Risk of inter-river transmission of Gyrodactylus salaris by migrating Atlantic salmon smolts, estimated by Monte Carlo simulation

Abstract

The possibility of Gyrodactylus salaris infection of wild North Atlantic salmon Salmo salar spreading to new rivers poses a major threat in Norway. This freshwater parasite can survive for some time in brackish water, and it has been suggested that smolts leaving infected rivers could transport vital parasites to new rivers. A Monte Carlo simulation model was used to estimate the risk that infected smolts would ascend a new river. Data from an infected watercourse in Norway, where the salmon population is maintained constant by cultivation, were used. The model included information on prevalence of infection, hydrographical conditions, survival of G. salaris in brackish water, fish population characteristics, and smolt behaviour during seaward migration. The annual risk was estimated for 3 neighbouring rivers situated at different distances from the index river. For the nearest river, which shares the same brackish water zone with the index river, the model estimated an annual risk of 31% that at least 1 infected smolt would ascend this river. The results of the simulation were highly sensitive to the water salinity along the migration route. For the other rivers, the annual risk was lower than 0.5%. Risk was positively correlated with the number of fish leaving the index river, indicating control of this number as a possible tool in risk management. KEY WORDS: Gyrodactylus salaris · Atlantic salmon · Inter-river transmission · Monte Carlo simulationResale or republication not permitted without written consent of the publisher

Geographic risk factors for inter-river dispersal of Gyrodactylus salaris in fjord systems in Norway

“…G. salaris was recorded on experimental fish after more than 200 h at 10 ‰ salinity and 72 h at 15 ‰ salinity, and 37 out of 52 Atlantic salmon smolts caught 25 km from the outlet of the River Drammenselva were infected with up to 1105 G. salaris specimens (Soleng & Bakke 1997, Soleng et al 1998. Risk assessment studies of fjord-wise dispersal pathways to salmon rivers near to the River Drammenselva estimated an annual risk of 31% for the event that at least one infected salmon smolt would enter the River Sandeelva (Høgåsen & Brun 2003). G. salaris readily transmits to a wide range of both salmonid and non-salmonid host species, suggesting that a range of species may serve as short-term transport hosts for the parasite (Bakke et al 2002).…”

“…However, an interesting case is the River Sandeelva in Region 1, with a relatively low predicted probability (0.39) of infection in the present study. The risk-assessment conducted by Høgåsen & Brun (2003) concluded that the risk of inter-river dispersal of G. salaris on migrating fish from the River Drammenselva to the River Sandeelva was high, but that their analysis was sensitive to assumptions with regard to salinity in the fjord area covering the dispersal pathway. Incidentally, G. salaris was recorded in the River Sandeelva in 2003, 16 yr after G. salaris was recorded in the River Drammenselva.…”

“…Outlets of infected rivers into fjords were found to be located sufficiently close to each other that this seemed a plausible proposition. Circumstantial evidence supporting this hypothesis has subsequently appeared: (1) the salinity tolerance of G. salaris (Soleng & Bakke 1997); (2) records of G. salaris infection in salmon smolts in brackish water (Soleng et al 1998); (3) juvenile salmon migration in fjords (Lund & Heggberget 1992, Finstad et al 2005; and (4) risk assessment models (Høgåsen & Brun 2003). Johnsen & Jensen (1991), Soleng et al (1998) and Høgåsen & Brun (2003) all give accounts of the mechanisms by which inter-river dispersal of G. salaris on fish migrating through brackish water in fjords may occur.…”

Quantification of free-living Gyrodactylus salaris in an infested river and consequences for inter-river dispersal

“…The low water concentration of parasites also indicates that the probability of attachment of drifting parasites to immersed equipment or swimmers is low. In comparison, the risk of inter-river transmission of G. salaris by migrating Atlantic salmon smolts or other species could be significant (Høgåsen & Brun 2003, Peeler et al 2004). All results support the hypothesis by Jansen et al (2007) that G. salaris dispersal in Norway is due to either anthropogenic movement of infested salmon, or migration of infested salmon in fjord systems.…”

“…In accordance with Høgåsen & Brun (2003), it was assumed that the sensitivity (Se) of the diagnostic method used by Soleng et al (1999a) was most likely 0.95, with a minimum value of 0.9 and a maximal value of 1, i.e. 90 to 100% of the fish infested by Gyrodactylus salaris were detected.…”