Model-Based Evaluation of How Stocking Pacific Salmonids Influences the Fish Communities of Lakes Huron and Ontario

We developed offshore fish community models for lakes Huron and Ontario to forecast the potential consequences to changes in Pacific salmonid Oncorhynchus stocking policies by the states of New York and Michigan and the province of Ontario.  The Lake Huron forecasting model included Chinook salmon Oncorhynchus tshawytscha, lake trout Salvelinus namaycush, steelhead Oncorhynchus mykiss, walleye Sander vitreus, and burbot Lota lota as predators; the Lake Ontario forecasting model included Chinook salmon, lake trout, steelhead, Atlantic salmon Salmo salar, brown trout Salmo trutta, and coho salmon Oncorhynchus kisutch as predators.  Alewife Alosa pseudoharengus and rainbow smelt Osmerus mordax were the primary prey species for both models.  Evaluated stocking policies ranged from a complete cessation in stocking of all Pacific salmonids to a doubling of the 2003-2005 average stocking rate.  Policies were evaluated using metrics that captured the desire to have high, stable catches of lake trout and Chinook salmon and large sizes of Chinook salmon.  The susceptibility of alewife and rainbow smelt populations being driven to low adult abundance levels was also measured for the policy evaluations simulations.  We found that for Lake Huron decreases in Pacific salmonid stocking rates were predicted to increase biomass and fishery yield of lake trout and age-3 Chinook salmon spawning weights throughout the lake; additionally, recreational yield of Chinook salmon was predicted to increase in the Main Basin of Lake Huron.  Biomass of alewife and rainbow smelt biomass also increased with decreases in Pacific salmonid stocking rates; however, even with a complete cessation of all Pacific salmon stocking, there still remained a high probability of prey abundance declining to below threshold levels as a result of presumably high wild recruit of Chinook salmon. Results will be discussed in context of large fish community changes that have recently occurred in Lake Huron.  The fish community of Lake Ontario was relatively unaffected by changes in Pacific salmon stocking policies because of higher prey productivity and presumably low wild recruitment of Chinook salmon.  The results for both lakes Huron and Ontario were sensitive to assumptions about predator wild recruitment, predator search efficiencies, and scaling of prey stock-recruitment relationships.  These sensitivities should be kept in mind in any proposed changes in Pacific salmonid stocking rates.