Modeling the spatial distribution of a tropical dry forest tree facing climate change

Abstract

Species distribution models have become an essential tool for the selection of conservation areas. Given this perspective, this research aimed to analyze the dynamics of the fundamental niche of Ziziphus joazeiro Mart. facing climate change. The MaxEnt algorithm was used to correlate species points of occurrence with bioclimatic variables of four periods: Middle Holocene, current period, and the optimistic and pessimistic future scenarios of the HadGEM2-ES model of general atmospheric circulation. The accuracy of the predictions and the influence of the variables was evaluated using the area under the curve (AUC index) and the Jackknife test. The predictions showed high precision (AUC > 0.895), and the bioclimatic variables related to precipitation were the ones that most contributed to determining potential areas for the occurrence of the species. A reduction of more than 60% was observed in the area with climatic suitability for the occurrence of Z. joazeiro, over time. The range that comprises the central region of the distribution of the species stands out as the region with the greatest suitability, as projected for the current and future period, this being the area that should receive the most attention for the conservation programs of the species.