Patterns in size, growth, and condition of juvenile chum and pink salmon in the northeastern Bering Sea

Abstract

The Bering Sea has alternated between warm and cool spring thermal regimes, as defined by May sea surface temperature, and in recent years has remained in a "cool" state. Differences in spring thermal regime influence the timing of sea ice extent in the southeastern Bering Sea (SEBS) region, with warm springs facilitating early ice retreats and cool springs resulting in later ice retreat. A recent conceptual model for relating production to higher trophic levels in the SEBS proposes that during years of early sea ice retreat, phytoplankton blooms occur in warm water and support small, lipid-poor species of zooplankton. Conversely, years of late sea ice retreat results in an ice associated bloom that supports large, lipid-rich species of zooplankton. As a consequence the energy density of prey sources available to higher trophic levels is reduced during warm years and enhanced during cool years. While the northeastern Bering Sea (NEBS) has consistently supported an ice-associated bloom, it is likely that productivity in the SEBS influences trophic-level connections in the NEBS. In order to examine this possibility, we extended this conceptual model to juvenile salmon and compared size and condition of juvenile chum (Oncorhynchus keta) and pink (O. gorbuscha) salmon in the NEBS between spring thermal regimes of the SEBS. We hypothesized that juvenile salmon would be longer in warm years and more energy dense in cool years. In years with cool springs, pink salmon were shorter and chum salmon exhibited greater energy density, but no other aspects of size and condition differed significantly between spring thermal regimes. We further examined relationships of size, growth, and condition of juvenile salmon with environmental variables within the NEBS. For both species, length increased over the time of the surveys; longer individuals were caught at stations with greater bottom depths and in cooler sea-surface temperatures, while individuals with high length-corrected energy density were associated with cooler temperatures and shallower depths. We used insulin-like growth factor-1 (IGF-1) concentrations as an indicator of relative growth rate for fishes sampled 2009-2012 and found fish exhibited higher IGF-1 l concentrations between 2010-2012 than in 2009. IGF-1 concentrations were positively correlated with temperature for juvenile chum salmon and with depth and length for juvenile pink salmon. The consistent appearance of depth (indicating distance from shore) in the best size and condition models was interpreted to indicate that as juvenile salmon moved offshore, they were allocating more energy to growth than fat storage over the course of the surveys. The association of cooler temperatures with greater energy density and longer lengths may reflect direct effects of temperature on salmon physiology as well as indirect effects on food quantity or quality indirect. Overall, recent conditions of the NEBS appear to successfully contribute to the growth and condition of the juvenile chum and pink salmon. Finally, we compared indicators of energy allocation between even and odd brood-year stocks of pink salmon and found the even broodyear stocks were more energy dense while odd brood-year stocks exhibited higher growth rates. These results reflect differences in energy allocation between brood-year stocks of juvenile pink salmon and suggest that the two brood-year stocks may respond differently to changing climate.