The spatial organization and habitat selection patterns of barren-ground grizzly bears in the central Arctic

I studied the population delineation, hierarchical habitat selection, home range requirements, and denning habits of barren-ground grizzly bears (Ursus arctos) in Canada's central Arctic. To meet study goals, I tracked 81 bears equipped with satellite radio-collars in a study area of approximately 235,000 km2, centred 400 km northeast of the city of Yellowknife, Northwest Territories. I identified three populations of grizzly bears in the study area using multivariate cluster analysis of movement data and population range analyses. High exchange among population units for both females and males, however, suggest that identified grizzly bear population units cannot be managed independently from one another. I documented highly selective patterns of habitat selection by grizzly bears in a central, 75,000 km2 portion of the study area. Using resource selection functions, I examined habitat selection at the level of the home range (second order selection). Coverage of habitat was determined from Landsat Thematic Mapper scenes. The general pattern was for bears to possess home ranges, relative to the study area, that contained preferential amounts of esker habitat, tussock/hummock successional tundra, lichen veneer, birch seep, and tall shrub riparian areas over other habitat types. I also examined habitat selection at a finer level of selection (third order selection), whereby habitat use was determined from individual satellite telemetry locations and compared to the availability of habitats within home ranges of individual animals. Overall, esker and riparian tall shrub habitats were the most preferred habitats by bears throughout the year. Annual ranges of males ('X' = 7,245 km2) were significantly larger than the annual ranges of females ('X' = 2, 100 km2). Annual ranges are the largest ranges yet reported for grizzly bears in North America. Multiple regression revealed that ranges increased in size as the proportional amount of poor bear habitat in the environment, supplying constant amounts of quality habitats. Compared to the proportional availability of habitat types in the study area, esker habitat was selected more than expected by chance. The majority of bears emerged from their dens in the first week of May. Den entrance occurred primarily in the last two weeks of October