Extensive and rapid losses of sea ice in the Arctic have raised conservation concerns for the Pacific walrus (Odobenus rosmarus divergens), a large pinniped inhabiting arctic and subarctic continental shelf waters of the Chukchi and Bering seas. We developed a Bayesian network model to integrate potential effects of changing environmental conditions and anthropogenic stressors on the future status of the Pacific walrus population at four periods through the twenty-first century. The model framework allowed for inclusion of various sources and levels of knowledge, and representation of structural and parameter uncertainties. Walrus outcome probabilities through the century reflected a clear trend of worsening conditions for the subspecies. From the current observation period to the end of century, the greatest change in walrus outcome probabilities was a progressive decrease in the outcome state of robust and a concomitant increase in the outcome state of vulnerable. The probabilities of rare and extirpated states each progressively increased but remained <10% through the end of the century. The summed probabilities of vulnerable, rare, and extirpated (P(v,r,e)) increased from a current level of 10% in 2004 to 22% by 2050 and 40% by 2095. The degree of uncertainty in walrus outcomes increased monotonically over future periods. In the model, sea ice habitat (particularly for summer/fall) and harvest levels had the greatest influence on future population outcomes. Other potential stressors had much smaller influences on walrus outcomes, mostly because of uncertainty in their future states and our current poor understanding of their mechanistic influence on walrus abundance.