Climate change-induced invasion risk of ecosystem disturbing alien plant species: An evaluation using species distribution modeling

Species distribution modeling is widely used for evaluating invasion risk, and prioritizing areas the control management of invasive species. However, selecting a tool that accurately predicts species risk requires systematic approach. In this study, five models (SDMs), namely, artificial neural network (ANN), generalized linear model (GLM), multivariate adaptive regression splines (MARS), maximum entropy (MaxEnt), random forest (RF), were performed evaluated their performance using mean value area under curve (AUC), true skill statistics (TSS), Kappa scores 12 ecosystem disturbing alien plant (EDAPS). The evaluation metric highest in RF (AUC = 0.924 ± 0.058, TSS 0.789 0.109, 0.671 0.096, n 12) lowest ANN. ANOVA AUC, TSS, metrics revealed was significantly different from other SDMs therefore selected as relatively best model. potential each EDAPS quantified. Under current climate conditions South Korea, average estimated to be 13,062 km 2 . future change scenarios, percentage relative predicted increased over 219.93%. Furthermore, climate, 0.16% country very high invasion, but would increase 60.43% by 2070. Invasion northwestern, southern, southeastern regions, densely populated cities, such Seoul, Busan, Daegu. By 2070, expand across whole except northeastern region. These results suggested induced invasiveness, could valuable tools assessment.