The vulnerability of endemic vertebrates in Sri Lanka to climate change

Abstract

Rapid climate change and ongoing habitat destruction pose a serious threat to global biodiversity. Understanding how species shift their geographical distributions in response to climate change is important for planning conservation actions for the biodiversity of isolated islands like Sri Lanka. Here, we used Maximum Entropy (MaxEnt) modeling to predict current and future (by 2100) distributions of 233 vertebrate endemics to Sri Lanka under three general circulation models of the atmosphere (GCMs) and three shared socioeconomic pathways (SSPs). We used six bioclimatic layers, together with land use, human population, elevation, and the distance to water, as environmental factors for SDMs. Resultant maps were used to calculate the area of habitat (AOH) of each species. Using general linear mixed models, we identified a significant influence of GCM, SSP, and current elevation on the change in AOH. The southwestern wet zone and the montane areas of Sri Lanka were found to be the most suitable regions for the species currently and in the future. However, all climate change scenarios indicated endemic species extinctions, which may be proportionally fewer for mammals and birds, but are expected to occur for all SSPs in amphibians and reptiles, ranging from 1.4 % to 22.5 % of these taxa in SSP5. Also, species declines were specific and severe at higher elevations. Our study highlights the risk to montane endemic vertebrates, yet given the vulnerability of the wet zone to urbanization, even lowland endemic species face an uncertain future.