Walleye pollock (Theragra chalcogramma) has a central role in the eastern Bering Sea (EBS) food web. It is a key forage species for many intermediate and upper trophic level predators and is the dominant consumer within the EBS food web. Pollock diet composition is known to differ with pollock size and location in the EBS, the geographic distribution and size composition of the EBS pollock stock changes over time, and the sampling of pollock stomachs varies among years in intensity and geographic distribution. Weighting diet information by stomach fullness and biomass for six size-categories (0-19, 20-29, 30-39, 40-49, 50-59, and > 60 cm fork length, (FL)) was used to provide a consistent average diet description of walleye pollock in the bottom trawl survey from 1987 through 2011. An index of partial fullness was also calculated for major prey over this time-period.
Partial fullness is an indicator of the average relative consumption of each prey type in each year (Fig. 1). The average index was calculated and plotted for euphausiids and copepods for three time periods – 1987-92, 1993-2004 and 2005-11, with the middle period being 12 years of continuous below average (0.22) partial fullness values for copepods. Partial fullness values for euphausiids were found to be serially random (P > 0.5), the probability that the partial fullness values for copepods were serially random was lower (P = 0.10), and the difference between the partial fullness values of euphausiids and copepods were found to be serially non-random (P = 0.02).
Hyperiid amphipods were not important in the diet of EBS pollock in most years, but in 2010 they were the dominant prey type (Fig. 1). The majority of these amphipods were likely Themisto libellula based on the locations and water temperatures where they were consumed. Themisto libellula is associated with Arctic water masses, and in the Barents Sea the highest concentrations are found near the Polar Front where the majority of reproduction and production of this species likely occurs due to the greater amount of food available to them compared to the Arctic Ocean. We speculate that recent conditions of cold water temperatures over the EBS shelf coupled with early ice-melt in the Arctic or more extensive ice-free zones in the Arctic may contribute to conditions that have favored the reproduction and survival of T. libellula. This may provide a vector for the enhanced retention of primary production in the pelagic zone in the northern Bering Sea and could have an effect on benthic-feeding invertebrates, fishes, and marine mammals.
The relationship between average annual weight composition (%W) of copepods and euphausiids to the average annual stomach fullness (%bw) varies by pollock size (Fig. 2). Each trend-line extends over the range of %bw values observed over the 25 years for each size category. The correlation between %bw and (arcsin transformed) %W of copepods in the diet was generally positive, especially for 30-39 cm pollock (r2 = 0.20, P < 0.05) and 40-49 cm pollock (r<2 = 0.15, 0.10 > P > 0.05). The correlation between %bw and %W of euphausiids in the diet was positive for the largest and smallest size groups of pollock (r2 = 0.05, 0.20 > P > 0.10; r2 = 0.01, P > 0.50; respectively), but negative for other size groups, especially 40-49 cm pollock (r2 = 0.20, P < 0.05). For 40-49 cm pollock, euphausiids became more important in the annual average diet when the annual average stomach fullness was lowest. Euphausiids may provide a more consistent food source from year to year than other potential prey of intermediate sizes of pollock. This may be similar to the role of euphausiids as a more consistent, baseline food source in seasons when other prey becomes less available.
Euphausiid biomass estimates have recently become available for 2004 and 2006-10 in the EBS (and include some hyperiid amphipod biomass). The increase in euphausiid biomass over this period was reflected in the %W of euphausiids in the diets of pollock ≥ 60 cm FL and possibly pollock < 20 cm FL (Fig. 3). Although correlations were not significant, the positive correlation between euphausiid biomass and %W of euphausiids in the diets of the smallest and largest size groups in the EBS, had r2 values of 0.41 and 0.61, respectively.
The strongest trophic link between pollock and euphausiids appears to be limited to the smallest and largest pollock while intermediate sizes of pollock appear to have a closer link to copepods in the summer. Size-specific relationships between pollock and specific subsets of the zooplankton community should improve the interpretation of patterns in pollock abundance and distribution whether due to interannual changes in physical conditions or longer-term climate change.
