Abstract:In this paper, super efficiency SBM-DEA model and exploratory spatial data analysis method (ESDA) are used to study the spatial-temporal evolution characteristics and spatial correlation of green innovation efficiency in 108 cities in the Yangtze River Economic Belt from 2004 to 2018.The results show that the green innovation efficiency of the Yangtze River Economic Belt shows an inverted V-shaped development trend in the early stage and a slow rise in fluctuation after the financial crisis. The pure technical efficiency is the key factor affecting the efficiency of green innovation.The green innovation efficiency of the Yangtze River economic belt, the middle and the lower reaches of the Yangtze River are different. The efficiency level of the lower reaches is the highest, and the average value of the green innovation efficiency of the upper and middle reaches of the Yangtze River alternately leads in each year.The spatial pattern of green innovation efficiency in the Yangtze River Economic belt has gradually changed from "U-shaped distribution" in the upper and lower reaches, which is higher than that in the middle reaches, to "multi-center distribution".With the advance of time, the positive spatial correlation of green innovation efficiency between cities is increasingly significant, but the scale of low-efficiency city agglomeration is greater than that of high-efficiency city, and the phenomenon of local polarization is intensified.