Habitat Suitability modeling of Urial (Ovis orientalis arkal) in the Samelghan plain by using Maximum Entropy Method

Abstract

Habitat suitability models are useful tools for a variety of wildlife management objectives. Distributions of wildlife species can be predicted for geographical areas that have not been extensively surveyed. The basis of these models’ work is to minimize the relationship between species distribution and biotic and abiotic environments. For some species, there is information about presence and absence that allows the use of a variety of standard statistical methods, however, the absence data is not available for most species. Nowadays, the methods that need presence-only data are expanded. One of these methods is the Maximum Entropy (MaxEnt) modeling. The purpose of this study is to model the habitat of Urial (Ovis orientalis arkal) in the Samelghan plain in the North East of Iran with the MaxEnt method. This algorithm uses the Jackknife plot and percent contribution values to determine the significance of the variables. The results showed that variables such as southern aspects, Juniperus-Acer, Artemisia-Perennial plants, slope 0-5%, and asphalt road were the most important factors affecting the species’ habitat selection. The area under curve (AUC) Receiver Operating Characteristic (ROC) showed an excellent model performance. Suitable habitat was classified based on the threshold value (0.0513) and the ROC, which based on the results 28% of the area was a suitable habitat for Urial.