Species distribution modelling of medicinal and culturally plants of Rwanda

Abstract

Species overexploitation poses a critical threat to biodiversity, particularly in the Global South where reliance on natural resources for livelihoods is substantial. In Africa, a significant portion of the population relies on medicinal plants and animals for primary healthcare. With Africa's escalating population growth, there is an urgent need to understand the potential consequences on medicinal plant species, as increased harvesting may heighten extinction risks. Therefore, this study aims to address this concern by modeling the potential distribution of 14 medicinal and culturally significant native plant species in Rwanda. Using spatial occurrence records obtained from the Global Biodiversity Information Facility (https://www.gbif.org/), and climatic predictors downloaded from WorldClim (https://www.worldclim.org/), I fitted species distributions models (SDMs) using bioclim envelope algorithms in the dismo R package. Model performance was evaluated using both area under the curve (AUC) of the receiver operating characteristic plot and True Skill Statistics (TSS). High-performing models were integrated into weighted average ensemble models to predict potential geographic distributions of each of the 14 species within Rwanda. The predicted suitable areas were refined by excluding urbanized and degraded areas, providing a comprehensive overview of habitat suitability for each species. Response curve method was used to assess the contribution of the predictor variables to the ecological niche, while the niche breadth of all species was calculated. The results of this study have significant application for conservation efforts in Rwanda by highlighting existential threats to medicinal plant species and identifying areas for conservation prioritization