Predictive Model for the Probability of Occurrences of Asiatic Black Bear (Ursus thibetanus japonicus) using a Logistic Regression Model with Environmental Indices

Abstract

Habitat estimation of Asiatic black bear is considered highly significant from the perspective of effective habitat management, because this species is known both as a destructive animal as well as an umbrella species. In this study, we developed a predictive model for the probability of occurrences of Asiatic black bear using environmental indices. Based on this research, the optimal logistic regression model was log(p/(1-p))=(-1.486e+01)+(7.335e-04)*x1+(9.470e-03)*x2. where x1 is the distance to small paths (m), x2 is the altitude (m), and p is a probability of occurrence of Asiatic black bear in this study area. This model is applicable for habitat management. Additional research will examine the application of this model to different local Asiatic black bear populations in other regions.