| Wise land-use planning is particularly important in the urban-wildlife interface where conservation and development efforts come into conflict. Due to limited empirical information about landscape composition and wildlife movement, conservation professionals often rely on species distribution models for strategic planning. Maxent generates distribution models using species occurrence and environmental datasets. Although the built environment can influence the distribution of species, human land use is rarely included in these models. Does including land-use improve the predictive ability of a species distribution model? To address this question we used Maxent to construct and compare models with and without land-use variables for gray foxes, carnivores sensitive to the wildlife-urban interface. We created both models using a combination of gray fox presence-only (n=209) and presence/absence (n=52) records, and 9 environmental and anthropic land-use variables. Including land use resulted in a more accurate model, with land use ranked in the top 3 variables and explaining 60% of the model's importance. Gray fox distribution was correlated with locations in low-density development and closer to roads. Species distribution models that include land use can be used to target priority habitat for conservation, thereby minimizing current conservation conflicts and maximizing resilience to future climate change.
|