Modeling the impacts of a changing and disturbed environment on an endangered beluga whale population

Climate change and disturbance from human activities are key threats facing many wildlife populations worldwide. The ability to quantify the effects of such threats on individual health and population dynamics is critical for effective management and conservation. We used Stochastic Dynamic Programming (SDP), a method for implementing state-dependent life history theory, to explore the impacts of changes in prey availability and anthropogenic disturbance on survival and reproductive success of Cook Inlet belugas (Delphinapterus leucas, CIB), an endangered and isolated population in decline. We predicted behavioral decisions (whether to forage, travel to a new location, or rest) of pregnant CIBs within a spatially and seasonally dynamic prey landscape. We used those decisions to explore time-activity budgets and spatial use under a variety of hypothetical environmental scenarios and estimate the resulting impacts on body condition and vital rates. In all scenarios, foraging activity was highest during summer to capitalize on abundant prey, which in the model was assumed to be eulachon (Thaleichthys pacificus) and salmonids (Oncorhynchus spp.). This resulted in large increases in blubber reserves that pregnant CIBs relied on during October to April when prey was assumed to be less abundant. Prey availability outside of summer months was still critical, as it either exacerbated or buffered against reductions in prey availability during summer months. Spatial predictions of habitat use indicated that pregnant CIBs should forage in areas used historically that now appear to be abandoned, suggesting that prey availability alone is unlikely to explain the recent range contraction of CIBs to upper Cook Inlet. Reductions in prey availability from late spring to early fall adversely affected vital rates, but intermittent disturbances that resulted in lost foraging opportunities, such as those caused by anthropogenic activities during the ice-free season, had little impact on body condition or vital rates if prey were abundant during the summer and early fall. Accurate assessment of the effects of anthropogenic disturbance on CIBs requires robust data on both disturbances and year-round prey availability, as intermittent disturbances adversely affected survival and reproductive success when they occurred in environments with reduced prey availability. Our model represents an initial effort to fill a critical information gap for informing CIB management decisions, providing insights into conditions under which reductions in vital rates might be expected and highlighting key data needed to increase the applicability of the model to this endangered population.