Tree species diversity analysis using species distribution models: A Faidherbia albida parkland case study in Senegal

Around the world, SDMs have been widely used to support forest management planning and biodiversity conservation. Beyond prediction of species distribution provided by SDMs, this study aimed analyze spatial tree diversity using SDMs. The area is a Faidherbia albida parkland in Central Senegal. It characterized tree-based farming system dominated albida. Using robust representative dataset 9258 occurrence, we first determined an SDM current potential 16 main forming parkland. Specifically, 6 algorithms applying several modeling techniques with different categories predictor variables (e.g., climate, topography, soil properties human impact) benchmarked 576 achieve best model predictions for species. Then, maps were created on basis resulting predictions. Finally, dynamics discussed relation landscape characteristics, including heterogeneity, composition impact. results showed that there no single ‘best’ algorithm (among tested) or approach all Benchmarking allowed strengthening performance, achieving AUC values ranged from 0.64 (intermediate accuracy) 0.87 (very good accuracy). related heterogeneity composition. In Sahelian agroforestry systems (AFS), sustained anthropization. A significant negative correlation distance village was found, i.e. closer you get village, greater trees. This could be crucial analyzing when abundance information not available.