A Multi-Data Geospatial Approach for Understanding Flood Risk in the Coastal Plains of Tamil Nadu, India

The coastal plains of Tamil Nadu, India, are prone to floods, the most common disaster experienced in this region almost every year. This research aims to identify flood risks in the coastal plain region of Tamil Nadu, delineated through a watershed approach with 5020 micro-administrative units covering an area of about 26,000 sq. km. A comprehensive flood risk assessment covering hazard, vulnerability, and exposure parameters was carried out using multiple datasets derived from field surveys, satellite data, and secondary data sources. The flood hazard layer was prepared on a probability scale (0-1) with the help of Sentinel-1 Synthetic Aperture Radar data coupled with GIS-based water rise modelling using Shuttle Radar Topography Mission Digital Elevation Model (SRTM-DEM) and reports of the District Disaster Management Plans of 13 coastal districts. In addition, the National Resources Conservation Service-Curve Number (NRCS-CN) method was adopted to estimate surface runoff potential for identifying low probability flood-prone regions. The vulnerability and exposure of the population to flood hazards were determined using census and household data-based indicators. The different categories of built-up areas were delineated and intersected with the flood hazard layer to estimate elements at flood risk. An exhaustive field survey was conducted at 514 locations of the study area, targeting deprived communities of all major settlements to validate the flood hazard layer and understand the public perceptions. The amalgamation of results shows that very high flood risk prevails in the northern parts of coastal Tamil Nadu, especially the stretch between Chennai and Cuddalore. In addition, to provide baseline datasets for the first time at micro-administrative units for the entire coastal plains of Tamil Nadu, the study offers a pragmatic methodology for determining location-specific flood risks for policy interventions.