Figure 1. MSM estimates of spawning biomass of walleye pollock from the Eastern Bering sea (blue line) and projections of unfished and 40% unfished biomass from MSM (solid and dashed blue lines, respectively), Ecosim (orange lines), and the 2011 stock assessment (horizontal gray bar).
Multispecies Stock Assessment Modeling (MSM) is a three-species model of some of the most productive stocks and key predators in the Bering Sea – walleye pollock, Pacific cod, and arrowtooth flounder.
Their biological interactions may influence natural mortality estimates and subsequent harvest recommendations. Thus, MSM is a logical tool for implementing ecosystem-based management of fisheries resources in the Bering Sea.
MSM combines traditional catch-at-age stock assessment models with multispecies virtual population analysis models (MSVPA) in a statistical framework and uses abundance and diet data to estimate fishing mortality, recruitment, stock size, and predation mortality.
MSM typically models the latter as a series of functional bioenergetics responses to derive temperature-dependent predator rations. Since MSM can capture critical threshold effects that characterize many ecological interactions, such an approach also provides a statistical framework to evaluate and manage both the direct and indirect effects of fisheries harvest on multiple species.
Figure 2. Daily size-dependent ration estimates for walleye pollock, Pacific cod, and arrowtooth flounder from the eastern Bering Sea. Shaded gray polygons represent the mean proportional weight of stomachs (gray dots; g stomach / g body weight), as compared to digestion corrected values (green lines; "livingston"), von Bertalanffy consumption estimates from two different fitting methods (blue and red lines), and Wisconsin bioenergetic model estimates (solid red lines).
Additionally, we used projections of the model to derive biological reference points (BRPs) for various harvest control rule approaches under variable climatic conditions. In particular, we have generated projections for the three species under various target harvest rates including 1) no fishing and 2) fishing to 40% of the unfished spawning biomass.
Model predictions were compared to single species assessments from 2011 as well as Ecosim predictions for the Bering Sea (Fig. 1). We are currently synthesizing results for publication.
As part of the MSM modeling effort we have also been conducting comparative analyses of annual predator rations estimated from 1) digestive corrections applied to the data, 2) von Bertalanffy derived estimates of consumption, and 3) estimates from Wisconsin bioenergetic models for the three species (Fig. 2).
Various ration estimates are compared to data from the food habits database and provide indices of predation pressure over time. These results are part of a manuscript currently in preparation.
