COMPARISON OF PREDICTOR SELECTION PROCEDURES IN SPECIES DISTRIBUTION MODELING: A CASE STUDY OF Fagus hayatae

Abstract

Selecting predictors for species distribution models (SDMs) is a major challenge. In this study, we evaluated a comprehensive set of 62 environmental predictors that may be related to the occurrence of Fagus hayatae. We modeled F. hayatae as a case study to compare model performance through different environmental predictor subsets according to three selection procedures, namely correlation coefficients between predictors, contribution level of predictors, and expert choice of biologically relevant predictors. The three selection procedures provided satisfactory results with high performance using about 10 valid predictors but had their respective limitations. Consequently, we suggest an approach of predictor selection. Accordingly, the first step was identifying and eliminating ineffective variables with nonidentifiability, such as coldness index, by using bivariate scatterplots. Next, correlation coefficients between other candidate predictors were calculated. Finally, predictors were selected within lower correlated (|r| < 0.7) candidate subsets on the basis of high contribution level predictors and expert knowledge of biologically relevant predictors for target species.