Modeling Gray Wolf Habitat in the Northern Rocky Mountains by Michael E. Houts

The reintroduction of the gray wolf into the Northern Rocky Mountains has created many difficult management issues; ranchers fear for their livestock, hunters worry about game populations, ecologists and biologists support the reintroduction, and resource management agencies are caught in the middle. To date (2000), the one hundred plus wolves in the Yellowstone National Park area and approximately 200 more in Idaho and Montana have been monitored and managed using action/reaction techniques. Consequently, as wolves disperse through these areas, little preparation or planning has been conducted due to a lack of information about wolf locations, movements, and habitat preferences. To improve the management of the gray wolf in the Northern Rocky Mountains, and to assist with pro-active planning, a geographic information system (GIS) was developed to create a wolf habitat suitability model. There were two main objectives to this project: the first was to identify the most effective mapping unit (Idaho Game Management Units, 33x33 km grid, 16x16 km grid) for modeling wolf habitat, and the second was to analyze four landscape variables (land cover, land ownership, road density, elk density) using logistic regression to predict wolf habitat suitability. Habitat models were built using data from Idaho, then the best model was tested using data from Montana and Wyoming for out-ofsample validation. Results indicated that a 33 x33 kilometer grid worked best for modeling wolf habitat, and that road density and land cover were the most influential landscape variables for determining wolf habitat suitability. Overall accuracies of 83%, 89%, and 75% were achieved for predicting the presence/absence of wolves in Idaho, Montana, and Wyoming, respectively. The resulting habitat suitability maps can assist management agencies in the identification of potential wolf habitats, areas where human/wolf conflicts are likely to occur, and areas that should be considered essential wolf habitat, thus contributing to the improved management and protection of the gray wolf in the Northern Rocky Mountains. Acknowledgments Research support and technical facilities were provided by the Kansas Applied Remote Sensing (KARS) program, a division of the Kansas Biological Survey, at the University of Kansas. Thanks to the administrative staff at KARS including Edward Martinko (Director), Kevin Price (Associate Director), Stephen Egbert (Research Scientist and principal advisor), Mark Jakubauskas (Research Scientist), and Jerry Whistler (Research Associate) for all their support and advice. Additionally, thanks to Dietrich Kastens, who helped me through many technical difficulties as well as Dana Peterson, Aimee Stewart, Jerry Griffith who helped me with the numerous “little things” that always came up. Thanks to the Idaho Department of Fish and Game, the Wyoming Department of Game and Fish, and the Montana Department of Fish, Wildlife, and Parks for their assistance in providing data regarding elk populations and current management practices. Finally, a special thanks to my fiancé, Jeannie Jackson, for putting up with my long hours of work and keeping me sane through the whole process.