Sense and sentinels

I have always been sure that it was unnecessary for Scottish Ministers to appoint a regulator to manage sea lice, writes Dr Martin Jaffa. As the weeks go by, I am also increasingly convinced that the appointment of SEPA to this role was a major mistake.

Having watched their efforts from the outset, I firmly believe that SEPA have little understanding of sea lice and the interactions with wild fish. I suspect that SEPA have simply listened to a narrative promoted by Scottish Government scientists based on mathematical modelling and have adopted the narrative as their approach to regulation but critically, without any real understanding of what it means.

The regulation is based on a sea lice dispersal model which assumes that sea lice found on farms will produce millions of infective larvae which are dispersed away from the farm, infesting all the fish they subsequently encounter. Unfortunately, they have been unable to validate their model because so far no-one has ever found the clouds of sea lice larvae predicted by their or anyone else’s sea lice dispersal model.  Instead, the salmon industry is to be regulated based on the outcome of a theoretical model.

The Scottish Government has collected masses of data on wild salmon and sea trout since the beginnings of the salmon farming industry. This data provides clear evidence that salmon farming is not the reason why wild salmon and sea trout numbers have collapsed. SEPA are not interested in this data because of their failure to understand such interactions.

SEPA plan to validate their model by using sentinel cages. However, sentinel cages will never validate the model because the infective stages that the model predicts do not exist in the real world. If sentinel cages were such a good method for predicting sea lice infestation, then why have they not been in regular use? The last time they were employed in Scotland was over 10 years ago and over a period of three years, these sentinel cages did not provide a shred of evidence that modelled infective sea lice larvae caused any threat to wild salmon during the key migration window.  This has not deterred SEPA from pursuing this approach.

Sentinel cages are small, enclosed net pens which are stocked with up to 50 farmed salmon smolts and left tethered to the seabed in key locations. After a prescribed length of time, the cages are recovered and the numbers of lice on each fish counted. If sea lice are found on the fish, then the presence of a salmon farm is blamed even though the infective lice larvae have never been recovered in large numbers from the surrounding seas. The possibility that any lice are the result of transfer from passing wild fish is dismissed out of hand because the model says otherwise.

Back in March, SEPA’s Head of Ecology, Peter Pollard had told the audience at the Fisheries Management Scotland Conference: “We now have got a grant funding arrangement with the Institute of Marine Research [IMR] in Norway, and they will be working alongside us on the sentinel cage work.”

This leads to the obvious question: why does SEPA need to work with IMR on sentinel cages? Certainly, Marine Directorate scientists have used sentinel cages previously, with the biggest study running for three years from 2011 to 2013 in Loch Linnhe. The data from that study was used in the more recent SPILLS project involving scientists from the Marine Directorate, Scottish Association of Marine Science and from some salmon farming companies. If sentinel cages were considered such a key indicator of lice infestation, why were they not deployed as part of the SPILLS project especially as the project concerned salmon parasite interactions.

Turning to Norway, IMR still use sentinel cages as part of the Traffic Light System assessment but interestingly not across all 13 Production Areas (POs). In fact, last year, they deployed sentinel cages in just two of the POs, which suggests that their value is extremely limited. If sentinel cages were considered a key part of the assessment process, then surely IMR would deploy them in every PO, and they don’t. The most effort they have employed to utilise sentinel cages was in 2017 and 2018 when they were installed in seven out of the 13 POs but by 2020, the number was reduced to just one. For the last three years, IMR have deployed sentinel cages in two of the POs but not always the same two. IMR clearly do not consider sentinel cages to be an integral part of the assessment process, yet SEPA does.

Sea louse

A costly programme

Whilst SEPA see sentinel cages as an important tool in validating their model, they are obviously not very confident how this should be achieved which is why as Peter Pollard told the FMS conference, SEPA have a grant funding arrangement with IMR to seek their advice.

Of course, the obvious question is what this grant funding arrangement is and to answer that a Freedom of Information request was submitted.

The official response stated: “The recipient of the grant from SEPA to develop an international standard for sea lice monitoring is the Institute of Marine Research in Norway. The grant value is £200,000.”

I had to read this more than once to make sure I was not reading it wrong. SEPA are paying IMR £200,000 to help develop a sentinel cage strategy to prop up their already weak sea lice framework. SEPA say this is to develop an international standard but as SEPA operate just in Scotland, why do they need an international standard?

It should also be remembered that if sentinel cages were such a good measure of sea lice infestation, then why have the Marine Directorate not continued a programme of sentinel cage monitoring rather than just fund the wild fish trusts to assess lice numbers on a few fish they catch?

However, SEPA’s willingness to splash the cash when it comes to regulating salmon farming does not stop with sentinel cages.

In his presentation to the FMS conference, Peter Pollard also stated: “We’ve also just awarded a big grant to develop a new model, a better model for Shetland and parts of the West Coast.”

SEPA’s response to the FOI included reference to this project: “The recipient of the grant from SEPA to extend and improve sub-area models of the Scottish Shelf Model, is the National Oceanography Centre. The Scottish Shelf Model is held by Marine Directorate and made available as an open service to support development in Scottish coastal waters. The grant value is £400,000.”

I had mentioned that the only area SEPA appear to be confident about was their modelling yet, they appear to be willing to spend £400,000 trying to improve the model as well as extend it to include the Northern Isles. I am left wondering how much confidence SEPA have in any part of the framework. Certainly, there are questions as to why SEPA want to extend the framework to include the Northern Isles.

The Scottish Government’s own catch data shows that there is no data recorded for Orkney despite the well-publicised activities of the Orkney Trout Fishing Association. By comparison, there is a record for fishing in Shetland, although this is mainly for sea trout as salmon hardly feature in the catch record. Shetland sea trout catches since 2000 are shown in the following graph and as can be clearly seen, are not affected by the presence of salmon farms.

Instead of rushing to spend £600,000 on the sea lice framework, SEPA might want to consider talking to those people in Scotland who possibly know more about sea lice than the official regulators appear to know.