InforSim has developed innovative modeling techniques for analyzing the migratory processes of large fish and marine mammals based on electronic tag records of their movements and synoptic environmental conditions. Using dynamic Bayesian networks and nonlinear programming methods, the movement data are synthesized and correlated with environmental features and used to predict population-level movement dynamics. A preliminary model has been used to study the migration of northern elephant seals from central California to Alaska (see illustration) and the movements of crabeater seals in the Antarctic. Working with the researchers at the University of California Santa Cruz, this model will be enhanced to treat life cycle processes as well as movements over a span of years in order to statistically predict the concentrations of marine mammals and large pelagic fish under alternative environmental scenarios.
	Original techniques have also been developed for handling parameter uncertainty in multi-species population models and predicting population changes in response to various natural and man-induced perturbations. These simulation models have been used to describe:       • movements of northern bluefin tuna between eastern and western Pacific boundary regions and their exploitation by         international fisheries       • collapse of the sardine population off California during 1930-1960 due to a combination of over fishing and natural         environmental fluctuations       • impacts to local reef and bottom fish caused by Ocean Thermal Energy Conversion (OTEC) plants operating off western         Oahu in Hawaii       • recovery of the Pacific herring population from the Exxon Valdez oil spill in Prince William Sound, Alaska