干旱区生态环境知识资源中心

Estimation of connectivity for multiple species could increase the efficiency of resource management and elucidate trade-offs among maintenance of connectivity for different taxa. We identified potential areas of high connectivity for 5 species of mammals on the Navajo Nation and adjacent lands in Utah, Arizona, and New Mexico, USA: mountain lion (Puma concolor), mule deer (Odocoileus hemiorns), desert bighorn sheep (Ovis canadensis nelsoni), American black bear (Ursus ataericanus), and pronghorn (Antilocapra americana). These species were identified by the Navajo Nation as relevant to the benefit of their present and future generations. We used telemetry data to calculate utilization distributions, derive model permeability (the probability that a given location facilitates animal movement), and assess connectivity (the realization of permeability across a landscape) for desert bighorn sheep, black bear, and pronghorn. We based models of connectivity for mountain lion and mule deer on expert-identified environmental variables and corresponding permeability values. We used Circuitscape software to model omnidirectional connectivity for each species, and then used maps of connectivity to identify potential dispersal areas. Different environmental features were associated with connectivity for each species. The rank correlation between the geographic distribution of connectivity for pairs of species ranged from -0.45 to 0.95. All but one of the estimated pairwise overlaps in potential dispersal areas were greater than would be expected if dispersal areas for each species were independent. The percentage of overlap generally decreased as a greater number of species was considered, but was greater than expected in 6 of 10 cases for 3 species and all cases for 4 or 5 species. Potential dispersal areas for all 5 species occurred within 83 km(2) of the approximately 72,000-km(2) analysis area. Our work illustrates use of a flexible method for estimating connectivity and potential dispersal areas, particularly where data on the distribution and movements of populations are limited.