Study on inter-seasonal and intra-seasonal relationships of meteorological and agricultural drought indices in the Rajasthan State of India

Use of rainfall anomaly based Standardized Precipitation Index (SPI) and satellite-derived Vegetation Condition Index (VCI) are becoming common to assess the impacts of drought on crops. This study analysed spatio-temporal intra-seasonal and inter-seasonal relationships for 24 years between rainfall and NDVI and between SPI and VCI to understand crop response to water availability in the Rajasthan State, India. To separate the effect of weather and technology on crop growth over time, a modification in VCI was proposed and called Trend Adjusted VCI (VCITadj). The VCITadj was computed for early, mid, late and whole crop seasons by deriving pixel wise crop phenology metrics from NDVI profile. Significant linear relationships were found between NDVI and rainfall but phase of crop season affected the strength of this relationship. The SPI and VCITadj were linearly related in all the four seasons, the strength of relationship improved with the progress of crop season and these relationships were stronger than between rainfall and NDVI. These relationships broke down in irrigated croplands. As a result, the anomaly indices of SPI and VCITadj and their intra-seasonal relationships can be used to study the response of crops to water availability for early detection and better prognosis of agricultural drought. (C) 2013 Elsevier Ltd. All rights reserved.