Predicting the potential geographic distribution of Fagus orientalis Lipsky under climate change using an ensemble...

Oriental beech (Fagus orientalis Lipsky) is an ecologically and economically significant species, covering 8.5% of Türkiye's total forest area. However, climate change threatens its distribution due to increasing temperatures and decreasing precipitation. This study integrates geospatial informatics and ensemble modeling (EM) to predict the potential geographic distribution (PGD) of F. orientalis under future climate scenarios using Biomod2 within the ShinyBIOMOD framework. An EM model has been developed from six models [Generalized Boosting Model (GBM), Generalized Linear Model (GLM), Multivariate Adaptive Regression Splines (MARS), Generalized Additive Model (GAM), Artificial Neural Networks (ANN), and Maximum Entropy (MaxEnt)] using 76 validated occurrence records and 19 environmental predictors. Model validation achieved high predictive accuracy (AUC = 0.96, TSS = 0.85). Spatial projections for SSP2-45 and SSP5-85 scenarios indicate significant shifts in PGD. Notably, high-suitability habitats will decline under SSP2-45 but expand under SSP5-85. Bio2 [Mean Diurnal Range (mean of monthly (maximum temperature - minimum temperature))] and Bio4 [Seasonal temperature fluctuation (temperature seasonality (standard deviation x 100))] emerged as the dominant drivers of distribution changes. Based on geospatial analyses, F. orientalis is expected to migrate to higher altitudes in the Black Sea region and expand into southern and inner Türkiye. This shift reflects a broader trend of temperate forest adaptation to climate change. This study underscores the power of ensemble modeling for ecological forecasting and conservation planning, demonstrating the value of computational tools in assessing climate-driven species distribution changes. The findings contribute to predictive modeling for biodiversity conservation and ecosystem management.