Brown bear ( Ursus arctos ) foraging in a mosaic of spatially discrete and variable habitats over 25 years of shifting Pacific salmon densities

• Published in: Canadian Journal of Zoology
• Main Authors: McElroy, Katherine N., Hilborn, Ray, Cunningham, Curry, Quinn, Thomas P.
• Other Authors: National Science Foundation Graduate Research Fellowship, Cooperative Institute for Climate, Ocean, and Ecosystem Studies
• Format: Article in Journal/Newspaper
• Language: English
• Published: Canadian Science Publishing 2024
• Subjects: Animal Science and Zoology; Ecology, Evolution, Behavior and Systematics; Ursus arctos
• Online Access: http://dx.doi.org/10.1139/cjz-2022-0178; https://cdnsciencepub.com/doi/full-xml/10.1139/cjz-2022-0178; https://cdnsciencepub.com/doi/pdf/10.1139/cjz-2022-0178

Many foraging models assume “perfect information” and “free movement” when describing predator foraging behavior, although this is rare in nature. Here, we quantified predation by brown bears ( Ursus arctos Linnaeus, 1758) on adult sockeye salmon ( Oncorhynchus nerka (Walbaum, 1792)) in a series of spatially proximate ponds that largely satisfied both assumptions. Salmon abundance varied among years, but pond area and depth were fixed, allowing us to examine interactions between prey abundance and habitat features. We applied versions of two models to 25 years of data on the number and proportion of salmon killed by bears, modifying these models to include habitat features and temporal variability. The functional response model with a year effect fit the data well, indicating bears could take almost all salmon in ponds when salmon were scarce, but bears were sated when salmon were abundant. The proportion of salmon killed by bears was similar across habitats after correcting for pond depth and area. Overall, bears foraged across all habitats but killed higher proportions of salmon in smaller and shallower habitats, consistent with ease of capture.