Spatiotemporal distribution of drought and its possible associations with ENSO indices in Bangladesh

Abstract

Droughts and related water stress are the major constraints of the sustainable socioeconomic development of Bangladesh. Large-scale atmospheric oscillations are the major drivers of climate fluctuation and droughts in Bangladesh, like many other regions of Asia including the Indian subcontinent, the largest entity in the world, with over 1.7 billion people. Therefore, it is crucial to insight into the spatiotemporal distribution of drought and its linkage to large-scale atmospheric indices to provide early warning and alleviate drought impacts. However, regional droughts and their linkages to large-scale oscillation indices like El-Nino Southern Oscillation (ENSO) are not explored adequately in Bangladesh. This study intends to evaluate the spatiotemporal distribution of droughts in Bangladesh using the standardized precipitation evapotranspiration index (SPEI) and the standardized precipitation index (SPI) for multiple timescales,?-?3,?-?6,?-?12, and 24-months, and to investigate the relationship of drought characteristics with ENSO. The monthly rainfall and temperature records from twenty locations for 38 years from the period 1980 to 2017 were used for this purpose. The results revealed that the droughts are region-specific and are in agreement with the warming trends observed in the different regions of Bangladesh. The droughts, particularly short-term droughts, are increasing significantly in the North-western region, indicating the worsening drought conditions in the drought-prone region. The SPI and SPEI showed a significant (p?<?0.05) positive association with the percentage of precipitation anomaly (Pa). However, the association of drought indices with ENSO and potential evapotranspiration (PET) were not significant. The polynomial regression model demonstrated that Indian Ocean Dipole (IOD) could explain SPEI-3 (4.7%) variations better than SPI-3 (4.2%).