DOI: https://doi.org/10.36347/sajb.2026.v14i10.005

Pages: 681-693

Ximenia americana L. is a multifunctional indigenous woody species of West Africa, exploited for its food, medicinal and energy uses, but whose distribution could be affected by climate change. This study aimed to model the current and future potential distribution of the species in Benin and to identify the main environmental factors that determine it. The modeling mobilized 902 occurrences (field data and GBIF) and four algorithms (Random Forest [RF], MaxEnt, generalized linear model [GLM] and generalized additive model [GAM]) under the biomod2 package, integrating bioclimatic, topographic and edaphic variables. Future projections were made from the MIROC6 climate model under scenarios SSP2-4.5 and SSP5-8.5, at 2050 and 2070 horizons. The four models showed satisfactory predictive capacity (AUC: 0.888-0.949; TSS: 0.660-0.788), with the RF showing the best performance. Seasonal rainfall, soil type and altitude emerged as the main determinants of the species' distribution. Under current conditions, the favorable habitat is estimated between 70599 and 89068 km2 according to the algorithm. Projections indicate a gradual contraction of this habitat, particularly marked under SSP5-8.5 by 2070, where GLM and MaxEnt predict a near disappearance of the favorable habitat, unlike GAM and RF which maintain a substantial area. These results, despite significant inter-algorithm uncertainty, indicate a potential vulnerability of X. americana to to climate change and call for prioritizing its conservation in areas of climate stability and its integration into Beninese agroforestry systems.