Spatiotemporal Evolution of Vegetation Cover NDVI and Its Response to SPEI in Guizhou Province in 2001—2015

Climatic factors causing vegetation cover change are dominant.Variation characteristics of long time series of vegetation index in Guizhou province in recent 15 years，characteristics of drought-temporal variation based on standardized rainfall evapotranspiration index （SPEI） and correlation between them were explored，which could provide a theoretical basis for vegetation cover protection and restoration in Guizhou Area.Temperature and precipitation data of MODIS NDVI（1 km）and 19 meteorological stations in 2001—2015 were selected，and a linear regression method was used to explore temporal and spatial evolution characteristics of vegetation cover NDVI in Guizhou province in the past 15 years，and its temperature，precipitation and standardized precipitation evapotranspiration index（SPEI）.Results showed that，annual average NDVI from northwest to southeast showed a low-medium-high distribution in Guizhou province，showed a significant increase in growth rate of 0.051/（10a）at p<0.01.Spatial distribution of average temperature and precipitation from northwest to south showed characteristics of low-medium-high，and its prone to arid areas were mainly concentrated in the northwest of Guizhou.There were obvious differences in drought areas，and the most prone areas were concentrated in northwest and southwest of Guizhou.The correlation coefficients of average annual NDVI and average annual SPEI，average annual temperature and average annual rainfall were positive，indicated that NDVI had a certain correlation with its response factor，but correlation with SPEI was not significant，and correlation with temperature and precipitation was significantly positive at p<0.01，and correlation coefficients were all more than 0.5，degree of correlation was higher.There was no significant negative correlation from January to June（p<0.01），and there was no significant positive correlation from July to September and November except September（p<0.01）.