Abstract: Yellow crazy ant, Anoplolepis gracilipes, a newly recorded invasive species in southern China, poses serious threats to native biodiversity. In order to reveal the potential risk of its expansion and adaptive distribution, the occurrence points were divided into two parts: the native points and the global points, with local and global prediction models being constructed, respectively. Seven climatic environmental factors, which strongly influence the survival of A. gracilipes, were selected for model analysis. The maximum entropy (MaxEnt) model was adjusted by using the ENMeval data package in R software. The prediction of the niche area of A. gracilipes in China was constructed using the local and global prediction models with the default and refined parameter settings. The pROC protocol was used to test the reliability of the models. The results showed that under the same setting, there was an obvious difference in the potential geographic distribution of A. gracilipes, based on the global model and the native model, while the influence of the refined parameter settings on the model’s prediction was minimal. Overall, the potential distribution of A. gracilipes with high suitability was Yunnan, Guangxi, Guangdong, Fujian, Hainan and Taiwan, whereas Hunan, Guizhou, Jiangxi and parts of Sichuan were areas of intermediate suitability. Moreover, for the native model, central Africa and northcentral America were the potential distributions of the highly adaptive A. gracilipes. The definition of the local scope of A. gracilipes has a considerable impact on the accuracy of the prediction results of the model.

Key words: extreme, winter planting, chilling, return period