Comprehensive stochastic assessment of meteorological drought indices

Drought indices are among the most important tools used to monitor and detect drought around the world. Most of these indices have been experientially developed using data extracted from specific geographic locations and climatic conditions. Regarding the inherent complexity of drought phenomenon itself as well as local climatic conditions, there have been great uncertainties when using such indices to provide accurate and logical assessments of drought in different regions. Few stochastic assessment studies, if any, have been published to evaluate and compare the inherent performance and capability of drought indices. Drought monitoring studies are usually based on a historical data record with relatively short length, e.g. 20-50 years, and have also focused on some particular aspects. In this study, seven meteorological drought indices and Monte-Carlo simulation method have been used to monitor drought characteristics in 12 diverse parts of the world endowed with various climatic conditions. Theoretical features of the indices as well as their inherent performance have been considered in the course of drought monitoring assessments. Accordingly, 50-year historical rainfall data of 12 stations in diverse parts of the world have been employed to generate synthetic rainfall time series of 1000 sequences in terms of monitoring drought characteristics and quantifying the inherent aspects of drought indices. The results confirmed the comparative advantage of the standardized precipitation index (SPI) as an accurate and realistic analysis. Similarly, Nitzche Index introduced itself as the best candidate for initial analyses. Meanwhile, other indices indicated significant deviations from the expected occurrence of normalized natural processes and their results could not be rationally and sufficiently applied to predict drought events.