Abstract:
Climate change impacts global ecosystems, agriculture, health of individuals, and water availability. The most severe effect of climate change is drought, which is becoming severe and intense. Drought and associated risk management are typically tracked using a variety of drought indices. The key components in determining drought are temperature, precipitation, and other hydro-meteorological factors. Four agriculture drought indices – VCI, TCI, SPEI, and SMADI, were computed and analysed over 12 years (2010-2021) using Earth Engine. The possible future changes in the intensity and spatial distribution of agricultural droughts were analysed based on the MME mean of thirteen GCMs of CMIP6 in two different warming scenarios (SSP 2-4.5 and SSP 5-8.5) using the Random Forest algorithm to improve the prediction accuracy. A Drought Composite Map was generated by an innovative method using AHP established MCDM approach. Each index was given AHP-derived weightage with a consistency ratio of 3.10% and a consistency index of 2.79. The study also integrates multiple satellite products, including CHIRPS and IMERG rainfall, MODIS, and ERA-5 temperature datasets, used for the quantification for the period 2010-2021 across different agro-climatic zones. The results showed significant spatial and temporal variability in drought severity, with arid and semi-arid regions being the most vulnerable. Quetta, Kalat, and Panjgur are the most affected areas under current conditions; under future scenario (SSP 2-4.5), Badin, along with Kalat is severely affected by drought, and under SSP 5-8.5, and all these regions are under extreme drought conditions in 2100. Effective drought management requires simple, applicable models that consider multiple factors and can accurately forecast agricultural droughts despite meteorological outliers.
