COS 103-2
Carnivore mortality risk at multiple scales: Evaluating the influence of human activity on source-sink dynamics

Thursday, August 14, 2014: 8:20 AM
314, Sacramento Convention Center
Rae J. Wynn-Grant, Ecology, Evolution, and Environmental Biology, Columbia University, New York, NY
Background/Question/Methods

Understanding the impact of human activities in shaping source-sink dynamics across landscapes is of critical importance in practical applications of ecology. The American black bear (Ursus americanus) is a large-bodied carnivore often involved in human-carnivore conflict across North America.  In the Western Great Basin, human-induced mortality to black bears is frequent, with up to 60 incidents per year.

Analyses of mortality risk are frequently used to demonstrate the relationship between human land use and source-sink dynamics in wildlife populations. These analyses, however, are typically carried out with anthropogenic landscape variables at coarse spatial scales, thus failing to uncover the more complex nuances of human influence on natural systems.  We were interested in testing the impact of spatial and temporal scale on predictions of mortality risk for black bears in the Western Great Basin. We used locations and descriptions of human-induced mortality to black bears from 1997-present along with environmental and anthropogenic landscape variables to construct models indicating probability of black bear mortality across the Western Great Basin.  One of the parameters used in the final model was a previously constructed human influence index, taking into account seven indicators of human influence of biological importance to large carnivores.

Results/Conclusions

Course-scale evaluations of mortality risk indicate highest probability of carnivore mortality directly related to human population density and distance to major road, which is keeping with predictions based on similar previous studies.  However, analyzing mortality risk using parameters at fine spatial and temporal scales suggested that metrics such as human influence index score, land use category, and frequency of human recreation in backcountry forest were stronger predictors of mortality risk than both human population density and distance to major road. The general finding is that type, intensity, and configuration of human activity create a non-linear gradient of mortality risk for black bears in the Western Great Basin. These findings support the advancement of risk map efforts with use of multi-scale anthropogenic parameters in evaluations of wildlife behavioral patterns and population persistence in shared human-wildlife landscapes.