Daily monitoring of drought effects on vegetation cover using ERA-Interim precipitation data and MODIS Images (Case study: Kermanshah Province)

The purpose of this study is to evaluate the effects of meteorological droughts on vegetation cover using satellite images. For this purpose, vegetation images were extracted from MODIS sensor of AQUA satellite on a 16-days timescale, during the growing season for three years of Wet (2006), Normal (2009) and Dry (2008) for different land uses, i.e. forest, pasture and agriculture in Kermanshah province, Iran. After necessary corrections to the satellite images, the daily Normalized Difference Vegetation Index (NDVI) values were calculated. Then, based on ERA Interim gridded precipitation data, the summed value of precipitation for the dummy durations of 1-365 days before any given date were calculated without and with a time-dependent reduction function representing the effect of precipitation on vegetation. The results showed that there is a strong and significant correlation between Interim rainfall variability and NDVI in three different land uses. The summation durations having maximum correlation with vegetation were used to calculate Effective Drought Index (EDI). It was identified that the maximum correlations between NDVI and EDI had negative values in all three selected land uses. The reason for this may be related to the technique that EDI employs to monitor drought. While vegetation cover is affected by precipitation conditions of the same year, EDI is calculated on the basis of daily precipitation with respect to its long-term average. Accordingly, another index was derived based on effective precipitation (EP) data, which is called Standard Effective Precipitation Index (SEPI). Results of comparison between NDVI and SEPI series showed that both series were similarly changed.