LAUSR

Abstract Since achieving population recovery targets, management of the Saginaw Bay stock of walleye in Lake Huron is informed by a statistical-catch-at-age (SCAA) model providing estimates of abundance, spawning stock biomass, mortality rates, and exploitation rates. Movement was examined by an acoustic telemetry study and indicated that 37% of the adult population spends much of spring, summer, and fall in the main basin of Lake Huron and unavailable to the fishery in the bay. The current SCAA model used by managers accounts for this movement by including harvest estimates from main basin fisheries in addition to those within the bay. To quantify the effect of the inclusion of the migration information into the model, we constructed a reduced model version that was limited only to the bay harvest and data sources, and then compared these estimates to the full model estimates. All estimates from the reduced model deviated significantly from the full Lake Huron model’s estimates. Significantly different by year were 61% of population size estimates, 25% of total annual mortality estimates, 18% of recreational exploitation rates, and 52% of spawning stock biomass estimates. Differences between the two models were greatest in years of high walleye abundance (i.e., after recovery) and retrospective analysis indicated that this departure was not an estimation artifact. Generally, the reduced model underestimated predicted abundance. We concluded that incorporation of knowledge of movement of Saginaw Bay walleye from the acoustic telemetry study resulted in better informed stock assessment estimates.