Environment, Climate Change and Land Reform Committee

Meeting date: Tuesday, January 30, 2018

Official Report 559KB pdf

---

Salmon Farming Environmental Impacts Inquiry

The second item on the agenda is to take evidence on the Scottish Association for Marine Science Research Services Ltd’s report, “Review of the Environmental Impacts of Salmon Farming in Scotland”. I welcome to the meeting Professor Nick Owens, Dr Adam Hughes, Professor Paul Tett, Dr Lindsay Vare and Professor Eric Verspoor. I thank you all for the work that you have done on our behalf for the report. As you can imagine, we have a number of questions to get through, so we will just kick on, if that is okay.

First, can you outline for us briefly the expertise that was deployed in producing the report and the qualifications of the various scientists who contributed to it?

Indeed. The typical way that we do such a review is by assessing what work is needed. We all have very good international and national networks in our specific fields, and we chose to help us people whom we consider to be the best available experts. A number of those people are, of course, from within our own institutes: we particularly sought the advice of colleagues in the University of the Highlands and Islands and other institutes that are in the marine alliance for science and technology for Scotland.

Okay. As we know, the environmental impact of salmon farming is a contentious subject. How did you ensure, and satisfy yourselves, that the report is impartial, in so far as it can be?

The principal way in which we do that is by using the very well established peer-review system. A better way has not yet been found of ensuring complete objectivity and, as near as possible, accuracy at the limits of knowledge. In fact, all the written evidence that we pursued had, in some way or another, been peer reviewed.

That is fine. We will move the questioning on, with Kate Forbes.

I thank the witnesses for their work. The review is an update of the 2002 report, “Review and Synthesis of the Environmental Impacts of Aquaculture”, although with a slightly different focus. However, some of the scientific conclusions and some of the problems in particular seem to be similar to those in the 2002 review. In the process of writing the report, did you identify any significant changes in the environmental impacts of salmon farming since the review in 2002? If so, in which areas? If not, in which areas did you recognise that very little had changed? I realise that there is a lot in the review, so the question is more just about headlines.

I ask Professor Tett to kick off on that.

I would like to distinguish between effects and impacts. The scientific process that Professor Owens has described is designed to tell us whether there is a causal link between salmon farming and changes in aspects of the environment. If the scientific evidence for it exists, we can establish with confidence the link between salmon farming and an effect. However, effect is value neutral, and impact requires evaluation of the effect. That evaluation will depend on the criteria that are applied, which are formal legal criteria and understandings of ecosystem health, but they also relate to societal concerns.

I was involved in the 2002 review. One of the most obvious changes to me is the way in which societal concerns have altered over the period. Looking back at the conclusions of the 2002 review, I agree that not a lot seems to have changed. The first conclusion in 2002 was that waste and nutrients would be unlikely to limit expansion of the industry in the future. It is still the case today that, although we can detect effects on the environment of organic waste and nutrients from farms, they are not of concern in relation to the ecosystem as a whole.

Do you mean that they are not of concern scientifically, in terms of their impacts?

I drew a distinction in that regard. Science can tell us, for example, that the effect of the organic waste from a fish farm is to change the population of animals and micro-organisms in, say, 3 per cent of the sea loch beneath the farm. Then, there is the question whether society should be concerned about that. From an ecosystem point of view—I am an ecosystem ecologist—the answer is no, because we know that sites that are subsequently left fallow will recover, and the areas are only small proportions of sea lochs. In contrast, if a farm happens to be close to a protected habitat and the 3 per cent that is affected includes the protected habitat, that would be a considerable concern. That is where regulation comes in, to ensure that farms are not sited close to protected habitats.

I have a question on that point. Your report mentions that

“in 2003, 16 of 346 operating farms ... were sited above maerl beds.”

Do we have a more up-to-date figure?

We do not, as far as I know.

Maerl beds are protected features. Your report goes on to say that even two years of fallowing does not allow recovery of maerl beds.

You should remember that we were asked to review the scientific literature. There may well be such evidence available from Scottish Natural Heritage, which monitors such situations. All we can say is that no scientific papers have appeared since the paper that is cited in our report.

You are not aware of an update on that figure.

That is correct.

I just wanted to get that on the record. I apologise to Kate Forbes.

I think that you will get similar answers on a number of issues.

To return to the 2002 report, the second conclusion was that the most likely issues that would limit production were

“medicine usage and sea lice transfer to wild populations”,

which continue to be of concern today. Professor Verspoor is more qualified than I am to talk about that.

The third issue was that

“The rate of escapes of farmed salmon is probably unsustainable and represents a major threat to wild populations.”

Again, I will pass over to Professor Verspoor if members want further information on that.

The fourth issue was that

“Changes in fishmeal supply may affect the sustainability of the industry”.

Two decades ago, most of the raw material in fish feed came from wild fish. Today, much of that has been substituted with vegetable protein, although there are still concerns about the supply of fish oil and omega 3 fatty acids.

We will explore those topics during the meeting.

In general, therefore, we have the same set of concerns now as we had in 2002. It is worth remembering that the industry went through rapid expansion in the late 1990s—the early figures I have are that it produced 83,000 tonnes in 1996, rising to a peak of 170,000 tonnes in 2003—and the review was based on scientific information from the earlier part of that period, when production was lower. Since 2003, there has been no upward trend, but there has been fluctuation up and down. Production did not again reach the high of 2003 until 2015, when it got to 171,000 tonnes.

The industry has continued to produce between 130,000 and 160,000 tonnes in the period from 2002 to 2017. In looking at the recent literature, we were looking at a period in which production of salmon was roughly twice what it was when the literature was reviewed in 2002.

In very general terms, and with lots of caveats, there is no published evidence that effects are more widespread or more general now than they were in 2002.

I hear what you say about the early growth of the industry and the current production level of 171,000 tonnes. The prediction for 2030, however, is that 300,000 tonnes will be produced. How robust is the peer-reviewed evidence of what has come before, given the enormous expansion of salmon farming that is anticipated?

I will start on that question: colleagues may also wish to come in.

Extrapolating from today, and without additional mitigation, we would expect more widespread effects. We suggest in the review the additional mitigations that would be necessary: many are already in train or are being considered.

This may be an unfair question, but given the expansion over the past 15 years and, as Mark Ruskell said, what is to come, does it surprise you that the conclusions of the 2002 review are largely the same as those in this review? Does it surprise you that there has not been more scientific evidence published over the past 15 years and—I presume—that there have been very few changes in practices in salmon farming?

I will not comment on practices in salmon farming.

I noticed a pattern in the scientific evidence. In the first decade, many papers were published on harmful algal blooms. In 2002, there was concern that the industry was perturbing nutrient ratios in the sea and causing greater frequency of harmful algal blooms, which had harmful effects on farmed salmon and on farmed shellfish. That led to international reviews, work by the European Commission and other normal scientific work. The review publications all suggested that the cause of harmful algal bloom lies offshore—not in the salmon-farming industry. As a consequence, research into the topic died away.

Will you comment on possible effects of climate change and sea-level temperature changes?

That is the big question of our age.

It is possible to see changes in ecosystems in the west of Scotland. It is hard to understand what is causing the changes—whether it is climate change, fish farming or other human activity, including disturbance of sea beds through fishing and removal of top predators from the food chain. We do not have enough information to be clear on that.

In summary, then, there is a multiplicity of potential effects. It is a dynamic situation that will inevitably change constantly, with fish farming as part of it.

Yes. We need to accept that natural ecosystems fluctuate of their own accord, even without human influence.

Indeed.

On top of that, there is recovery from ice ages, human-induced climate change and a number of other human pressures. Those are changes that take place over decades.

We need long time series information—first, to understand what is happening, and secondly to stand a chance of being able to correlate what is observable with what human pressures on the sea have changed.

10:00  

We are collecting long time series information, and the routine monitoring of salmon farming is providing a lot of data, but I advocate that more attention should be paid to synthesis and analysis of that data in respect of what is changing in ecosystems, which does not seem to be possible at the moment.

That appears to be a gap in the research since 2002.

I agree; we are not doing as much of that sort of research as we did before 2000.

I will follow up that point with a layman’s question. Could you quantify for us the scale of the task if we were to address those points? How big a job would it be and how long would it take to get a body of robust science to inform our understanding and get on top of the situation?

That would depend on the particular aspect of the ecosystem and the particular area of concern. I can speak to benthic and pelagic impacts, which are not of concern at the moment. In terms of benthic impact, a very large amount of data is collected from routine monitoring of lochs that contain fish farms. A relatively small amount—a few person-years—of continuing activity is needed to put that data together with information on other causes of change.

If we were to embark on an extensive piece of work over the next five years and you were to come back to committee in five years, how much more confidence would you have—if the work had been put together properly—in coming to conclusions about the impacts of salmon farming on the environment? Is it a five-year or a 10-year job? Those are layman’s questions, but we need to get a handle on the matter. How far away are we from really understanding the issues?

I will try to answer that: it is a fascinating question that we are struggling to answer because it is not particularly specific, but I will try to help.

I have just returned from a conference at which we talked about observation of the global ocean. In order to pick up some of the big ecosystem changes that are happening because of climate change and so on, work needs to be done over decades and by whole nations. In Scotland, that would mean doubling or trebling our current efforts to get a better understanding of the natural changes and the climate-change induced changes versus the changes that are much more local. My sense is that we probably need a decade of really intensive work: we would have to up our game by some order of magnitude. That was certainly the conclusion of the conference that I was at recently. Whole nations need to invest very seriously. We are talking about a very specific issue in fish farming, but the ocean has a considerably wider impact on the whole of society, particularly in Scotland.

That is useful.

In terms of the scale of the problem that you define and the order of magnitude of increase in research that is required to reach a definitive view of the whole picture, is there any part of the picture in which you perceive a problem that is in much need of research and could be addressed, and that we in Scotland should focus on?

Yes. There are probably two key areas identified in the report. There is the sea-lice issue on which SAMS is beginning to embark on some tractable work that will help, but that work needs investment. Professor Verspoor is better able to comment on that than I am.

The question of organic material is also interesting. Paul Tett knows better than I do about it, but I think that we could do more research in that area.

It seems to me, from the little that I know, that an engineering solution is required to capture organic material by harvesting it—scooping it up, taking it away and doing something intelligent with it.

Perhaps Adam Hughes can respond to that.

There are certainly engineering solutions out there. The Norwegians are leading the technology in that area with closed-containment systems from which sludge is removed. That engineering solution, however, obviously has an economic cost.

We will explore that subject in greater detail as we move along.

We mentioned maerl beds earlier. Are there any areas of environmental impact that the report does not address? There may be impacts that you are not aware of, but are there any that you are aware of that are not covered in the report?

We began with a process of trying to identify all possible environmental effects by drawing up a matrix of pressures caused by the human activities that are involved in fish farming. Against that, we tabulated areas of ecosystem function from the marine strategy framework directive. I would say that we picked up all the major issues, with the exception of plastic, which is of current concern but on which there is very little literature relating to its effects in Scottish waters.

The report is quite comprehensive and it takes some time to get through. What are the top three environmental impacts of salmon farming in Scotland and, on the back of that, what are the likely outcomes of an increase in salmon farming in relation to those three top concerns?

I refer back to my distinction between effects and impacts. The effects have been established from reviewing the scientific evidence, whereas the impacts, to some extent, depend on judgment. You might get a different answer about impacts even from the different experts who are here today, so I hope that I am not the only person to reply to your question.

I am a systems ecologist, so I am interested in the health of ecosystems as a whole rather than individual populations. One of the top two issues for me is the global impact of getting the ingredients for fish feed. Going to 300,000 tonnes of production will clearly increase the demand for ingredients. Scotland will be in competition for that with Norway, which is talking about going from 1.5 million tonnes to 5 million tonnes; there is also Chile and other world industries. That means that there are global issues around the impact of demand for fish feed on land use if most of the protein comes from terrestrial sources. There will also be an impact on fish stocks if we still need to get fish oil from marine sources. That is my top issue, because of the scale. Of course, Scotland is playing only a part in that global demand.

The second issue, for me, is the low-level and long-term effects of chemicals on the environment. That is of concern because we do not know enough about the long-term effects. We have a good system of regulation, which involves environmental quality standards that set maximum tolerable levels, but in recent years the long-term protection of those standards has been questioned. We do not know enough about this area, and it could affect ecosystems as a whole through harming their essential components—from the small animals that live in the sea bed and bring oxygen to the sediment by burrowing through it, to those that live in the water column and are an essential part of the food web.

I have given you only two issues. I understand concerns about the impacts of farming on wild salmon—both the impacts of sea lice and those of the effects of escapes and genetic transfer. I do not see those as a threat to ecosystems as a whole, but my colleagues might differ.

Whether something is a concern depends on which sector people are in: some will see certain things as a concern and other people will see others. As a scientist, in trying to answer such questions my main concern is the lack of the information that one needs in order to do so. In my area, which is Atlantic salmon, farm-wild interaction relates primarily to genetics. In the period from 2002 until now, specific knowledge about the level of interbreeding between farmed salmon that have escaped and wild salmon has progressed very little.

We might ask why that is. I think that it is because investment in gathering such information has not happened. We are very far behind Norway. If you are asking what levels of funding are required, I suggest that members go to Norway, where they will see them; I have had indications from colleagues that their spend on sea lice research alone is larger than the entire budget for all research on farm-wild interactions here. As a scientist, that is my concern. The literature is incredibly sparse, particularly with regard to Scotland and Ireland and anywhere else outside Norway.

Who funds the scientific research in Norway? Is it the Government or is the sector required to fund it?

I am not 100 per cent sure of who funds the total spend, but the industry contributes and the Government puts in a great deal of money through various departments. One example is that, a few years ago, there was a budget of £23 million just for sea lice research.

Okay. Mark Ruskell has a question.

I want to return to the issue of environmental impacts that are perhaps not covered by the report. There have been very well-publicised incidents in which dead salmon have been transported on the roads and there have been biosecurity issues, with waste leaking out of trucks and potentially getting into watercourses. To what extent do you see that as a problem? Do you see the increasing expansion of the industry causing any issues for the welfare of the fish?

Perhaps Adam Hughes could talk about the welfare aspect, at least.

Where there are large numbers of fish deaths, a robust system of disposal needs to be in place. That is really difficult, because a lot of the locations are remote and the events might occur only once in five or 10 years. Planning for them will therefore be difficult and expensive, and it will need to be proportionate to the risk. At the moment, work is being done on the process of dealing with large-scale fish kills, but I do not know what stage it has reached as regards evidence.

I am sorry—what was the second question? It was about welfare.

I asked whether you see specific biosecurity issues with regard to leaking fish waste getting into watercourses.

All salmon producers take biosecurity very seriously, and I suggest that the committee talks to them about the plans that they have in place. I cannot comment on individual cases that have happened up on Lewis recently.

On fish welfare, there will always be a proportion of society who are uncomfortable with the farming of animals and fish. I believe that the welfare standards are very high at the moment, but that is a personal opinion. As the industry expands, there is no reason for those welfare standards to degrade. The expansion is based on current best practice. It is for society to decide what it is acceptable in terms of its food production and animal welfare.

10:15  

Is it acceptable that we see 25 per cent higher mortality rate within the livestock? Only 75 per cent of farmed salmon actually make it to market and 25 per cent are dying. Is that comparable with other production systems, such as chickens or pigs?

I have no idea whether that rate is comparable to that of other production systems. I do not know whether those figures are across the board for the industry and whether it is acceptable. As I said, that is for society or yourselves to decide.

How does that mortality rate compare to those of other countries?

I do not have those figures, I am afraid. Compared to places such as Chile, which has had big problems with disease for the past three or four years, I think that Scotland probably has a lower mortality rate, but I do not have the figures.

If you were able to source those figures in due course, it would be useful for the committee to have them. Norway would be a particularly interesting comparison.

I have a supplementary question on the earlier point about the Norwegian research. The SAMS report says that there is no specific data for Scotland but you have looked at studies from elsewhere. What are the restrictions or limits on applying the Norwegian research to the Scottish environment?

On sea lice or genetic interactions, the research is accessible and relevant in that it informs the potential for impacts and often shows the degree of impact. However, it also shows that the impact can be very local, unpredictable and dependent on local circumstances such as the sea loch environment, the direction of wind and that type of thing. It is therefore difficult to say whether we could predict what the situation will be in a given location. That is very much where you have to have local information and that is generally what is lacking, at least in the public domain.

The review is based upon what is in the literature and what is accessible, and not just on raw data but what has been analysed. There is very little out there. There have been studies but they are not systematic and, in most cases, they are not up to date. Transferring from Norway to here gives only a general idea of the potential for a problem. We need to collect information that shows what is actually happening.

Given the lack of information and available science that we have discussed today and which is touched upon in your report, I am struggling to see where the precautionary principle has been applied in allowing this sector to expand in the way that it has. Is that harsh?

There have been attempts to find a way to work together on this. I recommend that you contact your counterparts in Iceland and look at what they are doing. Iceland is expanding its fish farming industry and it has introduced a new regulatory framework that takes a learn-as-we-go approach. It is a neutral framework, in that if you have evidence that there is no impact—in contrast to there being no evidence for impact—the industry will be allowed to grow. If there is evidence of impacts, the industry may have to contract. That is a flexible and adaptive system. However, it is contingent on the information being collected and in that respect the onus will be on the industry to contribute to the monitoring of the environmental parameters, such as sea lice levels and genetic introgression.

How important is the difference in the sea temperatures in considering the research from Norway and Iceland? My very limited knowledge of chemistry leads me to believe that, as a rule, warmer temperatures make things happen quicker. Although we are not measuring those impacts, might the problems be worse in our waters, given that they are warmer than those farther north in Norway and Iceland?

I can try to answer that. What you say is true: the rate of biological reactions roughly doubles with every 10° increase in water temperature. However, the Norwegian coast is extremely long and fish are farmed all the way up it. In the southern part of Norway, water temperatures are not very dissimilar to those in the west of Scotland. The water is colder in the north of Norway, but it also benefits from flows of warmer water from across the north Atlantic—northern Norway is not as cold as one might expect.

There are differences. For example, in terms of lice control, the Norwegians seem to have a preference for using lumpsuckers, which are cold water fish, to eat the lice, rather than wrasse, which are warm water fish that are used for preference in Britain.

Would the water temperature affect the breeding of the sea lice? Would they breed more quickly and readily in our warmer waters than they would in Iceland?

Yes, the lifecycle would be shortened.

Typically, a salmon farm has a two-year cycle, with maximum stock held in the summer of the second year. That is the time when the fish are growing the fastest because the water is warmest and their metabolic activity is greatest, so it is in summer that they would have the greatest local environmental impact, through excretion.

As water temperatures increase, it is likely that metabolic activity will increase. I am now talking about climate change. We have seen an increase in temperature of about 1° over the period that I have been working. That will have a small effect on metabolic rates. The same things will apply to the sediments.

That issue is coupled with the solubility of oxygen in seawater, which decreases as the water gets warmer. Cold water is better for salmon, because it is higher in oxygen than warm water. That is one of the factors that favour the growing of sea bass and sea bream under Mediterranean conditions.

Given the projected increases in farmed salmon and your answers on the top two or three environmental impacts, can you tell us what role alternative approaches and technologies could have on those impacts?

I guess that you are referring to recirculation systems, which are closed containment systems that take the production out of the environment, giving producers much greater control over such things as biosecurity and where the effluents go and how they are treated. There has been interest in recirculation systems for a long time and we are beginning to see one or two commercial salmon recirculation systems here in Scotland and in Norway. The technology is coming online. It is a question of economics and the cost of production; it is about the cost of capital expenditure—it is much more capital intensive to build an onshore facility, although they are now moving to offshore closed containment systems in Norway—versus the running costs and the environmental benefit.

Recirculation systems are a technology that has been coming for the past 10 or 15 years and it is still on the cusp. It will have environmental benefits. No food production system is without environmental impact, and there are other environmental impacts associated with it. It is just a question of economics and of whether it is cost effective to produce the salmon in closed containment systems.

Societal perception is also an issue. Work has been done to ask consumers whether they think that fish farmed in a closed containment system are more environmentally friendly than fish farmed in an open-water system, and the consumers believe that the open-water system is better because it is more natural, so there is greater consumer acceptance for cage farming than there is for recirculation systems.

What is driving the change towards more containment? Is it driven by economics—in terms of increased production, because of fewer losses—or is it driven by regulations relating to environmental protection?

My personal opinion is that the main driver is that it allows better control of the production cycle through biosecurity, with fewer losses and a better prediction of what the end product will be. It makes it much easier to control the environment, so producers can have a better idea of the product that they will get at the end of the production cycle.

I would like to ask briefly about the capex and the economic costs. How does Government support in Scotland or the lack of incentives compare with the situation in Norway?

I am afraid that I do not have the answer to that.

That is fine.

We will look at the sea lice issue in more detail.

I was involved, as a complete layperson and with some trepidation, in consideration of the Aquaculture and Fisheries (Scotland) Bill, so rather than dipping into the areas on which the committee may want to make recommendations, I want to start with the science, which is what you are here for, and with where the science has got to and what would be useful for the science in the future, to inform us further. However, I am tempted to highlight a quotation from your report, which states at section 2.1:

“Sea lice ... are ectoparasites and a key impediment to the expansion of the Scottish salmon farming industry in the marine environment.”

You know that, of course, but I wanted to read it out for the record. Perhaps I should have held back on that, but there is concern about that issue not only among the public but among scientific communities globally as well as in Scotland. Let us start with the science, please.

What is your specific question about the science? What aspect of it do you want us to address?

Each of you could cover whatever it is appropriate to cover—for example, the effect on wild fish; any views highlighted in the literature about welfare effects on farmed salmon; the trigger levels of sea lice on smolts, post-smolts and adults that might be appropriate for action in view of the science; or transparency and whether it is appropriate for the scientific analysis to be more publicly available. I lodged an amendment in 2013 that addressed that issue on a farm-by-farm basis, but it was rejected. I also invite comments on any other aspects of the science that it is appropriate to comment on. I am simply trying to open up the discussion, because I and other members have questions on this important issue.

10:30  

Professor Verspoor can answer the questions about the impact and levels of sea lice. I will follow up on that by providing some information about and thoughts on ecological modelling and how we might make a step-change in improving that.

I would also like to ask whether there are any figures for the volume of salmon that is being lost specifically to sea lice annually.

I do not know the answer to that. Professor Verspoor might.

I believe that there are estimates of that number. However, this review focused on the impact on the wild populations of sea lice from farms—that is the literature that we reviewed—and we did not cover exactly what is going on in the fish farms, so you will have to forgive us if we do not have those particular figures. The science tells us that an excessive sea lice burden on wild salmon can have a negative impact on their survival.

You say “can”, but the research literature that you highlight in your report says that more than 11 sea lice on a smolt or post-smolt would lead to mortality, so there is scientific evidence on that specific point. Perhaps you are coming to that.

I am somewhat hesitant to say that mortality will result from two sea lice per 10.3g of fish. It will depend very much on circumstances and will vary with the various factors affecting the fish, such as its size. The science would probably suggest that, if you increase sea lice burden on a wild fish for a significant amount of time, you will increase the likelihood of there being a negative impact on its growth or survival. Sea lice do not benefit salmon.

The extent to which sea lice are attached to salmon will depend on the local circumstances, such as the time of the year and the environmental conditions. You can say that there is a risk and a potential problem. However, whether it is a concern in a given location is another question. In Scotland, a location that I am familiar with but whose identity is not in the public domain has had historically high numbers of salmon even though those salmon have to go past a very large salmon farming operation. It would be difficult to say that sea lice are having a negative impact, given that the numbers of salmon in that location are at historical highs. However, there is a large number of confounding factors. For example, the river is stocked; does that account for the historical highs? Is it perhaps the case that the marine conditions where those salmon happen to go have been particularly good? Not all salmon go to the same place in the ocean—they do not all have the same journey. It is difficult to take information from one controlled study in one location, under a certain set of environmental conditions to do with temperature, feeding regimes and so on, and extrapolate that in a way that would allow us to say how many sea lice per gram of fish would be critical or would be an indication of how big the problem is.

I would think that there is also a significant issue about the amount of sea lice that are on the salmon in the farm that the wild salmon pass.

Absolutely, and you did ask about access to data. If we are going to do good science on these questions, we need access to data. As far as I know, that data is being collected to some degree, but it is not generally accessible. It is not universally in the public domain for scientists to analyse and say what it tells us. Is it sufficient data to answer the questions? If it is not, how should the data collection be altered to improve the data so that we can say more about, for example, what the impact might be or how the sea lice might be controlled? We were asked to review the literature, but that information is not in the literature, so we cannot comment on it.

Can you comment on how transparency and the public openness of real-time data on a farm-by-farm basis could help develop the science as we go forward?

It would help advance the science, but I understand equally that such information is misused on both sides of the fence in the debate. The science takes a long time to crunch the numbers and assess the implications, but when data is made public there are often knee-jerk reactions over what it means or does not mean, which is an obstacle. However, I accept that, in principle, if we want the best science to inform the debate, we need transparency about the available information.

But surely how that information would be used by anybody beyond the scientists would be the same as when, on land, if there was a discharge into a burn, the Scottish Environment Protection Agency will release information publicly. What is the difference?

I am talking about the misuse of the data by certain sectors.

But is that a reason not to have the science made public?

No, but I am just saying why that is not happening.

Okay. I am misunderstanding what you are saying.

I am saying that the science should be made public because if we want science to inform the debate, we need everything to be accessible ultimately for analysis.

Thank you. That is helpful. Can we go to Professor Owens, please?

Certainly. That is a pertinent point on which to come in. One of the things that we are in the business of doing is modelling the distributions of sea lice. We combine a biological model with an ecosystem—

Is that the same as biophysical modelling?

Yes. One of the limitations that we have in getting a predictive model as to where sea lice will be transported to from fish farms and so on is the availability of live, real-time data. If we had farm-by-farm data on the distribution and numbers of sea lice in the cages, we would have a considerable improvement in our predictive capabilities, certainly of the distribution of the sea lice. We then move into a biological question as to what might happen to those sea lice and the impact that they might have on the salmon.

To take the example that we have just heard about, one of the possible explanations for the apparently high levels of wild salmon in an area where there are cages is the very fine-scale variations in the distribution of the sea lice coming from the cages versus where the wild salmon are travelling. We do not have that information because we do not have the real-time data.

Thank you.

The science of the impact of sea lice on salmon is perhaps something that we do not hear being discussed in the pub on a Friday night. However, when “The One Show” ran its two evenings of programmes on salmon fishing in Scotland, I think that people were horrified to see the amount of dead salmon that was being put into lorries and shipped halfway across Scotland. There is therefore a genuine need for the public to know about the issues.

As has been said, the report does not highlight the disease in salmon farms and does not go into the detail of why so many fish are being slaughtered. I know a local farmer who has a dairy herd; if 25 per cent of his cattle were being slaughtered every year because of disease, that would mean that there was a serious problem across the industry. As Mark Ruskell said, the dead fish are being transported around Scotland.

Regardless of whether data can be misused depending on who has it, is there a lack of farm-by-farm data on the amount of disease that is in the fish stock, of which 20 to 25 per cent is being slaughtered? If fish production has doubled since 2002, what has happened to the number of fish that are slaughtered as a result of disease? Do we just keep compensating for that? Do we double production over the next 10 to 15 years to compensate, or should we do something to tackle the reason why the disease is there?

That is outwith the remit that we were given. The people who can answer your questions and who have whatever information is available on the issue are Marine Scotland science, the fish health inspectorate and those who are associated with them. I would direct those questions to them.

We dealt specifically with the impacts on wild stocks and not with what you referred to. We have no access to those organisations’ databases, so I cannot say what information they have.

If you are to look at the disease interaction between farmed salmon and the wild population, should you have access to such data? If the level of disease continues to increase and is compensated for by producing more, does that create more risk for wild salmon in our rivers up and down Scotland?

Yes—that is potentially the case. However, that area of work involves Marine Scotland science and the fish health inspectorate team. Information on the subject has been published—scientific studies are referred to in the report—but the level of information that you ask for is not in the public domain. You would have to check on that with those organisations.

We will have the opportunity to do that with the witnesses who appear before us next week.

I return to sea lice burdens—given the lack of published information, we are in the realms of conjecture, so I will invite you to speculate. Am I right in deducing from what you have said that there are a number of key variables for the likely sea lice burden? They include the genetics of the salmon, whose ability to fend off or absorb sea lice varies; the feeding regimes, as what the salmon are fed might attract sea lice; temperatures; and the water conditions in which the fish find themselves. Am I right in deducing that, given those variables, if we were dealing with a differential equation, there would be so many variables that it would be almost impossible to solve? Do you have to contend with those variables in arriving at a conclusion?

Yes. The Norwegian research programme perhaps gives the best understanding of the effect of a lot of the variables that you mentioned. That research has also considered whether there is an association across Norway between the health of salmon populations and the levels of sea lice. In general, it has found that there is an association, in that the higher the sea lice levels in an area, the larger the effect on the wild populations. It is correlative, although there are exceptions, which are probably because of the sort of local circumstances that we have discussed.

10:45  

I have one final question on this issue. The report says that

“sea lice populations ... appear to be developing”

widespread and serious

“resistance to many existing treatment medicines”.

Is there any Scotland-specific evidence of that, or are we talking about global evidence?

It is global evidence.

So that is another gap in our understanding at a Scottish level.

It appears to relate to Scotland as well. Section 2.1.4 of the report states that the treatment methods

“appear neither to be succeeding in controlling sea lice, nor capable of addressing the environmental effects of the lice.”

Sorry, I was addressing the question whether the evidence for the developing resistance of sea lice to chemicals is gained from Scottish work.

Okay.

Let us look at the discharge of waste nutrients.

Information is provided on what the literature says about environmental quality standards for emamectin benzoate. What does the review suggest in relation to possible changes to EQSs for EMB in Scotland?

I will start by explaining a little bit about emamectin and how it is used. It is an in-feed treatment that is supplied in salmon food. It is carried in the blood of salmon and gets to the sea lice in that way, and it certainly damages their growth—think of it as a systemic insecticide. It reaches the sediment in the faeces of the salmon and can then penetrate into the food chain if animals on the sea bed eat salmon faeces or eat the bacteria that have eaten the salmon faeces.

Environmental quality standards to ensure a minimum safe level of the chemical emamectin are set by the regulators in several different respects. One is in respect of anything that is to be eaten by humans—for example, an EQS is set for mussels that are intended for human consumption, in order to avoid emamectin getting into the human diet. Lower concentration levels are also set for the material in sediment, in order to protect animals that live in the sediment. In that case, the EQS is based on laboratory experiments with a number of test animals—the marine equivalent of white mice. The test animals are things that can be grown under laboratory conditions and are therefore pretty robust. Basically, the experiments involve determining the minimum dose of the chemical that is necessary to have a harmful effect on the test animal and then working out what concentration has no effect on the test animal. That gives us the NOEC—the no observed effect concentration. It is then necessary to introduce a precautionary factor, because the test animals are very robust. The precautionary factor might be 10, 100 or 1,000. The aim is to introduce sufficient precaution so that we can rely on the EQS to provide adequate protection of the animal community that lives on the sea bed.

That is now seen to raise two issues. The first is whether we are introducing enough precaution into the EQS and whether we know how sensitive certain animals are. Secondly, when we are developing the EQSs, we talk about the direct effects on particular animals, but are there more general, diffuse and long-term effects on ecosystems, such as on the behaviour and reproductive capacity of animals, that will not show up as mortality but will interfere? It has proven difficult to get evidence on that other level.

The reasons for that include the necessity of relating sediment concentration of emamectin to the state of the animal community in the sea bed, with which there are two particular difficulties. One is that most of the monitoring surveys of emamectin are not sensitive enough to measure the levels that might be causing harm, although recent improvements in techniques have begun to remedy that. The second issue is that samples for chemical pollutants in the sediment are taken in different places and at different times from the samples for the biological content of the sediment. Therefore, it is difficult to do a reliable statistical analysis of the relationship between the emamectin content of the sediment and the biological content.

One of my colleagues has attempted such an analysis, which is published in a report by the Scottish agricultural research forum. It suggests that there are lochs where levels of emamectin are being detected some way away from the fish farms and that that correlates with changes in the community of animals in the sea bed. From a scientific point of view, the confidence that we have in those conclusions is only moderately strong. The statistical analysis is the best that can be done but it is limited by the available data.

This is one of the research areas in which there is a need to do a specific, probably long-term investigation of a few sea lochs, including research studies of the level of these chemicals in the sediment and of changes in the communities of animals, including the effects of those potential changes.

Another factor that compounds the medicines’ effects is the organic input from the farm. A big farm will have a lot of organic input and will use a lot of medicines, and it is difficult to distinguish between the two.

The two might interact.

Yes. I suppose this is an area—

This is an area that requires further research. From what I read in your report, it appears that the breakdown of evidence shows that the distribution of emamectin is almost linear in distance from the cage. That might be obvious, but I am more concerned about the effect on other species of the breakdown of ivermectins in the sea—the impact that it might have on other sea life. You are really saying that it needs more research to even begin to measure that.

Is the breakdown of ivermectin, or whatever it is properly called—

It is called emamectin.

I should declare an interest as a land-based farmer. Is the breakdown process clear and well-defined? Does it break down into components that are not dangerous?

It eventually breaks down into non-dangerous components. I do not know of research on the breakdown products, but there is research into the breakdown time, which is typically approximately half a year, although it seems to vary between sediment type and condition.

The sediment type can have an effect on the breakdown process, and the sediment type is a function of what the fish are fed on in the first place as well as what the sea bed is composed of.

Exactly. One of the conclusions that I have come to from looking at all the papers that have been published is that there is no standard sea loch. There is a wide variety of physical types with a wide variety of sediment types. It is clearly desirable for each farmer to understand the local conditions, which might favour rapid breakdown of the chemicals in some cases, or their retention in others. There is little published information about the spread of conditions.

Nevertheless, it would be an area hugely worthy of further investigation to enable guidelines to be provided on suitable locations for future fish farms in respect of the composition of the sea bed and its sludges, rather than just species.

I agree strongly. We have a precedent in the locational guidelines that Marine Scotland brought out in 2002—and updates regularly—which consider effects on the sea bed and the water column. They could be expanded to take into account variations in sediment quality.

I may have already asked this question, but are there any data and analysis gaps relating to the discharge of medicines and chemicals into the environment? If so, how might those gaps be filled and what would the benefits be?

I had not read that question. It appears that you may already have answered it.

I answered in relation to emamectin, the sea-lice treatment medicine, but there are two other categories of chemicals, the first of which is antibiotics. Our evidence is that antibiotics are little used in Scotland.

I am very pleased to hear it.

The reason for that seems to be the vaccination of fish against disease. The other category of chemicals is those used in anti-fouling—the paints that are applied to farm structures and the steeps for nets that are used to prevent seaweeds and barnacles from growing on them. They are used not only on fish farms, but on any moored structures, such as sailing boats.

As you may be aware, over the past 20 or 30 years, there has been a big change in the nature of anti-fouling compounds. In the late 1980s, it was discovered that tributyltin had very harmful effects on molluscs—it caused them to change sex, which had an unfortunate effect on shellfish farms. That discovery, along with other evidence, led to a major change in the kind of anti-fouling chemicals that are used.

We now use compounds based on copper and zinc, with some organic ingredients, and, as far as we know, those are less harmful. However, there are some indications that we do not know the adequate environmental quality standards for some of those chemicals in respect of ecosystem function. That has not been raised as an area of concern so far, but we need to keep it under review if the industry expands—particularly if it expands in large, off-shore floating structures, which will need anti-fouling treatment.

Of those three issues, which is the key one towards which Government money for research should be directed?

At the moment the key issue is the direct impact of anti-lice chemicals.

We are told in the report that the DEPOMOD model is now thought only to be accurate to 63 to 85 per cent, which is somewhat at odds with an original accuracy estimate of 13 to 20 per cent. I ask the question as a layman, but from a scientific perspective, with how much concern should we view that difference?

Are you talking about the fact that it has been considerably improved?

Yes.

We should be pleased that it has improved. We are still working on the models to make them even more accurate and useful.

That was my clumsy attempt to get that point clear. I read that part of the report three times and two different ways. I now understand that it is an improvement.

Yes, it is an improvement.

I am glad that I asked that question on the committee’s behalf.

My question comes from my experience of using ivermectin in cattle and the resulting very slow breakdown of animal dung. When the fish equivalent is used, is the impact of withdrawal on breakdown of the sediment considered? Has any work been done on how quickly the sediment breaks down the use of the pesticide? Should there be?

11:00  

I will try to answer that. In general terms, the rate at which fish faeces, or any organic inputs, break down depends on bacterial activity. That in turn depends on the rate at which seawater containing oxygen can get into the sediments.

One of the key roles for the larger animals that live in the sea bed—the worms and the prawns and so forth—is to burrow into the sediment to rework it and to let a flow of water in. If those animals are harmed and their activity slows down, the rate of reaeration of the sediment will slow down. The rate at which the waste material breaks down will slow and the fallowing period will have to be longer.

So should we bear it in mind that if those chemicals stop being used, the sediment is likely to break down far quicker and there will be an environmental impact?

We cannot say that we know that. It is possible to turn it around to say that, if those chemicals are widely affecting the macrobenthic organisms—the reworkers, such as the worms—over the base of a loch, the general rate of breakdown of organic material will slow down.

We will now cover the discharge of waste nutrients and their interaction with the wider marine environment.

I want to linger a bit longer on the chemicals and ask about cocktail effects. Is that an area where there is a research gap? Are there proposals from the Scottish aquaculture research forum or elsewhere to study the interactions of some of those chemicals?

The simple answer is that it is a research gap. I do not know whether there are any new proposals for research.

On nutrients, my understanding is that SEPA is feeding into a sector review that will look at revised environmental quality standards for emamectin. The review will look at introducing a new depositional zone regulation—or DZR—that, on the face of it, could allow the industry to expand but could also increase environmental compliance.

What are your thoughts, particularly on SEPA’s DZR proposal? It is something that we have known about for some time, but it has not come to the committee yet. How does it reflect on the research base that you have been looking at?

The changeover to DZR has allowed a review of the current way that the fish biomass is consented for a site, and that is to be welcomed.

The prescribed maximum limit of 2,500 tonnes of salmon per site had no real basis in evidence; it was an arbitrary figure. The DZR will allow a more adaptive and responsive management of the biomass, which will be allowed to increase or decrease depending on the impacts on the benthos.

Scientists do not have any clear understanding of the detail of the mechanisms behind the DZR. The proposal has gone out for consultation and we do not have the results. It is difficult to say whether the scientific evidence supports a move to DZR, because we do not know what such a move would mean.

Do any other members of the panel have views on that?

I do not have anything to add.

Perhaps I could pitch it in a slightly different way. You have already mentioned the regulatory regimes in Norway and Iceland, which are very much focused on achieving environmental objectives. There has been a consultation on DZR, so you are aware of what is coming, in broad terms. How does DZR compare to regulatory regimes that are focused on delivering environmental objectives first and foremost?

Could I return us to the topic of adaptive management versus the precautionary principle? Adaptive management is learning by doing. It allows development to go ahead without there being absolutely clarity on what the environmental effects will be. It assumes that the environmental effects will be monitored and that knowledge of those effects will change management practice as necessary.

One can see the change to DZR in that framework. If it allows or encourages the industry, particularly farms, to monitor the condition of the sea bed in such a way that management practice changes, it will be successful. My understanding is that that is what is meant when people say that it is preferable to set the standard and then allow a fish farm to find its own way of achieving that standard, rather than trying to regulate by setting the maximum stock that can be held at a site.

If I am heading in the right direction, I would like to go on a little more with my answer.

Yes.

The important question is: what are the circumstances in which adaptive management can be properly implemented and will succeed? That clearly requires some changes in the way in which we all think about it, including changes in the way in which the public think about what the regulator is required to do. If the regulator is seen as being in a policing role—there to enforce specific regulations for that stock—that will set up a confrontational situation.

There is a degree of adaptive management at the moment because, in my experience, regulators talk with farmers and, in many cases, are able to guide them in how they might change their practice or stock without the need for a confrontational court case. Nonetheless, I have been associated with the European research programme, which has done research into public attitudes, and what I have just described leads to some public concern about whether the regulators are doing their job properly.

An improvement to adaptive management would include bringing in two additional groups of people. One would be research scientists, and the other would be those whom I think of as citizen scientists. Citizen scientists are members of the community who are sufficiently interested in the issues, from either the pro-industry side or the anti-industry side, and are willing to contribute some time to take part in some aspects of the monitoring process. SAMS has a good example of citizen scientists who are looking at seashore communities.

That would require full transparency from the industry.

It would require full transparency, and there are clearly issues around that. This is not something that I would recommend as a panacea, but it could be usefully introduced under experimental conditions to see how well it works. I am sure that there are some farms and farming organisations that would welcome such an approach and be willing to go along with it.

Do you believe that the DZR approach should be applied to every fish farm, including the existing inshore sites, or should it be restricted to expansions or new sites in more exposed locations? If it is a good thing, where do you draw the line?

I am saying that adaptive management is a good thing. I do not have strong views on the change from the allowable zone of effects to DZR.

I would like to see some of what we might call environmental social science—some monitoring not only of the environmental conditions at farms that have switched to the new system, but of the way that management works and interacts as part of an adaptive management process and of how the local community feels about it—whether they are engaged and whether it is changing their views of the impacts of the industry.

I have another question, which relates to nutrients, the issue of efficiency and the prospects of multi-trophic aquaculture systems, which produce multiple products. What is the potential for those systems? If we accelerate to 2030, given the industry’s anticipated growth, where do you see multi-trophic aquaculture sitting?

I will pass that on to Adam Hughes.

The concept of integrated multi-trophic aquaculture involves using the waste products from one production level in another. In the case of salmon, those products can be utilised by seaweed or mussels. It reduces the environmental impact and increases the growth of the species that is being co-cultured.

The idea is really attractive, but the practicalities are difficult at farm scale. We really need to go back to the question of what we are trying to achieve by implementing a multi-trophic system. It is quite difficult to balance out nutrient budgets at the farm scale, because there is a spatial mismatch between the amount of space that it takes to produce 1,000 tonnes of salmon and the amount of space that it takes to produce 1,000 tonnes of mussels. Roughly, a 1,000-tonne fish farm might be a hectare. Taking up 10 per cent of those nutrients through IMTA will require about 10 hectares of mussels or seaweed. It is really difficult to imagine that happening at farm scale, because it would mean a huge increase in production at a site.

If we start thinking about it at an ecosystem level, in which we are trying to balance the nutrient inputs from aquaculture and the nutrient reductions from activities such as mussel or seaweed farming, we might get a more viable model—one that works better when we move away from the farm scale to the ecosystem scale.

The benefits of IMTA might be greater than just the impact on nutrient budgets. There is the diversification of aquaculture; the development of new business, more businesses and rural economies; and the social acceptance of aquaculture. If you are looking into IMTA solely to balance nutrient budgets at the farm scale, there are a lot of logistical problems with it. If you are looking at it as a more holistic tool—taking the ecosystem approach to aquaculture, balancing social need and so on—there is more value in it.

Let us move on to the impact of escapes from fish farms.

Good morning. Eric Verspoor has already touched on the issue of escapes. However, we see in the report that, in Scotland, between October 2002 and October 2017, approximately 2.2 million Atlantic salmon were reported to have escaped. We also see references to “drip escapes”, which

“are difficult to identify and quantify and not encompassed by reported escape events, but”

which have

“been estimated in Norway to be substantial”.

We know that the causes of the escapes are human error, holes in the nets, predators and, of course, the weather.

How concerned should be we about escapes? The report says that

“the majority of salmon that escape from farms will not survive to interact with wild fisheries populations”.

How can we be sure of that?

The research that has been done in Norway and, in part, in Ireland indicates that even moderate and low levels of introgression of farm genes into wild populations can affect the normal life history characteristics of the populations in the rivers where that takes place. Once we disturb those characteristics, we will most likely increase mortality rates—it will compound the rates of mortality caused by other factors. In general, such populations also show declines, on average. Therefore the genetic mixing of farmed stocks with wild stocks will almost inevitably have negative consequences if it occurs.

11:15  

Norway has found levels of mixing to be highly variable—from about 5 to 10 per cent up to about 50 to 60 per cent. Those levels are generally associated with rivers that are in farming areas. Equally, there are some rivers in farming areas that are not impacted and others that are outside farming areas that are impacted, so it is difficult to predict. Norway has a system of monitoring and has developed genetic markers with which to estimate levels of introgression to get an indication of the extent to which populations have been impacted.

On the assumption—which is quite well supported by the science—that introgression will have negative impacts, we can then look at managing such situations in the knowledge that we have to reduce the levels of escapes in those areas in order to bring the levels of genetic mixing down. That is the adaptive management approach that the Icelandic Government will be putting in place. It has been approved, and will guide the industry in future. If there is no evidence of introgression, the industry will be allowed to expand, on that criterion. If there is such evidence, the industry will have to take measures to reduce those levels before it will be allowed to expand, or it will be asked to decrease production levels to a point where introgression is no longer a problem. That is the principle of adaptive management.

In Scotland, we have very little information on levels of introgression. We have historical evidence, going back to the early 1990s, from which we know that escaped farmed fish ascend rivers and reproduce. We also have some evidence of subsequent introgression, but it is imperfect. We had to use the Norwegian molecular markers—which are designed specifically for Norway—in Scotland. That did not allow us to get an accurate assessment of introgression, but it was suggestive of it.

On the other hand, I studied a very small river in the middle of a farming area on the west coast. We did not even know that there was a wild salmon population there. There was no evidence of introgression despite the population probably being composed of a few tens of breeding individuals.

The situation can be highly variable, and the only answer is that we need to monitor levels of introgression in Scottish wild stocks regularly and then manage them according to their actual effects, as we know that, if introgression occurs, it is extremely likely that there will be negative impacts to some degree, which will probably scale with the level of introgression.

Are you aware of any molecular or genetic marking that is going on in Scotland?

There are different ways of addressing the issue. The Norwegians have markers that will indicate the level of introgression. They also have markers that allow them to associate farm escapes with particular cages. If there is a farm escape, they can go to local farms, get samples of fish from them and see where the escaped farmed fish might have come from. They do that either by using genetic markers or by profiling the lipids in the fish, because the feed that the fish are given can be quite unique. That approach has been quite successful.

In Scotland, I and my co-author Mark Coulson tried to apply those Norwegian markers, but they were not accurate enough in terms of distinguishing farmed and wild fish. We have just completed a UK research council grant. We are identifying genes for domestication, which should give us better markers that will allow us to go into any river and identify farmed fish and hybrids between farmed and wild fish, and measure the extent of introgression—not fully, but we should be able to do what the Norwegians are doing. We now have a European structural and investment funds studentship that will enable us to look at historical samples and contemporary samples to identify evidence of introgression.

That is good to hear.

Convener, I have a couple of questions on wrasse. Do you want me to wait before asking them?

Yes, we will come to that in a moment.

I would like to wrap up this section by seeking clarity on the statement in section 5.8.3 of the report that

“experiments to develop triploid strains have so far not proven commercially successful.”

Does that mean that it can be done but that it is just too costly, or is there a bigger issue than that?

It is fairly inexpensive to produce triploids. The question about triploids concerns their performance and the economics—are they more susceptible to disease, and do they grow as well? Since the early 1990s, people have been playing with triploids to see whether they would be suitable, but they find that sometimes the performance is equivalent, sometimes it is superior and sometimes it is inferior. The fact that the industry has not taken up triploids suggests that, for some reason, they do not work for it. It may be public perception. Do people perceive triploids as being genetically modified? It depends on the definition of genetic modification.

Is there any other way to physically inhibit the breeding characteristics of fish, were they to escape into the wild?

Potentially there is. Farmed strains are currently selected for the traits that are of economic value, such as growth rate, delayed maturation or disease resistance, but they are also inadvertently selected for domestication—fish that are happy to live in cages tend to survive and are more docile. There are other traits that are not of relevance to production that could be selected in breeding programmes—for example, the tendency to migrate and the ability to reproduce successfully. There is a whole behavioural repertoire that is associated with successful reproduction in the wild. Those traits are not needed in a farm setting because the fish are spawned artificially. There are also stress-related traits that are important in the wild, because you have to be a little bit nervous in the wild in case you get predated upon, and in a farm context those traits can be changed, because stress-related traits are not advantageous.

There are indeed other traits that could be brought in. The immediate economic gain in terms of production is not there, but there could be a longer-term environmental gain in making the fish less able to breed in the wild—so that they do not run up rivers to spawn, for example.

Fish that want to stay at home, near their cage and their source of food, are likely to be the ones that do best anyway.

Exactly. There is some potential that has not been explored yet, but there is some hope that, in the coming years, we might explore those possibilities.

Fascinating. Thank you.

Let us move on to feed supplies.

Good morning. Paul Tett touched on my question earlier. The report states that the sustainability of feed supplies is one of the main issues, and section 6 is headed “Sustainability of feed supplies including substitution with plant-derived ingredients”. Farmed salmon production has been between 130,000 tonnes and 160,000 tonnes, but the intention is to almost double the higher figure to 300,000 tonnes. How can those farmed salmon be fed? Can the feeding be sustained, particularly if Norway is also increasing salmon production? Your report states:

“Increasing salmon production in Scotland and elsewhere (e.g. Norway) will necessarily increase the demand for the raw materials for feed ... required additional, sustainable, source of ‘omega-3’ could be obtained from—”

wait for it—

“transgenic oilseed crops”,

which are commonly known as genetically modified crops—oops! Nobody wants them in Scotland and we banned them, so what do we do?

I will pass that on to my colleague to answer.

The inclusion of marine ingredients in salmon feed is mainly an issue driven by the consumer. The marine ingredients in fish feed are down to about 20 per cent in Norway, but the figure is 25 per cent in Scotland. That is because the United Kingdom consumer prefers a product higher in marine ingredients, which is seen as more natural because it is fed on fish.

There has been complete substitution of marine oils by vegetable oils in salmon feed that has shown very little difference in growth rates. However, the product at the end of that process is not full of omega 3 oils, which are good for public health. The substitution of marine ingredients by terrestrial ingredients can continue as far as we want, but that will mean losing a lot of health benefits and a lot of consumer acceptance of the product.

So, with the greatest of respect, what is the answer? Is it to allow GM crops? A colleague pointed out earlier that 2 million salmon have been lost to escape and 25 per cent have been culled, but they still had to be fed. Does the industry need to get better at being sustainable in order to grow? Or do we just let rip and let everything come back in? I do not think that a lot of people would agree with doing that.

The issue of GM is something for society to decide. As I understand it, Scotland has made a very clear statement that it does not want GM products to be farmed in Scotland. That is a societal decision, but it does not necessarily have anything to do with the science.

On other options, a large proportion of the marine ingredients is sourced from the anchoveta fishery on the Peruvian coast in South America. That fishery is at the limit of its sustainable exploitation; although most of it is well managed, there is no real room for expansion there. There is also increasing use of discards and fish by-products to create fish oil, which in some respect has met demand. However, there are developing technologies, such as microalgal oils and bacterially produced oils, that might produce the oils that are required to go into the fish feed in order to get the public health benefit that comes from eating salmon. That should not be underestimated, because salmon is a major source of omega 3s to the Scottish population and it has huge health benefits.

So, my last question is: do you honestly think that the industry’s desire to double production in the next number of years is achievable? Or would you like to pass on that question?

In terms of feed?

In terms of feed.

I think that it is achievable because no matter what the Scottish industry does, the Norwegian industry is going to be larger than the Scottish industry by an order of magnitude of 10 and it will have to come up with solutions for exactly the same problem. So, there are technological and societal solutions out there, and it is about what we choose to adopt.

11:30  

You mentioned the South American anchoveta fishery. At what point in the next 10 to 15 years will that fishery start to tip over its total allowable catch? At what point will it exceed maximum sustainable yield and therefore be in a state of collapse?

I do not have that information. It is outside my area of expertise, as it is a fisheries management issue.

As I understand it, the fishery is sustainably managed at the moment, which means that it is managed at maximum sustainable yield and is giving us as much as it can. There is much fluctuation from year to year, depending on the El Niño cycle, but it seems unlikely that we can expect a much greater yield from that fishery. If it is not managed sustainably, we might get more fish from it for a few years, but after that we will have exhausted it.

There do not seem to be any other major sources of fish oil available from the natural world at the moment.

So we are at the limit.

Yes.

I could add just one more thing. As Adam Hughes mentioned, one of the possible technologies involves growing what the industry calls microalgae to produce omega 3. They are not what I would call algae, because they do not photosynthesise. The promising method seems to involve a sort of fermentation technology, so it is more like brewing. Scotland is pretty good at brewing, so perhaps that is an industry that we could develop.

I have a brief question about acoustic deterrent devices. Paragraph 7.1.5 in the report says that most ADDs are left to operate continuously. What evidence is there for that? I seem to recall that, in the previous session of Parliament, the Rural Affairs, Climate Change and Environment Committee was told that those devices should and would only be used in short bursts because, otherwise, they would be harmful to occupants of the wider marine environment. Therefore, the assertion in the report is quite concerning. Can you flesh that out?

I confess that I do not know the precise answer to that. I am aware that there is limited documentation available on what sort of devices are used, the conditions that they are used under, the length of time that they operate and the length of time that they do not operate.

So how have you backed up that assertion?

I cannot answer that directly. We will find out.

That paragraph was written by one of the other experts at SAMS. We can look into that and get back to you.

That would be helpful. At face value, the paragraph is concerning, because it raises the question of the need for consistent ADD monitoring and perhaps a licensing regime. That is quite an important point, so it would be good if you could come back to us on that.

Angus MacDonald will ask about wrasse.

During the course of the Aquaculture and Fisheries (Scotland) Bill, the Rural Affairs, Climate Change and Environment Committee went on a fact-finding visit to Lochaber. At that time, Marine Harvest was already using wrasse to control sea lice infestations. I am not sure when that policy was introduced, but there now seems to be more widespread use of wrasse. According to the report, official statistics show that 1.7 million lumpsuckers and 1 million wrasse were bought by the Scottish farming industry in 2016.

Overall, does the evidence show that the commercial rearing of wrasse and lumpsuckers is a sustainable approach to controlling lice the Scottish salmon industry? Do you have any concerns about future increased demand for cleaner fish, particularly given that salmon production is set to increase significantly in Scotland and in Norway? Presumably, there are only so many wrasse available to be bought, so demand will be pretty excessive.

The information that we have is that the cultivation of lumpsuckers seems to be capable of satisfying the demand.

The situation with wrasse is not so clear cut. Information from the industry suggests that by 2019 it would like to be able to cultivate all the wrasse that it uses, but it is not clear whether that is an achievable target. If it is not achievable, clearly the demand for wild wrasse will continue. In that case, there will be a need for fisheries management of the wrasse fishery, following the example that has begun to be developed in south-west England, where a local fisheries management board has successfully managed the wrasse fishery in such a way that it provides a sustainable source of employment for people in that part of the country. Many of those wrasse exports come to Scotland.

The killer question is this: if the use of cleaner fish is so widespread, why are lice still a problem?

That is where we need more research. The industry probably needs a portfolio of different lice control methods: some biological, some chemical and some physical. I have not seen published information as to what is the optimum mixture of the different methods. It might vary from site to site, depending on the cultivation conditions for the salmon and the hydrographic conditions in which the farm is situated.

Temperature certainly plays a part as well. Wrasse are warm-water fish and lumpsuckers are cold-water fish. The northern Norwegians prefer lumpsuckers because they grow more quickly when they are reared and they are more active in the cold temperatures in northern Norway.

The issue is still to be decided in the Scottish fishery. We hear a lot about wrasse but, looking at the numbers, we see that lumpsuckers are also important. If I remember rightly, lumpsuckers can be reared in a few months, whereas wrasse take more than a year to rear in a hatchery, so there are clearly economic aspects to the issue.

Before we wrap up, let us turn to mitigation.

I must confess that, when I saw the BBC programme, I thought that wrasse or lumpsuckers were the solution, but in the report you talk about recirculation aquaculture systems and enclosed systems, which seem to be the solution. What are the main environmental concerns about enclosed systems? Could they become the main means of salmon production in the future? If not, why not?

On the environmental side, the main concern with recirculation systems is in dealing with solid waste: the same amount of solid waste is generated, but it is generated inland or in an enclosed system at sea. That waste needs to be dealt with in some way. It is saline waste, so it might not be suitable for standard means of waste disposal, such as the biodigesters that might be used on a farm. A biodigester can be used for saline waste, but it needs to be dedicated to that purpose. There might be other ways of reusing that material.

There is obviously much better disease control, including lice control, as the water that goes in and what goes out are controlled. That system is an option for development of the industry; reduction in the number of sea lice is one of the reasons why the industry is so interested in it. I reiterate my comments about developing a system that is economically viable for the industry.

As I said, we also have societal concerns and perceptions of what is natural versus perceptions of what is unnatural. To make the economics for recirculation systems work, we would increase the stocking density well beyond the level at which an open-water system would operate. That would have animal welfare implications and societal implications in respect of where we judge those standards to be.

I suppose that it will depend on what we define as “natural”, given the amount of salmon that we see packed in to cages. As John Scott said earlier, you are really looking for an engineering solution. Will you say a bit more about the economics of that? Would the salmon-producing companies have to invest much more in the enclosed systems that are much safer environmentally?

The capital expenditure that is required to set up recirculation plant is much more than what is required for an open-water cage system, and the same is true of the running costs, to a degree. However, that cost could be offset by better control of the life cycle and lower disease impact. I cannot make an economic analysis—that is for the salmon-producing companies to do—but at the moment the economic balance is obviously not tipping towards recirculation systems, or else there would have been more widespread adoption of the technology.

Throughout this evidence session, we have heard that there is a lack of data and research in many areas. The Norwegians seem to be much further ahead than we are on investment and research. Is that a fair conclusion to draw?

In my opinion, that is an absolutely fair conclusion. The level of investment in research in Norway is much higher than that in Scotland, and technology development in the Norwegian aquaculture industry is much more advanced than it is in Scotland.

Finally, you have spoken at length about adaptive management systems, and you have pretty well said that it would be a good idea to pursue that direction of travel. Would you like to agree that that is the case?

I would like to agree to that.

I just wanted to have that very definitely on the record.

Adaptive management is essentially evidence-based management being done in an adaptive way. As evidence accumulates, management gets better because it is based on that expanding body of information and understanding of the system.

You also said that that happens on a real-time basis, which relates to the ability that we now have, and that we probably did not have 10 years ago, to understand in real-time populations of lice, for example.

Indeed. Paul Tett also made an important point about including communities, individuals and society in ways in which they have not previously been included, in order to minimise conflicts.

Yes. We take that on board.

Finally, we will hear two brief supplementaries from Mark Ruskell and Claudia Beamish.

On that last point, you spoke earlier about adaptive management versus the precautionary approach. Do we need a kind of hybrid? We have talked about the cocktail effect of chemicals, which is an issue on which the precautionary approach may be needed, but there are other elements of which we have some understanding, but there is a lack of monitoring and research. Is that a fair way to characterise the situation, or do we just have either the precautionary approach or the adaptive approach?

I think that what Mark Ruskell says is apt: we need a mixture of precaution and adaptive management. Of course, there is a judgment to be made about how much precaution is needed, and that judgment goes back to the risks from any particular issue and whether they are likely to be unmitigable in the long term or to be risks from which we can recover.

Research has to be done to monitor the levels and to look at compliance with standards, but there also has to be research on the standards, because we often do not understand enough to know whether a standard is appropriate. Adaptive management involves monitoring as well as researching the standards.

Do any of the witnesses wish to comment from a scientific perspective on the relationship between fish farms and marine protected areas and other marine protections?

I was one author of a paper last year that dealt with that issue across Europe. The working group that put the paper together concluded that, given that an MPA is usually designated for a specific purpose—maerl beds or transient cetacean populations, for example—if aquaculture is to be sited within MPAs, as much aquaculture in Scotland already is, we need to understand the impact of the industry on the specific objective of the protection.

If there is an MPA for maerl beds, for example, you would not put a fin-fish farm on top of it, but if the MPA was for wading birds or a shoreline feature it might be possible to have a farm off the coast, so long as the appropriate processes were gone through. Scotland has a robust system: if the risk assessment says that there would be no impact on the conservation feature that the MPA is there to protect, there is no reason why the MPA and aquaculture cannot coexist in the same space.

Thank you for your time. The discussion has been useful in teasing out some of the detail of the report.

11:46 Meeting suspended.  

11:51 On resuming—