In this study, the spatial-temporal characteristics of vegetation status, drought conditions, and their correlation were studied through three phases of the normalized difference vegetation index (NDVI), land surface temperature(LST)-moderate resolution imaging spectroradiometer (MODIS) database during the growing season in Xinjiang from 2000 to 2015. In addition, a multivariate linear regression method was used in this study to analyze the meteorological factors that could influence the relationship between NDVI and LST of different phases. The spatial-temporal characteristics of the relationship between NDVI and LST were analyzed by dividing different land use cover changes (LUCC). The results showed that (1) there was a significant correlation between LST and NDVI in the third quarter of the growth season, and drought conditions could be evaluated using the relationship between NDVI and LST. (2) The influence of meteorological factors on the NDVI-LST correlation was different in various phases in Xinjiang. Moreover, there were significant differences in NDVI-LST correlations among different LUCC. (3) It was effective to monitor vegetation health and drought conditions in most areas of Xinjiang using a vegetation health index in the middle of the growing season, whereas earlier and latter drought assessments needed to be supplemented by other drought indices.