NDVI-Based Long-Term Vegetation Dynamics and Its Response to Climatic Change in the Mongolian Plateau

The response of vegetation to regional climate change was quantified between 1982 and 2010 in the Mongolian plateau by integrating the Advanced Very High Resolution Radiometer (AVHRR) Global Inventory Modeling and Mapping Studies (GIMMS) normalized difference vegetation index (NDVI) (1982–2006) and the Moderate Resolution Imaging Spectroradiometer (MODIS) NDVI (2000–2010). Average NDVI values for the growing season (April–October) were extracted from the AVHRR and MODIS NDVI datasets after cross-calibrating and consistency checking the dataset, based on the overlapping period of 2000–2006. Correlations between NDVI and climatic variables (temperature and precipitation) were analyzed to understand the impact of climate change on vegetation dynamics in the plateau. The results indicate that the growing-season NDVI generally exhibited an upward trend with both temperature and precipitation before the mid- or late 1990s. However, a downward trend in the NDVI with significantly decreased precipitation has been observed since the mid- or late 1990s. This is an apparent reversal in the NDVI trend from 1982 to 2010. Pixel-based analysis further indicated that the timing of the NDVI trend reversal varied across different regions and for different vegetation types. We found that approximately 66% of the plateau showed an increasing trend before the reversal year, whereas 60% showed a decreasing trend afterwards. The vegetation decline in the last decade is mostly attributable to the recent tendency towards a hotter and drier climate and the associated widespread drought stress. Monitoring precipitation stress and associated vegetation dynamics will be important for raising the alarm and performing risk assessments for drought disasters and other related natural disasters like sandstorms.