Demography and behavior of polar bears summering on land in Alaska

Abstract

Polar bears (Ursus maritimus) in the southern Beaufort Sea population (SB) are spending increased time on the coastal North Slope of Alaska between July and October (Gleason and Rode 2010). The duration spent on land by polar bears, satellite collared on the sea-ice in the spring, during the summer and fall has also increased (USGS, unpublished data; Figure 1). This change in polar bear ecology has relevance for human-bear interactions, subsistence harvest, prevalence of defense kills, and disturbance associated with existing land-based development [e.g., National Petroleum Reserve of Alaska (NPRA), Arctic National Wildlife Refuge (ANWR)], Native Alaskan communities, recreation (ANWR) and tourism (e.g., bear viewing in Kaktovik, AK). These activities have the potential to impact, in new ways, the status of the entire SB population. Concomitantly, the change in polar bear ecology will impact these human activities, and a base-line characterization of this phenomenon can better inform mitigation (e.g., industry permitting under the Endangered Species Act and Marine Mammal Protection Act). In this study we aim to characterize the demography, habitat-use, and aspects of foraging ecology and health of polar bears spending fall on land. The SB population is characterized by a divergent-sea ice ecology, where polar bears typically spend most of the year on the sea-ice, even as the pack ice retreats northward, away from the coast, to its minimal extent in September (Amstrup et al. 2008; Durner et al. 2009). From 2000 – 2005, using coastal aerial surveys, Schliebe et al. (2008) observed between 3.7 and 8% of polar bears from SB (~ 60 – 120 of 1526, Regher et al. 2006) on land during the autumn. Sighting probability was not estimated in these surveys, and therefore the numbers represent minimum numbers of bears on land. Our analysis of USGS data suggest an annual average of 15% (± 3%, SE) of polar bears satellite-tagged on the spring-time sea ice (total n = 18 of 124 satellite tags, 2003 – 2009) come to land during July – October. Based on these data, and an assumption that bears satellite-tagged on the spring time sea ice are representative of the entire SB population of independent bears, there would be an average of 230 bears on land each fall. In contrast to the SB population, in five of the world’s 19 polar bear populations (Obbard et al. 2010), polar bears spend significant periods of time on land (1 – 5 months) when ice completely melts. In these seasonal-ice populations (Amstrup et al. 2008), polar bears are largely in a hypophagic condition (e.g., Hobson et al. 2009), relying on fat stores from the spring hyperphagic season, when ringed seals (Phoca hispida) pup. In general, these seasonal-ice populations are demographically productive (Taylor et al. 2005), although recently an increase in the ice-free season has resulted in a population decline in western Hudson Bay (Stirling et al. 1999; Regehr et al. 2007). There have been measured declines in the body condition and productivity of polar bears in SB, and changes in these parameters have been linked to declining optimal ice habitat (e.g., Durner et al. 2009; Regehr et al. 2010). We do not understand the relationship between land-use and the overall status of the population. Individual polar bears that use land may have increased or decreased fitness, in comparison to polar bears that remain on ice in the autumn. This project, which focuses on the biology of animals that spend time on-shore, will help address this question. This project is funded by the Bureau of Ocean Energy Management (BOEM) under Agreement No. M09PG00025 and the USGS Outer Continental Shelf Program (OCS) for FY 2009-2014. Parts of this study are also funded by US Fish and Wildlife Service, Office of Marine Mammals Management; the Bureau of Land Management; and the North Slope Borough, Department of Wildlife Management. This report is comprehensive, describing results for achieving the overlap