Spatio-temporal variation in potential habitats for rare and endangered plants and habitat conservation based on the maximum entropy model

Abstract

Rare and endangered plants (REPs) act as key indicators for species habitat priorities, and can thus be critical in global biodiversity protection work. Human activities and climate change pose great threats to REPs, so protection should be a top priority. In this study, we used the maximum entropy model (Maxent) to identify current and future (2050) potential habitats of REPs in the Xishuangbanna tropical area of China. We compared potential habitats with existing protected areas (PAs) in gap analysis, and used a transfer matrix to quantify changes in potential habitats. By comparing the potential distribution obtained with existing land use and land cover, we analyzed the impact of human-dominated land use changes on potential habitats of REPs and identified the main habitat patch types of REPs. The results showed that the current potential habitat area of hotspots is 2989.85 km2, which will be reduced to 247.93 km2 by 2050, accounting for 15.60% and 1.29% of the total research area, respectively. Analysis of land use and land cover showed that rubber plantation was the human-dominated land use posing the greatest threat to potential habitats of REPs, occupying 23.40% and 21.62% of current and future potential habitats, respectively. Monsoon evergreen broad-leaved forest was identified as the main habitat patch type for REPs in Xishuangbanna and occupied the highest proportion of potential habitat area. Gap analysis showed that only 35.85% of habitat hotspots are currently included in existing PAs and that this will decrease to 32.26% by 2050. This emphasizes the importance of protecting current and future potential habitats of REPs in a dynamic conservation approach that can adapt to changes in future climate and human activities.