Climate change is fundamentally altering habitats, with complex consequences for species across the globe. The Arctic has warmed 2–3 times faster than the global average, and unprecedented sea ice loss can have multiple outcomes for ice-associated marine predators. Our goal was to assess impacts of sea ice loss on population-specific habitat and behaviour of a migratory Arctic cetacean.
Using satellite telemetry data collected during summer-fall from sympatric beluga whale (Delphinapterus leucas) populations (“Chukchi” and “Beaufort” belugas), we applied generalized estimating equations to evaluate shifts in sea ice habitat associations and diving behaviour during two periods: 1993–2002 (“early”) and 2004–2012 (“late”). We used resource selection functions to assess changes in sea ice selection as well as predict trends in habitat selection and “optimal” habitat, based on satellite-derived sea ice data from 1990 to 2014.
Sea ice cover declined substantially between periods, and Chukchi belugas specifically used significantly lower sea ice concentrations during the late than early period. Use of bathymetric features did not change between periods for either population. Population-specific sea ice selection, predicted habitat and the amount of optimal habitat also generally did not change during 1990–2014. Chukchi belugas tracked during 2007–2012 made significantly more long-duration and deeper dives than those tracked during 1998–2002.
Taken together, our results suggest bathymetric parameters are consistent predictors of summer-fall beluga habitat rather than selection for specific sea ice conditions during recent sea ice loss. Beluga whales were able to mediate habitat change despite their sea ice associations. However, trends towards prolonged and deeper diving possibly indicate shifting foraging opportunities associated with ecological changes that occur in concert with sea ice loss. Our results highlight that responses by some Arctic marine wildlife can be indirect and variable among populations, which could be included in predictions for the future.