Exploring Regional Profile of Drought History- a New Procedure to Characterize and Evaluate Multi-Scaler Drought Indices Under Spatial Poisson Log-Normal Model

Drought is recurrently occurring in many parts of the globe. In contrast to other natural hazards, drought has complex climatic characteristics. Several environmental factors are involved in the occurrence of drought hazards. However, the selection of an appropriate drought index may incorporate to make efficient drought mitigation policies. Moreover, the spatial distributions of extreme weather conditions (Dry/Wet) are necessary to avoid the consequences of future drought hazards. In this study, we aimed to explore and compare the regional profile of Dry/Wet episodes under various drought measures. Consequently, this research proposes a new spatial comparative procedure to assess and evaluate the spatial predictive distributions of Dry/Wet episodes under various drought measures. The study incorporates three Standardized Drought Indices (SDIs) and a spatial Poisson log-normal model to assess and evaluate the spatial predictive distributions of Dry/Wet episodes in Pakistan. Results of this study show that the segregated patterns of Dry/Wet counts are moderately consistent with the climatology of the region. However, the spatial patterns of Wet/Dry counts under various drought measures are significantly different under each index. Therefore, this research suggests the simultaneous use of multiple drought measures for accurate and precise drought monitoring at the regional level.