Sensitivity of vegetation toward precipitation in dry land of China using satellite images

There is a scientific need to present an objective, spatially explicit and quantitative measure for sensitivity of vegetation toward precipitation in dry land. It will be helpful to understand the spatial and temporal interaction between them. In the study, we used 1km-monthly MODIS (Moderate Resolution Imaging Spectro-radiometer) NDVI time-series (2001-2013 year) as proxy to indicate temporal context of plant variation. Accordingly, a 5km-monthly precipitation dataset was also produced by EWBMS (Energy and Water Balance Monitoring system) model, which derived from more than 800 ground rain gauges and hourly cloud imagery from GMS (Geostationary Meteorological Satellite). the yearly green NDVI/precipitation was computed by computing monthly data during growing season. Sensitivity of vegetation toward precipitation was assessed in arid and semi-arid regions of China: (1) the vegetation variability was assessed for each pixel by the statistical value of multi-year green NDVI. (2) The relationship of vegetation and precipitation variation was tested by a linear regression model based on multi-year green NDVI/precipitation data. (3) GlobCoverV2.2 (GlobCover 200412 -200606 V2.2) map was collected and converted into 1km-pixel size. By overlaying each class layer, plant community in dry land of China varied with precipitation was discussed. The results indicate that plants located in agricultural and pastoral zone or arid oases have greater variability than other regions in yearly vegetation production. Land covers in these regions have GNDVI variation thresholds as 0.14-0.17, referring to the difference between maximum and minimum value of yearly green NDVI. The correlation coefficient R indicates that GNDVI is positive to precipitation in most of study area except for water, ice, forest, and some desert regions. The high value of R mostly occurs in southeast of study area, this is coincided with the high yearly variation of vegetation production.