Abstract: This study takes 31 provincial administrative units in China (excluding Hong Kong,Macau and Taiwan of China due to statistic data shortage) as the study area.Firstly,we use quantitative methods such as entropy method and standard deviation to quantitatively study the evolution of the regional tourism development potentials of the 31 units,and then use multiple linear regression analysis model and geodetector model to screen out the leading factors that have a great impact on the potential.The units are grouped into different types of tourism development potential areas with particular development models and recommendations for each type.The results show as follows: (1) The development potential of interprovincial tourism in China presents a trend with high value in the southeast,low value in the northwest and the potential was gradually decreased from the southeast to the northwest.The interprovincial regional differences show a pattern which was shrunk first,then enlarged and then shrunk again along with the time. (2) The tourism development potential in the eastern region has the largest difference,while the differences in the central,western,and northeast regions are generally small. (3) The leading factors in China’s regional tourism development potential include the total tourism revenue,total number of tourist accommodated,number of beds,number of employees in the tertiary industry,number of students in tourism colleges,and total import and export volume at home and abroad. (4) The four types of the regional tourism development potential are: market demandled type,regional supplyled type,guaranteed developmentled type,and economy supportled type.According to the characteristics of each type,the corresponding development countermeasures and recommendations are proposed.

Key words: tourism development potential, geographical detectors, evolutionary, leading factors, China