Forecasting Changes in Habitat Use by Bowhead Whales in Response to Arctic Climate Change

Bowhead Whales

PI: Dr. Elizabeth Holmes

Co-Investigator for APL: Jinlun Zhang
The effects of climate change are projected to be disproportionately pronounced in polar regions, where changes in the extent of Arctic sea ice will have an effect on all trophic levels. The endangered bowhead whale (Balaena mysticetus) is one of the largest animals in the Arctic, yet they feed on some of the smallest Arctic animals, zooplankton. Therefore, physically-induced bottom-up changes may be quickly reflected in the distribution of bowhead whales. Some of the important threats to bowhead whales include shipping and offshore oil drilling. Loss of sea ice in the Arctic has the potential to increase negative interactions with these threats, as Arctic areas become more accessible to vessels and oil exploration. The objective of our work is to make forecasts of bowhead whale habitat on seasonal to decadal time scales using output from large-scale models of zooplankton biomass and sea-ice extent. The ability to identify the range of large marine mammals will be essential for marine conservation planning in a changing Arctic environment.

Our proposed work builds upon a forecasting system developed for a similar baleen whale species, the North Atlantic right whale (Eubalaena glacialis) (Pendleton 2010, Pendleton et. al. 2009, Pershing et al. 2009, Pershing et al. 2009). This system, developed with funding from the NASA Applied Sciences program, generated weekly predictions, at regional and local spatial scales, of potential right whale habitat. Predictive maps were generated from satellite-derived sea surface temperature and chlorophyll, a satellite-driven zooplankton abundance (right whale prey) model, bathymetry, and aerial and acoustic detections of whales. Predictive accuracy of habitat maps was good (Pendleton et. al. 2009). This prototype system was used to assess the impact of moving shipping lanes in order to reduce risk of whale-vessel collisions, a primary threat to right whales.

We propose to extend this approach to bowhead whales by using output from a currently supported NASA project (Jinlun Zhang, The pan-Arctic Biology/Ice/Ocean Modeling and Assimilation System (BIOMAS) model). BIOMAS has generated predictions of zooplankton and sea-ice extent, along with a host of other environmental variables, from 1988 to present. Model output from the time and location of bowhead whale sightings, from aerial and acoustic surveys, will be used to train an environmental niche model (ENM) that characterizes bowhead whale habitat. Two types of ENM techniques will be used, boosted regression trees and maximum entropy density estimation, to produce seasonal and long range (5-10yr) estimates of bowhead whale distribution in the Chukchi-Beaufort Seas. Once predictions have been validated, we will use forecasts for bowhead whale occurrence to help evaluate conservation actions and develop strategies for minimizing human-whale interactions, especially as Arctic sea-ice extent and whale populations change in the coming century.