Assessing the Spatio-Temporal Variability in the Frequency and Magnitude of Flash Floods and their Driving Mechanisms: Evidence from Haor Region of Bangladesh

There are a limited number of studies addressing the spatiotemporal variability of pre-monsoon flash floods and their driving forces in Bangladesh. This study examines long-term trends in temperature, rainfall, and the frequency and magnitude of flash floods in the five most vulnerable haor districts of northeastern Bangladesh. Temperature, rainfall, and surface water level datasets, up to 2018, were collected from the Bangladesh Meteorological Department (BMD) and the Bangladesh Water Development Board (BWDB). Based on the normal distribution of these datasets observed in quantile-quantile (Q-Q) plots, regression models were used to analyze the long-term temperature trends. The models predict a gradual rise in maximum temperature, ranging from 1.06% to 1.94% over decadal periods. Additionally, an average annual rainfall increase of 4.1% at Sreemangal and 2.28% at Sylhet stations are forecasted. Analysis of historical data from the past sixty years shows a relatively lower peak river stage in tidal rivers compared to non-tidal river stations during pre-monsoon months. The frequency of peak surface water levels at six non-tidal and ten tidal river monitoring stations was estimated using Gumbel's probability estimation method. Frequency analysis suggests a high probability of flash floods across most floodplain areas, with a return period of five to ten years, based on flood danger levels established by government agencies. Furthermore, MODIS satellite imagery (with cloud cover <10%) from the peak flood months (March to May) between 2004 and 2017 was analyzed to assess the extent of flash floods in the study area. Geospatial analysis revealed temporal variations in peak flood extents across different locations. While no clear trends were observed in the frequency of flash floods, their magnitude has significantly increased in recent decades, potentially leading to greater losses in agriculture and property. The increased vulnerability to flash floods in the region can be attributed to several factors, including a rise in pre-monsoon heavy rainfall in the upstream hilly regions of Assam and Meghalaya, high sediment loads in Transboundary Rivers, drainage congestion, poorly designed and maintained flood control structures, and the absence of a reliable flash flood warning system. J. Asiat. Soc. Bangladesh, Sci. 50(1-2): 33-50, June-December 2024