Comparative Patterns of Predation by Cougars and Recolonizing Wolves in Montana's Madison Range

• Published in: The Journal of wildlife management Vol. 71; no. 4; pp. 1098 - 1106
• Main Authors: Atwood, T.C, Gese, E.M, Kunkel, K.E
• Format: Publication
• Language: English
• Published: 2007
• Subjects: antipredator behavior; Canis lupus; Cervus elaphus; deer; elks; Odocoileus hemionus; Odocoileus virginianus; predation; predation refugia; predation risk; predator-prey relationships; prey selection; Puma concolor; refuge habitats; species reintroduction; wolves; Montana

Numerous studies have documented how prey may use antipredator strategies to reduce the risk of predation from a single predator. However, when a recolonizing predator enters an already complex predator-prey system, specific antipredator behaviors may conflict and avoidance of one predator may enhance vulnerability to another. We studied the patterns of prey selection by recolonizing wolves (Canis lupus) and cougars (Puma concolor) in response to prey resource selection in the northern Madison Range, Montana, USA. Elk (Cervus elaphus) were the primary prey for wolves, and mule deer (Odocoileus hemionus) were the primary prey for cougars, but elk made up an increasingly greater proportion of cougar kills annually. Although both predators preyed disproportionately on male elk, wolves were most likely to prey on males in poor physical condition. Although we found that the predators partitioned hunting habitats, structural complexity at wolf kill sites increased over time, whereas complexity of cougar kill sites decreased. We concluded that shifts by prey to structurally complex refugia were attempts by formerly naive prey to lessen predation risk from wolves; nevertheless, shifting to more structurally complex refugia might have made prey more vulnerable to cougars. After a change in predator exposure, use of refugia may represent a compromise to minimize overall risk. As agencies formulate management strategies relative to wolf recolonization, the potential for interactive predation effects (i.e., facilitation or antagonism) should be considered.