Evaluation of habitat patches importance to desert landscape connectivity for three fox species, using resistance kernel and graph network

Despite the importance of central Iranian desert as a refuge for threatened carnivores, conservation strategies for these species have been hindered by scarcity of data on distribution and habitat connectivity. The present study aimed to 1) determine the ecogeographical variables affecting distribution of three fox species including sand fox (Vulpes rueppellii), Blandford’s fox (Vulpes cana), and red fox (Vulpes vulpes); 2) identify important core habitats and dispersal corridors of the three fox species; and 3) evaluate habitat patch importance to landscape connectivity. At first step, ensemble models were built for each species using three distribution algorithms and 12 ecogeographic variables. Distribution of the foxes was affected by annual precipitation, shrubland density, human settlement density and topographic roughness. In the next step, negative exponential function was used to convert ensemble distribution maps to resistance surfaces. Core patches and significant corridors were predicted using resistance kernel approach. Most of the habitat core patches were located in Protected Areas (PAs), which could be related to high landscape resistance outside the areas under protection. Finally, the importance of identified core patches was evaluated using graph network. The patch importance to connectivity was significantly correlated with core extent, mean of relative density of dispersing individuals and probability of occurrence in patch. Results showed that the habitat quality indices are more effective than habitat quantity in predicting landscape connectivity. The obtaining results suggest that effective conservation of carnivores demands for an integrated landscape management aiming at functional connectivity among habitat patches.