Abstract: Rana zhenhaiensis is a newly named species endemic to China; its potential geographic distribution is still unclear due to lack of field investigation. Thirtyfive distribution localities ofR. zhenhaiensiswere collected from field investigations and literatures. The MaxEnt ecological niche model was used to predict the potential distribution of R. zhenhaiensis by importing corresponding data of 20 biological and environmental variables downloaded from WorldClim database between 1950 and 2000 corresponding to the above 35 localities. The predicted habitats with suitability probabilities of 0.5-1.0 forR. zhenhaiensis mainly included Zhejiang, Jiangsu, Shanghai, Anhui, Hunan, eastern Hubei, central, western and northern parts of Jiangxi, coastal regions of Fujian, western Tianjin, and several other parts in Sichuan, Chongqing, Shandong, Hebei, Guangxi and Guangdong. Besides, Beijing, Shandong, Henan, Chongqing, Guizhou, Guangxi, Fujian, Guangdong, southern Liaoning, eastern Tianjin, southern Hebei, southern Shanxi, eastern Sichuan, western Hubei, southern Hunan, northern and southwestern parts of Jiangxi as well as northern Taiwan were relatively lower suitable habitats (suitability probabilities, 0.09-0.5) forR. zhenhaiensis. The average AUC values for both training and test data predicted by MaxEnt were 0.993, indicating high accuracy and reliability of our predicted model. Jackknife test in MaxEnt ecological niche model showed that the contributions of different environmental variables to models were different, and precipitation in the warmest quarter (contribution rate, 44.6%) and standard deviation of temperature seasonal change (contribution rate, 20.4%) were the primary factors that affect the distribution ofR. zhenhaiensis.

Key words: low-carbon behavior, driving mechanism, scenic spot, World Heritage Site., redundancy analysis (RDA)