Including predation mortality in stock assessments: a case study for Gulf of Alaska walleye pollock

A separable catch-age stock assessment model that accommodates predation mortality. is applied to the Gulf of Alaska walleye pollock (Theragra chalcogramma) assessment. Three predators are incorporated in the model: arrowtooth flounder (Atheresthes stomias), Pacific halibut (Hippoglossus stenolepis), and Steller sea lion (Eumetopias jubatus). The effect of these predators is examined by defining the predation mortality as a type of fishery. The model is used to quantify changes in the relative fit to the survey, fishery, and predator data when the assumption of constant natural mortality is relaxed. Specifically, we examine the effect of assumptions regarding the functional feeding response, residual naturaly mortality, and uncertainty in predator biomass on stock assessment. Total natural mortality rates (including predation) tended to be higher than estimated from life history characteristics of the stock. Models that did not account for uncertainty in natural mortality underestimated uncertainty in current stock biomass by as much as 20%. Our results indicate that independent estimates of survey selectivity, additional food habits data, and estimates of the feeding responses of predators to different prey densities are all needed to improve our ability to develop stock assessment models that address ecosystem concerns. (C) 2000 International Council for the Exploration of the Sea.