centre, and both would connect at the same grid node, yet neither application acknowledges or assesses its relationship with the other. I raise two questions for the Council to put to the applicant, as a matter of transparency and consistency rather than as proven facts: first, whether a 600MW continuous load consumes the grid headroom the battery facility was approved to relieve; and second, whether the public-benefit basis on which the battery facility was consented remains valid if it is, in practice, supporting the data centre. I do not suggest the battery facility removes the need for generators (it is far too small to back up a load of this scale); the point is the absence of any assessment of the two developments together.

J. Reliability of the evidence base

24. The application documents are internally inconsistent to a degree that undermines their weight. Individually these are minor; together they show an evidence base that cannot be relied upon without scrutiny. The site area is given as 72.9 hectares, around 66 hectares, around 72 hectares and 68.6 hectares in different documents. The development area is given as around 25 and around 30 hectares. The generators are described four different ways (should not be required; high-capacity and tested regularly; none on site; and around 25 modelled), and then as emergency-only to exclude them from the noise assessment while being tested regularly in the air-quality assessment. Water demand was quantified for screening (around 1,000 cubic metres a day) but withheld from the energy statement. The screening's "no groundwater abstraction" is contradicted by the Ecological Impact Assessment's "significant abstraction". The SSSI was screened out on "no hydrological connection" while the ecology assessment records connectivity as "yes". The workforce is 50 in the screening request and 120 in the application. The castle is "a few sections of wall" in the Supporting Statement but a totally demolished castle of possible eleventh-century origins in the Heritage Assessment, whose history the Landscape Appraisal in turn sources to Wikipedia with a destruction date (1560) contradicting the Heritage Assessment's 1847. The drainage assessment carries another project's reference number and lists hazards irrelevant to this scheme. The site is placed both north-west and east of Auchtertool. The cumulative effect of these inconsistencies is that the Council cannot take the application's headline figures and conclusions at face value.

K. Statutory duties

25. Statutory duties. In determining this application the Council must have regard to its duties under section 44 of the Climate Change (Scotland) Act 2009 (to act in the way best calculated to contribute to emissions reduction and adaptation) and section 1 of the Nature Conservation (Scotland) Act 2004 (to further the conservation of biodiversity), as well as the duty to protect the features of the Camilla Loch SSSI. The unassessed emissions, the deferred SSSI hydrogeology and the unassessed nitrogen deposition engage these duties directly.

Conclusion and what I ask the Council to do

For the reasons above, this application is not in accordance with the development plan and is not supported by adequate environmental information. I ask the Council to:

  1. reconsider the EIA screening opinion in light of the contradictions in ground 1, and adopt a fresh screening opinion (I support the reconsideration request already on file);
  2. refuse the application as contrary to NPF4 Policies 1, 3, 4, 6, 7, 22 and 23 and the corresponding FIFEplan policies; or, at the least,
  3. decline to determine the application until the deferred and determinative assessments (the SSSI hydrogeology, the major-accident and fuel-storage case, the contaminated-land investigation, and the construction and cumulative-traffic assessments) have been carried out and consulted upon, so that the principle of the development is judged on the evidence rather than on deferral.

I would be grateful to be notified of the decision on this application and of any amendments to it.