The Voice of a Community

Written Evidence submitted by the Friends of the Sound of Jura to the Scottish Government, February 2018

The recent report carried out by SAMS for the Scottish Government states that science measures environmental effects, but society must decide whether their impacts are acceptable. In December 2017, five million One Show viewers were shocked to see a convey of trucks carrying away dead, diseased farmed salmon. The effects of such mass mortalities are just one of the environmental impacts laid out in the report; the unnecessary consequences of open-cage salmon farming, as it is being practiced in Scotland. These impacts are harming the industry’s profits, Scotland’s reputation for quality, and the health of our seas. Surely it makes sense to solve these problems before contemplating any expansion, let alone pressing on to double production by 2030?

Coastal community groups, worried about these threats to our sustainable jobs, have formed the Salmon Aquaculture Reform Network, Scotland (SARNS). Friends of the Sound of Jura (FoSoJ) is part of this coalition. Other SARNS groups have filed written submissions on various topics and we wish to be counted as agreeing with them.  We also agree with the submission by Salmon and Trout Conservation Scotland on the impacts on wild salmon and sea trout, as detailed in the independent review commisisoned by them from NINA (2018)1. Both species are Priority Marine Features and both are suffering harm at population levels from sea lice and escaped farmed fish. Mitigation must no longer be relegated solely to the Local Authority planning processes, which issues permanent permits, farm by farm, yet (according to Argyll & Bute Council’s written submission) cannot allow for the cumulative impact of multiple farms. Consenting must move to depend much more on community agreement and local knowledge of sites.

While the report’s review of the science is generally thorough, its opinions reflect SAMS dependence on aquaculture funding; for instance that for the foreseeable future fish farming must inevitably be in open cages at sea, and that the deliberate overuse of chemicals is very limited. The number of recorded EQS breaches and SEPA’s internal ‘PAMP 2 response options’ paper (below), imply otherwise.

SAMS promotes ‘Adaptive Management’ as a complete solution, but this approach of ‘going ahead and seeing what happens’, as Prof. Tett described it, then trying to clean up afterwards ‘adaptively’, is not new. The regulators and industry have always taken this approach, with each unanticipated crisis compounding the next. No wonder the SAMS panel could not explain how this method includes the precautionary principle. They did agree that AM is meaningless if the consenting process is not accurate and transparent, if environmental monitoring is not consistent, impartial and sensitive enough to detect every significant effect, if site-specific data is kept secret (making independent long-term analysis of environmental impacts impossible) and if the agencies do not do their own analysis to spot emerging problems, then set appropriate new thresholds. They must always penalise the companies that breach them too. SEPA and FHI should police their own rules, despite what Prof. Tett suggests.

Some long-term analyses are done occasionally, but it seems the regulators do little or nothing with the results. It is shameful, for example, that the 2002 precursor to this inquiry heard that maerl was being damaged by waste from 16 fish farms, then, after 16 years, SEPA had to admitted that there are now 29 farms at sites registered as having had maerl, but that it is no longer present at 13 of them. This is a dereliction of the agency’s legally binding biodiversity duty. SEPA should also limit farm biomass to protect wild fish, which it is empowered to do.

Even though SEPA’s records show there have been hundreds of breaches of its EQS/AZE/maximum biomass thresholds, we understand that no aquaculture CAR licence has ever been revoked and that, despite these and other breaches of FHI’s upper sea lice thresholds, and of other standards (eg on the disturbance of cetaceans with ADDs), no fines are issued and there has never been a successful prosecution. To this end the Lord Advocate has been asked to clarify whether self-reported data is admissible as evidence in court. If this amounts to self-incrimination, all voluntary self-reporting must be replaced by properly resourced independent monitoring.

Currently, consenting, oversight and enforcement are not sufficiently impartial, thorough or effective at preventing impacts and the regulating agencies seem to be failing in their function. The system should be overhauled to properly separate industry from its regulatiors.

This is especially important regarding Priority Marine Features (which are not always inside protected areas) and in the MPA/SAC protected areas themselves. A Februray 2018 analysis by NTS, using public data, shows that 33% of active salmon farms are sited inside protected areas.

The SAMS report highlights how little information there is for making safe decisions, for instance on the impact of aquaculture’s waste and pesticides on Scottish species and protected habitats, especially as they affect the environment in combination, cumulatively and at low levels over wide scales. Many Priority Marine Feature species are mobile. This does not mean they can escape these impacts by swimming away – they may live in the only places suitable for them.

Where data is lacking, Adaptive Management not withstanding, Scotland is legally obliged to apply the precautionary principle: ‘…a lack of full scientific evidence must not postpone action to protect the marine environment’; however, the report and hearings show how rarely it is applied.

As Prof. Callum Roberts, Environment Department, University of York, has put it: ‘The multiple environmental problems associated with open-cage salmon farming have been thoroughly established by scientific research. Individually, many impacts represent a serious risk to the conservation objectives of marine protected areas. Collectively, they guarantee an unacceptable level of impact. Open cage salmon farms are incompatible with the conservation objectives of protected areas and should not be sited within them.

Additionally, the industry harms its own interests by shooting seal and disturbing cetaceans. Both should stop & ADD use must be discontinued until they are proven not to harm or displace cetaceans.

FHI stated that many farmed fish are killed by ‘unpredictable’ harmful algal blooms (HABs), triggered by nutrient imbalances and warmer seas. In 2016, HABs caused an ecological and economic catastrophe in Chile’s aquaculture region2, while many smaller HABs have closed valuable Scottish shell-fisheries temporarily. Climate change makes HABs more likely. The report suggests that all HABs are natural events, triggered far offshore, but we should be really concerned when it states that 5-10% of the dissolved nitrogen in the Minch is already derived from salmon farming, and (Heath et al (2002)) that nutrient inputs from aquaculture might be more than 80% of the land-derived inputs on parts of the west coast. Astonishingly, a SAMS/PML project (MERP) intended to inform policy decisions by modelling everything from the west coast’s nutrient inputs to its commercial fish outputs, includes no inputs from aquaculture, even though they will reach 500,000 tonnes by 2030 (more than the untreated sewage of half of Scotland’s people (just solids)) plus twice that quantity in dissolved nutrients3, also doubling by 2030.

The SAMS report omits salmon farms’ visual impact on the environment, for which SNH is the statutory consultee. In Visit Scotland’s annual surveys, tourists consistently cite the landscape as their main reason for visiting. Tourism is worth about £11bn to Scotland, rising to c. £23bn by 20254. It provides at least 100 times more jobs than those directly employed on fish farms, many in small coastal communities. You will never see fish farm cages in tourism photo adverts, and this is why:

In one SARF-funded study5 48% of respondents said expansion of fish farming would negatively impact the scenery and 46% said it would negatively impact the natural environment. 25% did not want to see an increase in the number of fish farms. Over a third didn’t want to see them get any bigger and 10% said they would be less likely to visit these locations.

SARF’s members include SEPA, MSS, Crown Estates, SNH & SSPO. The SAMS report barely mentions another SARF-funded study, whose unwelcome findings were obscured by the industry and the Government agencies that regulate it. SEPA commissioned the ‘PAMP 2’ study6,7 into the environmental impact of emamectin benzoate (EMB), a pesticide fed to fish to combat sea lice. It breaks down so slowly that some can remain acfive in the seabed for 4.5 years, according to SEPA. The study found a correlation between EMB use & a c. 60% decline in crustaceans, at sea loch scales and at levels too low to detect with the process used for standard monitoring. SARF members agreed not to publish the study until the criticisms of five anonymous reviewers had been included. These focused on the limitations of the data collection methods and stressed that correlation doesn’t prove causality – one suggested it might be caused by the deposition of fish farm waste instead.

PAMP2 was published quietly online. SEPA made no public comment but their aquaculture specialist produced an internal ‘response options’ report (FOI8), stating, ‘…the waters in which salmon farming is practiced are usually  the  same  waters  in  which  Scotland’s  valuable  crustacean  fisheries  are located…it is not tenable for SEPA to adopt a position where commercial shellfish species are impacted by the day-to-day activities of fish farms, activities which SEPA will have knowingly authorised under CAR… SARF098 reveals that there is a significant risk of failure to provide such protection.’

He recommended a ban, which became SEPA’s policy, but it was shelved just before a public announcement, due to industry lobbying of the Government, according to the Sunday Herald 9. Further  Herald FOIs later exposed that 4 of the 5 anonymous reviewers worked for the pesticide’s manufacturer. Behaviour like this destroys the public’s confidence that regulation is able to control polluting industries and that they will not impact sustainable jobs or wildlife.

SEPA has since issued a lower interim EQS for EMB use in some protected areas, but did not apply it to the CAR licence issued for a farm expansion in an MPA & SAC, with qualifying species, at Loch Duich.

SEPA maintains that all industries’ CAR licences are issued in the same way, but their own ‘response options’ paper states that aquaculture has a unique status, at least regarding toxic chemical discharges: Fish farming is unique in that it is a sector which is allowed to discharge substantial quantities of biocides, some of them Priority Substances in terms of the Water Framework Directive and all at least List II substances in terms of the old EU “Dangerous Substances Directive”.

Aquaculture is also unique in being allowed to release so much organic particulate and dissolved waste into the sea, concentrated at its c.250 active marine salmon farm sites, without having to clean it up, even though ‘polluter pays’ is a legal obligation, as is the commitment to using best environmental practice.

In Norway, even the CEO of the parent company of Marine Harvest has conceded that open cage marine fish farming is not environmentally sustainable10 and that the industry must move to closed containment, encouraged by higher licence fees for the companies who do not invest in this innovation. The newly devolved Crown Estate gives Scotland an opportunity to do the same thing. We are experiencing the same environmental impacts as Norway after all, with the harm detailed in the SAMS report being done to wild salmon and sea trout populations by sea lice, disease transfer and gene introgression, and to seabed and other marine animals through pollution, acoustic disturbance of cetaceans, over-exploitation of wrasse and the shooting of seals.

SEPA’s CAR biomass consenting decisions aim to protect the environment from pollution by predicting how much waste will fall on the seabed & how much EMB it will contain, then calculating the allowable fish biomass, up to a maximum of 2500 tonnes. Crucially, the protection provided by CAR depends on computer modelling.

Regarding the recent CAR application for a 2500t fish farm at Dounie, SEPA confirmed to FoSoJ that it would be assessed using their computer model, AutoDepomod. On the map FoSoJ have submitted as evidence11 you can see the model’s prediction that deposited waste+EMB would fall close to the cages, entirely inside the Allowable Zone of Effect, where SEPA allow most living things to be killed. The tides are strong at Dounie and AutoDepomod also said 99% of the waste+EMB would be swept out of the model’s 1 km2 grid. Once outside it’s gone and SEPA cannot tell whether it would have a cumulative impact, with adjacent farms’ waste. AutoDepomod has another critical flaw – it assumes the seabed is flat, but Dounie is on the edge of an underwater trench. When SEPA used its revised NewDepomod, as an experiment, the distribution was very different, with waste landing up to 500m away, downhill and down current. Even then >80% was lost to the model’s larger 4km2 grid. This lost material would still be ignored by SEPA.

SEPA should review all previous CAR licences issued using AutoDepomod for situations like this, and alter or revoke their licences if necessary, starting with those in MPAs and where PMFs are present.

AutoDepomod was peer-reveiwed, but the many assumptions implicit in how NewDepomod models the transport, deposition and resuspension of waste are still opaque. Astonishingly SEPA admit (FOI12) that its developers at SAMS have not told them of any limitations. NewDepomod’s assumptions and limitations must be published for independent scrutiny before it is used as part of SEPA’s new DZR system.

SEPA have confirmed to FoSoJ (letter received after the hearing & now submitted separately to ECCLR) that Autodepomod was not intended to make accurate spatial predictions but that its thresholds are deliberately conservative, to flag the worst risks, ie Autodepomod helps set emission levels so that all organic pollution is predicted to fall within the AZE or to leave the model’s quite small grid. SEPA don’t consider the impact of the waste that leaves, or mind that the rest trashes the AZE. They acknowledge that AutoDepomod cannot assess large scale dispersion, how the waste is transported, what effect it will have, the cumulative impacts of several farms, or extreme events, which often redistribute waste.

From the letter it does seem though, that SEPA have decided to use the new model for all new applications after all – which is essential, and welcome news.

They should also commit to doing site-specific validation in every case, to always publishing the results, and to setting a low and biomass-limiting threshold on how much organic waste is allowed to leave the model to ‘fate unknown’. SEPA should also explain why Marine Harvest have applied for planning permission for a 3500 tonne farm13 (1000 tonnes above SEPA’s max), with its impacts modelled by NewDepomod, although neither new model or DZR has yet been scrutnised or approved by Parliament?

Scotland has adopted in law the principle that the polluter pays. Killing most life inside each farm’s Area of Allowable Effect is gross pollution, however you look at it, and allowing 80%+ of the waste and pesticide you are modelling to leave a predictive model, unaccounted for, does nothing to regulate or control that pollution.

The SAMS report details many concerns about our ignorance of the wide scale, low level impacts of pesticides. Dilution to oblivion, as a way to dispose of harmful chemicals, was debunked in the 1970s.

Aquaculture is unique in that it releases so much waste but pays nothing for its disposal. Instead it uses the sea – a vital common resource on which so much biodiversity and so many sustainable jobs depend. The industry’s argument that closed containment would be too expensive is in part because it would then have to match the standards applied to industries on land. This is no reason to set aside a legal principle.

Marine Scotland must look harder for evidence of harm to wild fish, eg from gene introgression and disease transfer from escaped farm fish, and it must address the enormous numbers of escapes, which number more than half the entire wild salmon population every year. The level of containment is clearly inadequate if severe storms can cause these regular failures, as FHI told the hearing. Climate change is making powerful storms more frequent. MSS told FoSoJ (p.comm.) that all salmon cages must withstand a once in 50 year event, but MSS rely on the applicants to assure them that they will, and they do nothing to check this or to assess the sites. Existing designs can clearly do nothing of the sort and now the industry is actively expanding into more exposed locations and pressing to be allowed much larger farms.

During the hearing, even the FHI reluctantly agreed that it is unacceptable for the industry to expand while HABs, disease, sea lice and their chemical and physical treatments are killing 25% of its stock.

The aquaculture industry must fix these problems and its proven, unsustainable impacts on wild salmonids, before considering further expansion. Separating farmed fish from the open sea would solve them all. Instead of investing £40m in a wrasse breeding plant that can only fulfil 10% of the predicted demand for sea lice cleaners, why does industry not invest that money in developing closed containment?

The Scottish Government could raise a levy to do the same thing, by varying the cost of seabed leases to encourage firms to move towards closed containment, as Norway is doing – making polluters pay.

Rather than cutting the budgets of all the relevant agencies, the Scottish Government should fund a publicly trusted programme of enhanced and fully independent monitoring, and the analysis of environmental impacts, by imposing a levy on salmon farm profits.

We must stop aiming to be as cheap as Chile and to ‘streamline’ the consenting process14 – Scotland’s farmed salmon should be the best in the world, and so should our precious marine environment.


3 McGarvin M (2000) Scotland’s Secret – Aquaculture, Pollution, Eutrophication and Toxic Blooms. Modus Vivendi for WWF Scotland
5 F Nimmo R Cappell (2011) Assessment of evidence that fish farming impacts on tourism. SARF045 in Aquaculture Research
6 SARF098: Towards Understanding of the Environmental Impact of a Sea Lice Medicine – the PAMP Suite,
7 SARF098C PAMP Refreshment: Wilding TA Hughes D and Black KD (2017). The Association between emamectin benzoate use and crustacean assemblages around Scottish fish farms.
8 SARF098 PAMP Refreshment Study – SEPA Response Options Paper.
SEPA Aquaculture Strategic Management Group. ‘Doc 131.pdf’ in ‘Final Response’ to FOI FO187415 at
9 Sunday Herald , 19 March 2017
11 Image 3:
12 SEPA EIR Response to FOI request FO188192. 24 November 2017.
13 Argyll & Bute Council Planning Application 18/00005/MFF

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s