Facing the climate change and restoring vegetation is an urgent problem in the karst area. Based on the long-sequence vegetation index and meteorological data from 1987 to 2017 in the karst area in Shilin County, Yunnan Province, the relationship between different vegetation types and climatic and environmental factors was studied by means of Sen's slope analysis, Morlet wavelet analysis, partial correlation analysis and Mann-Kendall non-parametric detection. The results showed that: The four vegetation types showed a decreasing trend during the study period, and there were time periods of about 21, 11 and 6 years, and the mutation years of evergreen broad-leaved forest, coniferous and broad-leaved mixed forest and coniferous forest occurred in 2009, and the rocky desertification shrubs occurred in 2004. There was a positive correlation between vegetation NDVI and precipitation, and a negative correlation with temperature. The cyclical changes are controlled by the rhythm of precipitation, and the mutations of the three vegetation types of broad-leaved forest, coniferous and broad-leaved mixed forest and coniferous forest are caused by the mutation of precipitation, while the mutation of rocky desertification shrub is caused by the abrupt change of temperature. Rocky desertification shrubs have the worst stability compared with other vegetation types. Evergreen broad-leaved forests, coniferous and broad-leaved mixed forests and coniferous forests are more stable than rocky desertification shrubs in the context of climate change. This study can provide scientific basis for vegetation restoration and protection in karst areas of southern China.