Comparisons of walleye pollock, Theragra chalcogramma, harvest to Steller sea lion, Eumetopias jubatus, abundance in the Bering Sea and Gulf of Alaska

Abstract

Steller sea lion, Eumetopias jubatus, counts made on rookeries in the Gulf of Alaska, Aleutian Islands, and the Bering Sea from 1976 to 1991 were compared to annual estimates of walleye pollock, Theragra chalcogrmma, harvest to examine possible relationships between Steller sea lion abundance and commercial pollock fishing. Comparisons were made between Steller sea lion counts and pollock fishery data from the same year and from 1 to 5 years prior. Two sets of study areas were used, one encompassing 60 nmi or 120 nmi blocks around six major rookeries and another with 13 sites defined by 20 nmi radius rings around major rookeries. Lacking accurate measures of local pollock biomass, harvest data were used as approximations of annual differences in pollock abundance. Following methods used in Loughlin and Merrick (1989), correlation coefficients were calculated using a linear model and tested for significance using a 2-tailed null hypothesis ( = 0.05). However, since the data are probably not normally distributed, Kendall's coefficients of rank correlation were also calculated and tested for significance   ( = 0.05).

Using the parametric approach, some significant positive correlations were obtained comparing sea lion counts to pollock harvests 1 to 5 years prior (2 in the first set of comparisons and 8 in the second). These may indicate circumstances where pollock abundance was adequate to sustain the fishery while sea lion abundance remained unaffected. Significant negative correlations (2 in each set of comparisons) may suggest instances where areas of low pollock biomass were effectively harvested, but the low or reduced availability of prey, or the fishing activity itself, could have affected Steller sea lion abundance 3 to 5 years later.

Only two of the significant correlations found in the parametric approach were detected using Kendall's rank correlation test. We conclude that any potential relationship between sea lion abundance and pollock harvest can not be properly determined from these data.


View Online  (.pdf, 668 KB).