Climate change is leading to significant variations in the intensity and frequency of drought events. The compositions, structures, functions as well as the processes of terrestrial ecosystems are directly affected by the uncertainty of extreme climate events. In 2009-2010, the southwestern China, an area that contains multiple types of fragile ecosystems and was frequently hit by drought events, experienced a once-in-a-hundred-year extreme drought. Over the past decade, large efforts have been made to evaluate the recovery of vegetation after drought, however, the responses of vegetation growth to this extreme drought event on a long-term time scale is still unclear. In this study, using multi-year normalized difference vegetation index(NDVI) data, ecosystem type data, digital elevation model(DEM) and meteorological datasets, we found a climate factor to predict vegetation growth based on linear regression model. Then the duration and the spatial variation of legacy effects of 2009-2010 extreme drought in Yunnan Province in southwestern China were determined, and the responses of diverse vegetation types on the extreme drought event were analyzed. The results showed that 1) the inhibition of vegetation growth occurred about 1-2 years in Yunnan Province after the extreme drought event, especially in areas where precipitation experienced a severe reduction. The mean deviation between observed NDVI and predicted NDVI in whole research area in 2010 and 2011 were -0.0404 and -0.0330, respectively, and the deviations were more evident in grid points with a significantly positive relationship between climate factors and vegetation growth, 2) The most sensitive area of vegetation response to drought event was around 2000 m above sea level. The magnitude of negative drought legacy effects showed an increasing trend along the elevational gradients at altitudes lower than 2000 m and a decreasing trend at altitudes between 2000 m and 4000 m. At altitudes higher than 4000 m, the vegetation growth was almost unaffected. 3) Compared with grassland and farmland, the inhibition of forest vegetation was stronger and longer lasting, suggesting a weaker drought resilience of forest under dryer conditions. Our study revealed the negative impacts of extreme drought on the growth of vegetation in Yunnan Province and provided a theoretical basis for coping with extreme drought and restoring vegetation effectively in the future. More in-situ observation data are needed to improve the understanding of drought legacy effects and predict trajectories of terrestrial ecosystems in response to a warmer and drier climate in Yunnan Province and southwestern China.