Delineation of flood risk zones and affected land use land cover for Sunamganj and Sylhet districts using remote sensing approaches

Bangladesh experiences annual flooding due to its geographical location in the delta of significant rivers, combined with climatic factors like heavy monsoon rains and the effects of climate change, which lead to increased flood risks. In the Northeastern part of Bangladesh, Sunamganj and Sylhet regions are particularly vulnerable, experiencing recurrent monsoon floods that inflict substantial economic losses, impacting agriculture, infrastructure, and livelihoods, incurring millions of dollars in damages. This study aimed to analyze the flooding situation for the Flash Flood of 2022 by creating classified flood extent and susceptibility maps to asses affected land use land cover. For result validation, a comparison was made between the results obtained and the flood map available at the secondary data source of the Flood Forecasting and Early Warning (FFWC) website, Bangladesh. In this study, Sentinel-1A SAR images were used to analyze flood extent using SNAP, and Sentinel 2A images were assessed using Google Earth Engine and ERDAS Imagine for land use land cover information. Flood susceptibility analysis was done with ArcGIS software and based on factors such as topographic wetness index, elevation, slope, rainfall, soil type, land use/land cover, drainage density, and distance from the river. Three risk zones were identified for this study area: low-risk, moderate-risk, and high-risk. The land use land cover classification was validated with primary ground-truthing data. The classes of agricultural land, vegetative cover, built environment, barren land, and dry haor land were identified, and their flooded and non-flooded areas in hectares were calculated. The flood extent maps were validated with the FFWC flood map which showed 81% accuracy. For the Sylhet district, 56% of the total district's area of 2,76,755 hectares was inundated, and for the Sunamganj district, 66% of the total district's area of 1,57,719 hectares. This research produced a quick, efficient, and cost-effective flood analysis technique for analysing flood extent, susceptibility, and impact on land use. It can aid disaster management stakeholders in formulating proactive rescue, recovery, and future mitigation strategy development initiatives.