Flood Mapping Using Sentinel-1 SAR Data

Satellite-based flood monitoring is a powerful tool to map inundated areas and distinguish water, vegetation, and urban settlements. In this study, Synthetic Aperture Radar (SAR) images from Sentinel-1 satellite were used for this purpose as images are available round the clock. Also, being a passive microwave sensing radar, the images obtained are not affected by cloud cover. The satellites Sentinel-1A and 1B combined have a 6-day revisit frequency which can be used to create a near-real-time flood inundation map. The study area is the areal extent of the Nilambur municipality in the Chaliyar river basin in Kerala, India. This area was very badly affected during the floods of 2018 and 2019. Google Earth Engine (GEE) was used to process the Sentinel-1A and 1B images. Manual selection of threshold backscattering coefficient values was used to determine open water (−16 dB) and submerged urban areas with water (−5 dB). The extent of flooding was analyzed by plotting the time series of the number of flooded pixels in the images of the area during the major flood events of 2018 and 2019. Flood maps were generated for pre and post-flood events.