Modelling dispersal of salmon lice in a large fjordic system: Loch Linnhe, Scotland

Abstract

Sea lice are marine ectoparasitic copepods that infect a wide range of fish and other species. Salmonid fish in their marine phase are infected by a variety of generalist Caligus species (in Scotland C. elenogatus) and the salmonid specialist Lepeophtheirus salmonis. Cost of salmon lice is estimated to be €305M damage worldwide and have also been associated with declines in wild salmonid populations. Sea lice undergo a planktonic larval phase that consists of two non-infectious naupliar stages and the infectious copepodid stage. During this planktonic phase the larvae are transported by marine currents and so it is of crucial importance for our understanding of how lice on different host populations may interact. This is important for protecting wild fish and for identifying groups of farms that would benefit most strongly from coordinated management. We therefore are developing a model of the Loch Linnhe system in the western Highlands of Scotland using a modelling approach first developed for the smaller Loch Torridon system. Loch Linnhe is Scotland’s largest sea loch (fjord) being more than 60 km long and containing 10 salmon farms operated by two companies. A coupled hydrodynamic-particle model is used to derive the distribution of lice under different environmental forcing. The hydrodynamic model used is POLCOMS with a 100 m horizontal resolution and 15 minute time-step in the first instance. This model is forced using tidal forcing at the mouth of the loch (Firth of Lorn and Sound of Mull), winds, and freshwater inputs from rivers and side lochs. The particle model describes the development of the larvae through the different stages and their mortality. Development rate is strongly temperature dependent; lice take about 3.6 days from hatching to become infectious at 10C but much longer at lower temperatures. Mortality rate is highly sensitive to salinity, increasing rapidly once salinity drops below 29 ppt. Initial results show that the larvae can be transported over large distances, sometimes 20-30 km from their source, before becoming infectious. Here they can form elevated but localised concentrations. Copepodids may be absent from their parents’ hosts. A sampling programme has been initiated to validate the model predictions. Data has been obtained for May 2011 and a second validation exercise is planned for October 2011; further validation data will be obtained in 2012 and 2013. The validation consists of the collection of hydrodynamic and forcing data, using plankton tows to detect larval lice (all phases), and sentinel fish to attract the infectious copepodids. The initial sampling was carried out in May 2011 under extremely difficult conditions of south-westerly winds reaching up to 85 mph. Despite this, planktonic tows were completed for 30 of the planned 31 locations, 9 out of 10 sentinel cages were recovered, and hydrographic measurements were taken. Useful data was thus obtained under poor weather conditions which give confidence that validation will be practicable under a representative range of forcing conditions. The model will be used to evaluate where concentrations of lice form under observed environmental forcing and to advise on the management of salmon farms.