Mapping the agricultural drought based on the long-term AVHRR NDVI and North American Regional Reanalysis (NARR) in the United States, 1981–2013

To provide a long-term perspective of drought variability from 1981 to present, we develop a new monthly agriculturally-based drought index called the Integrated Scaled Drought Index (ISDI). This index integrates Normalized Difference Vegetation Index (NDVI) from Advanced Very High Resolution Radiometer (AVHRR) data (available from 1981 to present), land surface temperature (LST), precipitation (PCP), and soil moisture (SM) data from North American Regional Reanalysis (NARR) project (available from 1979 to present). This new agriculturally-based drought index incorporates important components controlling agricultural drought, particularly soil moisture, for which there are limited in-situ observations through time and across space. The optimum weights for each component of the ISDI are determined by correlation analysis with commonly used in-situ drought indices, such as the Palmer Drought Severity Index (PDSI), the Palmer Modified Drought Index (PMDI), the Palmer's Z-index, and the Standardized Precipitation Index (SPI) at different time scales. Resulting ISDI maps are also visually compared with United States Drought Monitor (USDM) and Vegetation Drought Response Index (VegDRI) maps for empirical validation. ISDI shows strong agreement with these two national-wide drought monitoring systems. ISDI also shows strong linear correlations with corn yield anomalies in July and with soybean yield anomalies in August and strong spatial correspondence with county-level corn/soybean yield anomalies during major drought events. These results illustrate the robustness and usefulness of ISDI. This agriculturally-based drought index integrates the benefits of numerical model simulation and remote sensing technology to account for interannual variability of drought for the longest possible time-frame in the satellite era. This long-term monthly drought index provides a longer historical perspective of drought impacts since 1981. It can be generalized to incorporate other satellite data or in-situ observation and has the potential for operational drought monitoring and assessment.