Incorporation of Predation into a Population Assessment Model of Eastern Bering Sea Walleye Pollock
P.A. Livingston and R.D. Methot
- Price: $0.75 Sale: $0.00
The effect of predation by adult walleye pollock (Theragra chalcogramma), Pacific cod (Gadus macrocephalus), and northern fur seals (Callorhinus ursinus) was incorporated into a stock assessment model of walleye pollock in the eastern Bering Sea. Previous studies showed these three species are the primary predators of young walleye pollock. The stock synthesis model was used to incorporate the three predators as sources of mortality in a single-species catch-at-age analysis. Predators were entered into the model by defining (1) a time series of predator abundance (i.e., effort) over the entire modeled period, (2) a series of pollock consumption data (i.e., catch per unit effort) for each predator for each year when food habits data were available, and (3) the age composition of pollock consumed for each year where food habits data were available. The instantaneous annual predation mortality rates estimated from cod and fur seal predation were small, ranging from about 0.02 to 0.04. However, mortality rates of age-1 pollock due to cannibalism were high.
Adding cannibalism on age-1 fish increased model estimates of the number of age-1 pollock in recent years, which changed the apparent stock recruitment relationship at age 1 from one with declining recruitment at high spawning stock sizes to one where recruitment is more asymptotic at high spawning stock sizes. Our analysis suggests that mortality due to cannibalism on pollock may be an important factor in determining recruitment into the fishery at age 3. Adding predation, and particularly cannibalism, to the model improved the fit of the stock recruitment relationship at age 1. The largest outliers from the estimated relationship were years with generally warmer temperatures and higher inshore transport of surface waters. Oceanographic conditions and predation both appear to play an important role in pollock year-class success. Simultaneously accounting for both these factors in future models of pollock population dynamics may improve our understanding of the relative importance of
those factors in determining pollock recruitment.
- Item number: AK-SG-98-01ai
- Year: 1998
- DOI: https://doi.org/10.4027/fsam.1998.35