Distribution of Rhinella granulosa: integrating bioclimatic envelopes and ecophysiological responses

Abstract

We used two modeling approaches (correlative and mechanistic) to build models of potential distribution of Rhinella granulosa. The model generated with several climatic variables (BIOCLIM1) was more conservative, a second with temperature only (BIOCLIM2) presented patches of discontinuous distribution and the mechanistic model indicated thermal habitats more suitable along an extensive continuous area. The BIOCLIM1 had more accuracy in predicting the potential distribution than BIOCLIM2. The BIOCLIM1 also had the best performance, followed by BIOCLIM2 and the mechanistic model. The difference among areas of potential distribution in models is probably due to the use of different predictor variables and may reflect limiting interespecific biotic interaction. These results provide basis for improving species distribution models.