Skip to main content

NDVI-based monitoring of long-term vegetation dynamics and responses to multi-time scales droughts in Inner Mongolia

Cite this dataset

Tong, Siqin; Bao, Gang; Bao, Yuhai; Huang, Xiaojun (2022). NDVI-based monitoring of long-term vegetation dynamics and responses to multi-time scales droughts in Inner Mongolia [Dataset]. Dryad.


The characteristics of vegetation and drought for different seasons between 1982 and 2015 in Inner Mongolia were studied based on the Normalized Difference Vegetation Index (NDVI) and the Standardized Precipitation Evapotranspiration Index (SPEI). The response of vegetation to drought over various time scales for different seasons and vegetation types was investigated using the maximum Pearson correlation, allowing a discussion about the possible causes of any changes. The results indicate that the vegetation NDVI in Inner Mongolia showed an increasing trend in different seasons, with spring vegetation NDVI (April to May) having the largest significant increasing rate, followed by the growing season (April to October), autumn (September to October), and summer (June to August). Accordingly, the proportion of stations with decreasing SPEI was, in descending order, summer, growing season, autumn, and spring. Additionally, the magnitude of the SPEI decrease was greater in eastern Inner Mongolia. NDVI and SPEI were positively correlated in most regions of Inner Mongolia, indicating that changes in vegetation in most parts of this region were affected by the spatial and temporal characteristics of drought, the correlation being them being strongest in the growing season, followed by summer, then spring and autumn. Considering the different types of vegetation, forests were less affected by drought, with broadleaf forests more affected than coniferous forests. The meadow steppes and typical steppes were more affected by 12-month droughts in the growing season and summer, 6-month droughts in spring, and 3-month droughts in autumn, with desert steppes mainly affected by 3-month droughts. The shrubs, sandy vegetation, and cropland were mostly affected by droughts in summer, and show a greater response to 3-month droughts in autumn. Finally, the water balance was found to be the most important factor affecting the response of vegetation to drought in Inner Mongolia.


We upload two datasets used in the paper, including GIMMS NDVI3g and SPEI dataset during 1982-2015 in Inner Mongolia region.

(1) To monitor the vegetation, the GIMMS (Global Inventory Modelling and Mapping Studies) NDVI3g datasets were used because of their high quality, with 0.083° and 15-day spatial and temporal resolution, respectively (download from To reduce the effects of atmospheric and aerosol scattering, we used the maximum value composite (MVC) method to develop a monthly NDVI dataset.  Since the vegetation in Inner Mongolia shows minimal growth or is covered by snow in winter, the growing season (April to October) was selected for this study, with the other considered seasons defined as spring (April to May), summer (June to August), and autumn (September and October). 

(2)  The SPEI (Standardized Precipitation Evapotranspiration Index), was selected to indicate drought in the paper. It is designed to consider both precipitation and potential evapotranspiration when identifying droughts. It also captures the main impact of increased temperature on water demand and can be calculated over a range of timescales, reflecting the duration and cumulative effects of droughts. The SPEI is therefore a widely used index for evaluating, monitoring, and assessing drought within the background of global warming. The calculation of SPEI please check in the paper.

Usage notes



National Natural Science Foundation of China, Award: 42061070

National Natural Science Foundation of China, Award: 41861056

Research Foundation for Advanced Talents of Inner Mongolia Normal University, Award: 2019YJRC002