Spatially Explicit DroughtStorage Coupling and Storage Response Regimes in Bangladesh Revealed by TWSASPEI Analysis From GRACE and GRACE-FO

Monthly Terrestrial Water Storage Anomalies (TWSA) derived from GRACE and GRACE-FO CSR RL06 mascon solutions are integrated with climatic drought indices (SPEI) computed from CHIRPS precipitation and GLDAS-derived potential evapotranspiration. The analysis spans April 2002 to September 2025 and is conducted on a 0.25° (~25 km) grid, masked to the national land area of Bangladesh. Pre- and post-GRACE-FO periods are analyzed separately to assess changes in hydro-climatic coupling and groundwater dynamics. Key analytical components include: Spatial correlation analysis between TWSA and multi-scale SPEI with lag optimization and statistical significance testing. Trend detection using Mann–Kendall tests and Sen’s slope estimators for both groundwater storage and drought indices. Pixel-wise drought characterization, including drought frequency, duration, intensity, and affected area fraction. Unsupervised machine-learning classification (k-means and hierarchical clustering) to identify dominant groundwater depletion and drought-response regimes. Regime transition analysis to quantify structural shifts in drought–storage behavior between study periods. All processing is implemented in Python using reproducible workflows, with outputs provided as NetCDF, GeoTIFF, CSV, and publication-ready figures. The dataset supports national-scale assessment of groundwater stress, drought vulnerability, and regime shifts relevant to water-resource management and climate-adaptation planning in Bangladesh. Data Included GRACE/GRACE-FO CSR RL06 mascon TWSA (monthly, 2002–2025) CHIRPS monthly precipitation GLDAS-NOAH monthly potential evapotranspiration SPEI (scales 1–12 months) Derived products: correlation maps, trend statistics, drought metrics, regime classifications, transition matrices Publication-ready figures and summary tables Spatial Coverage Bangladesh (national land mask) Temporal Coverage April 2002 – September 2025 Data Formats NetCDF, GeoTIFF, CSV, PNG/PDF (figures) Methods / Keywords (Zotero “Tags”) GRACE, GRACE-FO, groundwater depletion, terrestrial water storage, SPEI, drought analysis, Bangladesh, machine learning, clustering, hydro-climatic coupling, Mann–Kendall, Sen’s slope, regime classification, climate variability