Ecosystem Modeling in Support of Alaskan Fishery Management from the Aleutians to the Arctic

Mathematical modeling of marine food webs has been an area of active research in Alaska, particularly with respect to the fisheries management. Food web analyses have played a role in managing some single species fisheries, in estimating overall yields that may be sustainably extracted from ecosystems, and more recently in modeling and predicting future ecosystem responses to climate. Here, I review a suite of ecosystem models currently used or in development for use in Alaska. Static food web models were first developed in the 1970s for the Bering Sea, and have since been developed for the Aleutian Islands, the Gulf of Alaska, and most recently for the Chukchi Sea. Dynamic multispecies models address key species interactions in the Bering Sea and in the Aleutian Islands, while full dynamic food web models are used in Bering Sea and Gulf of Alaska ecosystem assessments and management analyses. The newest and most spatially resolved model is FEAST (Forage-Euphausiid Abundance in Space and Time); a 3D, 10km resolution model of fish and zooplankton of the Bering Sea built on the ROMS (Regional Oceanographic Model) platform. The fish component models size-based interactions between key fish and zooplankton species of the Bering Sea based on both the results of the BSIERP field work and data from the collection of over 300,000 fish stomachs over the last 30 years in the region. The model includes a component for testing management strategies and for spatial prediction of fisheries and fish production given future climate (IPCC) scenarios as input drivers. With all of these models, we focus on predictability: using the wide range of resources at our disposal today, what’s possible, what’s missing, what predictions are robust, and what surprises we might have in store, as we move to attempting to forecast longer-term (20-50 year) outlooks for marine ecosystems.