Modeling the distribution and movement intensity of the Arabian Leopard Panthera pardus nimr (Mammalia: Felidae)

Abstract

This study attempts to link the distribution and movement rate of the critically endangered Arabian Leopard (Panthera pardus nimr) to environmental and anthropogenic features, and to identify environmental constraints and priority areas for the recovery of leopard in Arabian Peninsula. Generalized linear and additive models were used to fit leopard presence/absence locations to environmental and anthropogenic variables. Movement rates between the polygons of modeled leopard presence were inferred and mapped using the isolation-by-resistance model, where probability values of the species distribution model were treated as those of conductance. Our results suggest that currently the Arabian Leopard prefers to live and move in terrain that has high values of normalized difference vegetation index (NDVI) and is difficult for humans to reach. The current network of protected areas largely under-represents the species suitable habitat only covering its 11%, and most of the survey effort to detect and ensure the survival of the leopard in the peninsula has taken place outside the polygons identified by our models as core areas for the species. Our models coupled with existing data suggest the following scenario of the species biogeography: The Arabian Leopard accumulated genetic and phenotypic differences from its conspecifics at a series of glacial maxima during the last glacial period in the Yemeni refugium, from where it expanded elsewhere in the Holocene warming following the expansion of suitable landscape types. Humans expanded too, eventually restricting the source populations of the leopard to an area intersecting eastern Yemen and western Oman today. Our models may serve as a tool for planning future research and conservation for Arabian Leopard.