A brand new model
Mowi Scotland has been pioneering a new approach to sea lice modelling that suggests regulators may be over-cautious.
Mowi Scotland has pioneered new sea lice modelling as part of its application to grow post-smolt salmon at its farm in Loch Etive, which has now been approved. The model aims to offer the most effective control of sea lice and was developed by Dr Philip Gillibrand, Mowi’s Head of Oceanography and Environmental Monitoring.
Dr Gillibrand produced a complex sea lice distribution model for Loch Etive which compares the continuous production of sea-grown rainbow trout to the production of salmon smolts in two crops annually with fallowing events in every year. Sea lice levels were predicted to be vastly reduced under this farming strategy and, in mid-July, Dr Gillibrand and Ben Hadfield, COO Mowi Farming Scotland, Ireland, Faroes and Atlantic Canada, visited the site to analyse and validate the model against sea lice levels.
Dr Gillibrand has been developing sea lice models within Mowi for over six years and has collaborated with SEPA, Marine Directorate and other scientific institutes to achieve a robust and transparent model architecture. As with all environmental modelling, the complex parameters come with a degree of precaution, but intense work is needed to ensure accuracy over precautionary assumption, which can result in an inaccurate outcome.
The initial validation of sea lice modelling in Loch Etive, which was done with the largest ever deployment of post-smolt salmon (2.5 million), shows that the model has good accuracy, but still over-predicted sea lice levels in the first five months of stocking.
Dr Gillibrand said: “We used a sea lice dispersal model to establish the connectivity between our farm sites in Loch Etive and neighbouring sites, and then combined that with a population dynamics model to predict the lifecycle of the lice on the farms and estimate the daily production of lice larvae.
Both models include a degree of precaution, and the lice count data we are collecting will help fine-tune the models and improve the predictions in the future.”
Commenting on the initial validation of sea lice modelling, Ben Hadfield, COO Mowi Farming Scotland, Ireland, Faroes and Atlantic Canada, said: “The post-smolt salmon looked excellent: perfect form, healthy gills and low sea lice levels. Our model has a strong correlation to lice epidemiology, but still over-predicts to a reasonable level, due to the precautionary assumptions in line with the disciplines expected in environmental modelling such as the fact that we are not able to model the full extent of our farmed cleaner fish removing sea lice from our salmon.”
The research team at Loch Etive: (From left) Dr Philip Gillibrand, Sean Anderson, Allan Murdoch and Luke Plummer
Regulating the Scottish salmon farming industry
Sea lice modelling now forms a big part of regulation in Scottish salmon farming due to SEPA’s Sea Lice Risk Framework where the migration of wild salmon through salmon farming regions on the West Coast is modelled. This was first suggested by the salmon industry as part of the Salmon Interactions Working Group as a practical way of improving the relationship between the industry and wild fish interests.
Hadfield believes that a robust and validated model which predicts sea lice levels is an important part of the world leading aquaculture regulation but warns that it will take three to six years to do it properly and accurately, even with the intense resource deployed by Mowi Scotland in the case of Loch Etive.
He is also critical of what he describes as “ultra-precautionary and therefore unrealistic model architecture” that has been presented by wild fish ENGOs such as Wild Fish, Coastal Communities Network and Fisheries Management Scotland, which recently gave evidence to the Scottish Parliament: “I was disappointed to see Wild Fish and CCN state that sea lice levels were high and uncontrolled when the data shows they are at the lowest level for 20 years. I believe that this recent evidence was at best misleading and given with the express intent of damaging the salmon farming industry. Indeed, this was duly refuted by Charles Allan, the most senior operational regulator at the Scottish Government, who gave evidence to the Parliamentary Committee that sea lice were under control and at low levels.”
What SEPA’s SLRF Screening Model shows, the company says, is that out of more than 220 operational salmon farms in Scotland, only 19 have a theoretical risk, according to a model that vastly over-predicts effect.
On Fisheries Management Scotland (FMS), Hadfield believes that whilst both organisations agree that a robust and validated model is a major step forward, what FMS failed to say to the committee was unacceptable: “The current SEPA model over-predicts sea lice concentrations, possibly by a factor of 4-5, and uses a very low impact threshold which equates to detectable effects on wild salmon smolt behaviour but not levels that would induce high mortality. It assumes all salmon farm biomass is constantly at its maximum, which it is not, and, crucially, it is yet to undergo full validation to remove layer upon layer of over-precautionary assumption in order to attain a realistic correlation.
“If this is not changed, then it will over-regulate and force unnecessary treatment of farm-raised salmon which will challenge the high welfare of stocks, which all salmon farmers work for daily. Not explaining this clearly and calling for an ever more precautionary and rushed approach is misleading and is designed to achieve an over-prediction of impact. I fervently hope that this can be corrected in future sessions, and we can talk with confidence about the reality of conditions in the sea and how salmon farmers across Scotland understand and adapt their methods accordingly.”
Mowi lice modelling graphic: Etive post-smolt lice counts predicted and observed weeks 1-22 2024
Post-smolt salmon at Loch Etive
Mowi Scotland acquired the farms at Loch Etive as part of its acquisition of Dawnfresh Farming, Scotland’s largest trout producer, in 2023.
Mowi was granted consents to grow post-smolt salmon in Loch Etive in late 2023, and the first fish were put to sea in February 2024. Fish from Loch Etive will be used to stock the company’s larger sites in the Small Isles, Loch Linnhe and Loch Seaforth.
Mowi says it is confident that the post-smolt salmon production at Loch Etive is a sustainable alternative to land-based post-smolt production which incurs intense construction costs and energy consumption. Additionally, the company believes that this approach will deliver the most effective control of sea lice due to the new approach to fallowing.
Previously, individual sites at Loch Etive were fallowed at different times. Mowi’s new production plan includes six-month growth cycles to enable farms to synchronise fallow periods throughout the whole loch system. This break in production, combined with the loch’s naturally brackish waters, is intended to maintain a low sea lice population in Loch Etive (sea lice do not flourish in low salinity).
With less standing biomass on site, less feed will be used and therefore less organic waste generated which will reduce the benthic footprint of carbon on the sea bed.
