Abstract: Vegetation plays an important role in regulating the terrestrial carbon balance and the climate system, and also overwhelmingly dominants the provisioning of ecosystem services. However, changes occurred on small temporal scales and the persistency or robustness of the changes was often not fully understood. Documentation of changes in vegetation over the most recent years is limited. These documented changes are especially important for policy development and ecosystem conservation and recovery. In this study, with GIMMS NDVI (1982-2006) and MODIS NDVI (2000-2012) datasets in Xinjiang, the spatio-temporal patterns of changes in monthly NDVI and their linkage with change in temperature, precipitation, evapotranspiration, humidity index and human activity were analyzed from March to November at regional and pixel scales. To detect the trend of NDVI during a given period, a least -squares linear regression was applied. To further explore the climatic factors driving NDVI change, interannual correlations between NDVI and climatic variables were calculated using Pearson correlation analysis. To analyze the temporal patterns and dynamic process, we estimated trends of NDVI and the correlation between NDVI and climatic factors over progressively longer periods of 25 to 31 years since 1982 and calculated the percentage of the area that showed a positive or negative trend in the seven nested time series. The monthly NDVI was significantly increased in all months besides November, but there were two distinct periods with opposite trends in most of these months, especially in June to September, from which it increased before 1998 or 1997 or decreased after 1998 or 1997. The trend of reversal of NDVI change also led to that the rate of NDVI increase was notably slowed or stopped as the NDVI record grew in an incremental length from 1982-2006, 1982-2007, ..., to 1982-2012 for March-October. At pixel scale, the areas with significantly change of NDVI in size highly significantly (P<0.01) increased during seven periods in most months, especially for those with a significant (P<0.05) decrease trend. The rate of increase in size of areas with significant decreasing NDVI was larger than that with significant increasing NDVI in all nine months. Consequently, this caused stall or slow increase of regional scale NDVI. The response of vegetation NDVI to climate change for different months was different. The response of NDVI in March-June and September-November was more sensitive to thermal indicator, such as temperature and evapotranspiration. The correlation between NDVI and precipitation and humidity index was stronger in July-August, and areas with significant (P<0.05) correlation were larger. Moreover, the effects of spring temperature on vegetation growth were more substantial at high elevations, such as Altai Mountains, Tianshan Mountains, than that at low elevation. In addition, the impact of climate on vegetation became more significant over a longer time scale. Meanwhile, change in NDVI was significantly (P<0.05) affected by human activities. The change of planting structure and farming methods were also the driving factors. The promotion of agricultural production such as the rapid increase in the proportion of cotton cultivation and the change from flood to drip irrigation may reduce March-May NDVI in some farmland areas. The difference in study time spanning generates different results. Trend analysis during the multiple nested time series may contribute to a better and thorough understanding of NDVI dynamic. Extending the time series as much as possible and focusing on the course of change are particularly important in studies that monitor vegetation dynamics and its relationship with climate change.