Mixing It Up In Alaska: Habitat Use Of Adult Female Steller Sea Lions Reveals A Variety Of Foraging Strategies

From 2010 to 2015, satellite transmitters were deployed on 16 adult female Steller sea lions (AFSSLs; Eumetopias jubatus) in three regions of Alaska because there is limited information regarding the habitat use of this age class during winter and populations have yet to recover in western Alaska. Two approaches were used to assess how static (distance to shore, sea lion site, and continental shelf break, presence on/off the continental shelf, and bathymetric depth and slope), dynamic (proportion of daylight, fraction of lunar illumination, chlorophyll-a, wind speed, sea surface height, eddy kinetic energy, and sea surface temperature), and other (region, distinct population segment, and season) covariates affected the habitat use of AFSSLs. Multimodel inference was first used to examine diving behaviors (mean and maximum dive depths, dive frequency) with respect to covariates using linear mixed-effects models, whereas single model inference was used to examine kernel density estimates (KDEs) of individual monthly utilization distributions (n = 74) in western Alaska with respect to environmental covariates using generalized additive models. Additionally, weighted coefficients from these models were examined for the population as a whole, within each individual, between regions, and across monthly scales. Comparisons of foraging behaviors of AFSSLs over time and space revealed pronounced individual variability within overall broader patterns. Response variables of most models were related to various combinations of predictor variables, but distance to shore was the most influential variable across all models. As expected with a non-migratory central place forager, frequency of diving and KDEs were greater on the shelf and near shore, though maximum dive depths increased with distance to shore. Interaction effects (proportion of day light*on/off shelf) observed for mean dive depths suggested AFSSLs were feeding on benthic species when in shelf waters near shore, whereas they were likely feeding on vertically migrating prey species while off-shelf. Relationships regarding diving behaviors and KDEs of AFSSLs relative to dynamic oceanographic variables were not as prominent as those observed for static environmental variables, though some signals were apparent at different scales. Overall, static environmental features likely provided more consistent sources of habitat for prey resources, thereby making them more predictable for AFSSLs.