CLOUD-BASED HABITAT SUITABILITY MAPPING FOR CLIMATE CHANGE ADAPTATION OF Cinnamomum parthenoxylon IN CENTRAL HIGHLANDS, VIETNAM
DOI:
https://doi.org/10.71254/85btwn89Keywords:
Cinnamomum parthenoxylon, machine learning, Google Earth Engine, ecological suitability, species distribution modelingAbstract
Cinnamomum parthenoxylon (Jack) Meisn. is a tree species belonging to the genus Cinnamomum that is currently endangered in Vietnam. In this study, we developed a habitat suitability model and projected future climate change impacts on the distribution of species using 15 geospatial variables available on the Google Earth Engine platform. Future distribution projections were made for the period 2080 - 2100 under six medium-emission climate change scenarios (GISS-E2-1-G, BCC-CSM2-MR, INM-CM5-0, MPI-ESM1-2-HR, MRI-ESM2-0, and UKESM1-0-LL). The results identified ecologically suitable areas for C. parthenoxylon in Central Highlands region. The MaxEnt algorithm, based on presence–pseudoabsence data, was applied to model habitat suitability zoning for the species. Model performance was evaluated using the Area Under the Curve (AUC) metric. The results revealed that Dak Nong province had over 157000 ha of highly suitable habitat, while Lam Dong province had more than 151000 ha classified as very highly suitable for C. parthenoxylon. Climate change projections suggested that some areas currently identified as highly or very highly suitable may experience a decline in suitability in the future. The consistency observed across the six climate scenarios enhances the reliability of the projections. These findings provide a critical scientific basis to support the conservation and management of C. parthenoxylon in the context of increasing climate change impacts in Vietnam’s Central Highlands.



