AbstractsPsychology

The persistence of grizzly bears (Ursus arctos) in western Alberta is threatened by increasing human activities on the landscape. The Foothills Research Institute Grizzly Bear Program (FRIGBP) hypothesizes human-caused landscape change in Alberta causes long-term stress in individual bears, resulting in impaired biological functions and, when many bears are affected, decreased population performance. To facilitate the evaluation of individual grizzly bear health within the FRIGBP, the objective of my research was to develop and assess the usefulness of a health function score system for grizzly bears. From a large set of complex biological data collected from grizzly bears from 1999 to 2007, I merged 14 � constituent� variables into four health functions; growth, immunity, movement, and stress. For each health function, I calculated individual scores by adding ranked and weighted variable percentiles. I found that health function scores corresponded well with health status of individual bears based on values for multiple constituent variables. The score system facilitated quick screening of health in individual bears, identification of bears with reduced health, and comparison of health profiles between bears. I examined the usefulness of the score system by evaluating relationships presumed to exist under the working hypothesis of the FRIGBP. Results generated from health function scores were compared with those from constituent variable values using statistical and graphical techniques. I concluded that scores likely provided clearer depiction of wildlife health relationships than did constituent variables because they were not influenced by capture method, sex, or outlying observations. By using the score system, I found support for the proposed positive relationship between human-affected landscape condition and stress, but not for inverse relationships between stress and other health functions. The usefulness of the score system could be increased by minimizing use of redundant constituent variables, e.g., in growth and immunity, and removing the influence of potential confounding factors, e.g., capture.