Predicting habitat suitability for conservation of IUCN- red listed medicinal species Chloroxylon swietenia DC., in Tamil Nadu, India using ensemble modeling

Abstract

Climate change is profoundly affecting biodiversity, highlighting the need for a comprehensive understanding of its impacts on ecosystems and species to inform sustainable conservation strategies. This study focuses on Chloroxylon swietenia DC., an endemic species listed as Vulnerable under the IUCN Red List (criterion A1c), to assess its habitat suitability under various climate change scenarios using ensemble modeling. We incorporated 19 bioclimatic variables, four soil-related variables (bulk density), and five land use/land cover variables (tree cover, cultivated land, water, and built-up areas) to model the species' habitat suitability across four Shared Socio-economic Pathways (SSPs) for the periods 2021–2040, 2041–2060, 2061–2080, and 2081–2100, using two different climate projections: BCC_CSM2-MR and EC-Earth3-Veg. Among the algorithms tested, the Random Forest model exhibited the highest performance. The eight climatic variables that contributed most significantly to the model were Bio1 (annual mean temperature), Bio3 (isothermality), Bio4 (temperature seasonality), Bio7 (temperature annual range), Bio13 (precipitation of wettest month), Bio15 (precipitation seasonality), Bio18 (precipitation of warmest quarter), and Bio19 (precipitation of coldest quarter), along with one soil variable (bulk density) and four land use/land cover variables (tree cover, cultivated land, built-up areas, and water). The ensemble model identified the Western and Eastern Ghats, sacred groves, and scrub forests of Tamil Nadu as highly suitable habitats for C. swietenia. Notably, the habitat suitability range of C. swietenia increased in the EC-Earth3-Veg projection, whereas the BCC_CSM2-MR projection indicated a significant reduction in suitability at SSP5–8.5. In contrast, the EC-Earth3-Veg projection showed maximum habitat expansion between 2041 and 2060 and 2081–2100 across all SSPs. Notably maximum gain noticed in weighed mean ensemble type than the committee averaging. Furthermore, the results indicate an increase in the species' population range in the Western Ghats and its neighboring region and reduction noticed in the Eastern ghats region.