Characterization and assessment of flood inundated areas of lower Brahmaputra River Basin using multitemporal Synthetic Aperture Radar data: A case study from NE India

Abstract

Floods are known to be one of the greatest disasters in documented human history. In the Indian subcontinent, the Ganga‐Brahmaputra plains annually accommodate monsoon‐driven flooding of the Brahmaputra River and its tributaries, causing widespread flood inundation and geomorphic changes. In the present study, we present flood inundation maps of Assam and suggest a possible geomorphic explanation for the frequent flood inundation areas. We have used a novel SAR (Synthetic Aperture Radar)‐based approach for a comprehensive flood inundation mapping of Assam. The Google Earth Engine (GEE), a cloud‐based data computing and analysis platform, is used to access and analyse the Sentinel SAR 1 GRD datasets for 30 events from 2018 to 2020. The Sentinel‐2 imagery, WWF Hydrosheds, 90 m DEM (SRTM), and JRC Global Surface Water Dataset are also used to conduct the flood inundation mapping for 80,236 (1 × 1 km) grids in Assam and morphometric analysis in the lower Brahmaputra River Basin. We also carried out detailed mapping and geomorphic analysis on channel and sediment bar area and channel width in 100 (6.25 km wide) grids on the Brahmaputra River in Assam. Our results suggest that high flood inundation areas are well scattered in Assam, and Lakhimpur, Nagaon, Sibsagar, Sonitpur, and Barpeta are the most affected districts from 2018 to 2020. We also observe that the geomorphology can act as primary control over the frequent flood inundation in highlighted areas in Assam. On a basin scale, we recognize a low correlation between the stream power index, topographic wetness index, plan curvature, slope, and flood inundated areas, mainly due to the coarse resolution of the digital elevation model. In a braided river system like the Brahmaputra River, the channel and bar area and channel width are well correlated during flood events. Finally, the channel sinuosity also controls channel and bar area and width during regular and influential flood events.