Habitat Suitability Modelling for Tree Species in Binsar Wildlife Sanctuary, Almora, Uttarakhand Using Maxent Algorithm

To predict habitat suitability of tree species in the study area, we used Species Distribution Modeling (SDM) with MaxEnt algorithm. This research hopes to enrich conservation planning and ecological decision-making in the region by linking species occurrence data with bioclimatic, topographical and ecological variables. The main aim is to detect potential suitable habitats for six tree species like Quercus leucotrichophora, Pinus roxburghii and Rhododendron arboreum etc. Binsar Wildlife Sanctuary, study area is rich in flora and fauna with varied topography thus an ideal location for conducting ecological studies. Data for this study were collected through extensive field surveys scattering over the entire reserve to record the occurrences of tree species, as well as by compiling bioclimatic and topographical data in a GIS context. The MaxEnt model was selected due to its ability to deal with presence-only data robustly and offer the possibility of probabilistic predictions of species distributions. We measured the model performance with AUC (Area under Curve) and obtained good results reflecting high predicting ability. The distribution maps acquired consequently bring out significant information about the habitat preferences of the species, indicating areas with high suitability that overlap certain environmental conditions like elevation, slope and vegetation index values. These results also have significant implications for conservation work to determine where it is best to protect and restore wetlands. The study also considers the weakness of its model (MaxEnt) such as presence collection bias and needs to detailed environmental variable databases. Together, this information advances our knowledge of species-environment relationships critical for conservation planning and management in Binsar Wildlife Sanctuary.