Drought prediction in Bangladesh under changing climate: Integrating SDAT-SPI and SDSM-DCBC with CMIP6 data

Drought is a recurring environmental challenge that disrupts soil moisture and crop water balance due to prolonged dry spells. In Bangladesh and many other regions, it has emerged as an increasingly urgent natural hazard. This study employed rainfall data to develop statistically robust drought indicators using SDAT-SPI, a nonparametric tool. Additionally, CMIP6 datasets were analyzed to assess future drought trends under various climate scenarios, providing deeper insights into potential climatic changes. To further bridge the gap between global climate models and localized drought impacts, a novel drought downscaling and bias correction framework namely SDSM-DCBC approach for CMIP6 data was developed. The SDSM-DCBC approach was applied to improve the accuracy of drought assessments by comparing model simulations with observed data. Results indicate extended drought spells in certain years, with pronounced drought clusters projected for 2052–2053 and 2082–2083. Correlation analysis revealed a statistically significant relationships between global oceanic indices and regional drought. The Niño 3.4 index was identified as the primary driver of long-term drought in the Southern region, while the Pacific Decadal Oscillation (PDO) exerted the strongest unique influence in the North. These findings enhance drought prediction accuracy and provide an essential tool for developing effective, region-specific adaptation strategies. • Utilized SDAT-SPI to generate statistically robust drought indicators from rainfall data. • Analyzed CMIP6 datasets to project peak drought occurrences in 2052-2053 and 2082-2083. • Developed SDSM-DCBC to enhance drought assessment through downscaling and bias correction. • Identified significant correlations between drought indices, oceanic indices, and regional climate variables. • Improved drought prediction and management, contributing to a scalable global adaptation framework.