Modelling the biological invasion of Prosopis juliflora using geostatistical-based bioclimatic variables under climate change in arid zones of southwestern Iran

Invasive species have been the focus of ecologists due to their undesired impacts on environment. The extent and rapid increase in invasive plant is recognized as a natural cause global-biodiversity loss degrading ecosystem services. Biological invasions can affect ecosystems across wide spectrum bioclimatic conditions. Understanding impact climate change invasion crucial for sustainable biodiversity conservation. In this study, possibility mapping distribution Prosopis juliflora (Swartz) DC. was shown using present background data Khuzestan Province, Iran. After removing spatial bias by creating weighted sampling grids occurrence dataset, we applied six modelling algorithms (generalized additive model (GAM), classification tree analysis (CTA), random forest (RF), multivariate adaptive regression splines (MARS), maximum entropy (MaxEnt) ensemble model) predict under current future conditions both optimistic (RCP 2.6) pessimistic 8.5) scenarios years 2050 2070, respectively. Predictor variables including mean CHELSA (climatologies at high resolution Earth’s land surface areas)-bioclimatic geostatistical-based (1979–2020), physiographic extracted from shuttle radar topography mission (SRTM) some human factors were used process. To avoid causing biased selection predictors or coefficients, resolved autocorrelation presence points multi-collinearity predictors. As conventional receiver operating characteristic (ROC), area curve (AUC) calculated absence observations measure probability two error components are equally. All models evaluated partial ROC different thresholds other statistical indices derived confusion matrix. Sensitivity showed that diurnal range (Bio2) annual precipitation (Bio12) explained more than 50% changes played pivotal role habitat suitability P. juliflora. At all thresholds, significant difference comparison with single model. However, MaxEnt RF outperformed others models. Under scenarios, it predicted suitable areas will increasing climatically facilitate its distribution. These findings support conservation planning management efforts ecological engineering be formulating preventive measures.