A new remote sensing framework for drought monitoring

Drought is one of the most expensive but least understood natural disasters. The remote sensing based integrated drought indices could reflect drought conditions more comprehensively than single drought indices. However, one of the major issues for most remote sensing based integrated indices is that those indices were established and evaluated under specific environmental conditions. The second major issue is that for the typical composition method to generate integrated drought indices, the indices were developed based on the linear combination of single drought index to integrate drought information from multiple sources with weighting determined empirically. A third limitation is that traditional integrated drought indices only use single in-situ based drought index as the dependent variable to combine the multi-source remote sensing data. In this regards, the objective of this study is to develop and evaluate a new integrated drought index based on multi-sensor remote sensing data for meteorological drought monitoring under different climate conditions. We named the new index Geographically Independent Integrated Drought Index (GIIDI). For locations with available meteorological data, we also developed a GIIDI_Station version for more accurate drought monitoring using a different statistical approach. Our results indicated that GIIDI had the strongest correlation with in-situ drought indices than the compared remote sensing based meteorological drought indices in most climate divisions. The performance of GIIDI was not significantly affected by environmental factors. When local station data are available, GIIDI_Station performs even better. Our framework has considerable potential for improving our ability to monitor drought across a range of biomes and climates and to be applied globally.