Host density and pathogen transmission in salmon aquaculture: effects of increased production and pathogen control policies

Sustainable growth in the aquaculture industry is a stated political goal in Norway. Growth through increased production of farmed fish may, affect the spread of infection because the spread of pathogens often depends on the density of host organisms. A key question is thus how to increase production, while having control over the spread of infection. The overall goal of this project was to quantify the effects of growth and organization of production on the spread of infection and the occurrence of disease in marine farming of salmonids. To achieve this, extensive data from the fish farming production in Norway, together with data on infection and the spread of infections, were used. The data form basis for development of models for infection and spread of infection, which were used for exploring various scenarios for production growth, geographical allocation and for various strategies for infection control. (All studies marked * has been submitted for publication in peer-reviewed journals, and those marked with ** has been published) In the project, a model for salmon lice was developed. The model calculates the expected incidence of adult female lice and other mobile stages of lice on fish farms and the effects of lice treatments. The model provides insight into a number of important biological processes in salmon lice population dynamics on farmed salmon. This include generation time from adult female lice into other mobile lice and effects of conditions such as fish size, lice treatment and distance between sites on changes in lice deposits and the spread of infectious lice larvae.** The salmon lice model was used for the project's main goal, simulation of various scenarios for production and infection control. The results indicate that an increase in biomass leads to an increase in the total number of salmon lice, so that a doubling leads to an increase of 103.5 % and a fivefold increase to a 107 % increase at national level. According to simulations, it is hypothetically possible to keep the amount of lice at the current level, regardless of the increase in biomass, but it will require an increase in the number of lice treatments to a level that is considered unrealistic.* The model was used to simulate what on a theoretical basis will happen if there are fewer and larger sites in PO3. Results show that a gradual reduction in the number of sites results in a gradual reduction in the number of salmon lice. This indicates a good effect of distributing the biomass to fewer sites. The simulations also showed that this effect is greater if the sites that are closed are chosen on the basis of how much they contribute to the spread of lice. This study is described in a report 'Changed locality structure in production area 3 - assessed effect on the spread of salmon lice, pancreatic disease and infectious salmon anemia', prepared in collaboration with HI. The model was also used to simulate whether synchronized releases within zones contributes to increased lice control. Here, Sunnhordland was used as an example. We used a scenario where the production history of each site was randomly allocated to new sites at the start of the simulation period and one where sites are moved within the zones, so that the distance between sites becomes smaller within the zone but the distance becomes larger between zones. The results from the simulations indicate that zoning in Sunnhordaland has not had an infection-reducing effect. The project has also updated and further developed a model for the spread of pancreatic disease (PD) in salmon. The model has been used to simulate the spread of infection from sites infected with PD virus. The model can calculate the probability that each individual site within a radius of 100 km will be infected after introduction to a new site. This has been used to assist in the management of PD through knowledge support to relevant authorities. The model has also been used to simulate the effect of various control measures against PD, and results from this clarify the benefit of rapid culling to limit the spread of infection.* The PD model was adapted to simulate the spread of Infectious Salmon Anemia (ISA), and different scenarios for control were tested. The results indicate that the introduction of compulsory vaccination can be a fullworthy alternative to the current strategy of immediate culling. To reduce the number of ISA outbreaks, mandatory screening for ISA virus must be introduced.* In the project, preparatory work has also been done to identify costs related to lice control, in the first instance the biological costs in the form of mortality. This study shows that there is significant mortality associated with all types of treatment against lice, and that the mortality after thermal and mechanical de-lousing is highest. There is great variation in mortality between each individual treatment.**

I prosjektet er det utviklet modeller som etterligner spredningen av lakselus, pankreas sykdom (PD) og Infeksiøs lakseanemi (ILA). Modellene har blitt brukt til å undersøke forskjellige muligheter for mer bærekraftig forvaltning av akvakultur i Norge, med tanke på bedre kontroll av sykdommer, bedre velferd for fisken, samt effekten på miljø i form av spillover av patogener til villfisk. Modellene blir allerede brukt i forvaltningen av lakselus, PD og ILA. I tillegg har prosjektet bidratt til økt internasjonalt forskersamarbeid. To utenlandske forskere har vært på gjesteopphold i Norge, og bidratt med overføring av viktig kompetanse til det norske veterinærepidemiologimiljøet. Prosjektet har også muliggjort oppstart av ett PhD-forløp, med fokus på akvatisk helseøkonomi, herunder kost-nytte effekter av kontroll av patogener. Nasjonalt har prosjektet bidratt til økt samhandling på tvers av fagfelt og institusjoner.

Sustainable development and growth in the salmon farming industry is a primary goal for the Norwegian government. Due to host density dependence in the propagation and spread of pathogens in salmon farming, further growth in the production will come at a cost of increasing risks of pathogen infections and disease effects. A key question, at this point, is how to organize the production in order to balance production yield and risks of pathogen transmission. The overall aim of the proposed project is to assess quantitative effects of the spatial allocation of production, production growth, and pathogen control policies, on host-pathogen infection dynamics in the salmon farming industry. To accomplish this, the proposed project will make use of the wealth of data accessible for the Norwegian salmon aquaculture system to update and refine stochastic host-pathogen models, which will be used for scenario simulations. The scenarios will vary the spatial organization of salmon farms, production growth trajectories, and pathogen control policies. The simulations will focus on the viral pathogens causing pancreas disease and infectious salmon anaemia, and on salmon lice. The output from the simulations will consist of future production histories covering the total marine production of salmon in Norway, pathogen and disease incidences in the farmed host populations, and pathogen control efforts. This output will further feed bio-economic models to quantify costs and benefits associated with the various management strategies and production growth scenarios. Environmental aspects relating to fish health and welfare issues, or pathogen spill-over to wild fish populations, will also be addressed, albeit in a more qualitative way. Ultimately, the project will contribute to the development of rational production-systems to maximize production yield, while at the same time ensuring an economically and environmentally sustainable salmon farming industry.