Search for a command to run...
Changes in the abundance and distribution of wildlife populations are common consequences of historic and contemporary climate change. Some Arctic marine mammals, such as the polar bear (Ursus maritimus), may be particularly vulnerable to such changes due to the loss of Arctic sea ice. We evaluated the impacts of environmental variation on demographic rates for the Western Hudson Bay (WH), polar bear subpopulation from 1984 to 2011 using live-recapture and dead-recovery data in a Bayesian implementation of multistate capture-recapture models. We found that survival of female polar bears was related to the annual timing of sea ice break-up and formation. Using estimated vital rates (e.g., survival and reproduction) in matrix projection models, we calculated the growth rate of the WH subpopulation and projected population responses under different environmental scenarios while accounting for parametric uncertainty, temporal variation, and demographic stochasticity. Our analysis suggested a long-term decline in the number of bears from 1185 (95% Bayesian credible interval [BCI] = 993-1411) in 1987 to 806 (95% BCI = 653-984) in 2011. In the last 10 yr of the study, the number of bears appeared stable due to temporary stability in sea ice conditions (mean population growth rate for the period 2001-2010 = 1.02, 95% BCI = 0.98-1.06). Looking forward, we estimated long-term growth rates for the WH subpopulation of ~1.02 (95% BCI = 1.00-1.05) and 0.97 (95% BCI = 0.92-1.01) under hypothetical high and low sea ice conditions, respectively. Our findings support previous evidence for a demographic linkage between sea ice conditions and polar bear population dynamics. Furthermore, we present a robust framework for sensitivity analysis with respect to continued climate change (e.g., to inform scenario planning) and for evaluating the combined effects of climate change and management actions on the status of wildlife populations.