FUTURE ASSESSMENT OF CLIMATE CHANGE ON DROUGHT IN RAJASTHAN

Abstract

Droughts, exacerbated by climate change, threaten agriculture, water resources, and ecosystems. This dissertation presents an XGBoost model designed to predict drought indices (SPEI, SPI, RDI) using historical data from the India Meteorological Department (IMD) and US National Oceanic and Atmospheric Administration (US NOAA satellite) for Rajasthan, India (1901-2023). Future projections utilize CMIP6 data (2024-2100) under various SSP scenarios. The model is trained and validated on IMD and US NOAA datasets, showcasing superior performance in mean squared error, mean absolute error, and root mean squared error compared to other machine learning and deep learning models. This framework enhances drought prediction accuracy, critical for effective mitigation strategies and preparedness.