EXTENT MAPPING OF A MAJOR FLOODING EVENT ON THE ISLAND OF TRINIDAD USING SPACE-BORNE SYNTHETIC APERTURE RADAR

摘要

Flooding events around the world have been increasing both in their occurrence and their intensities within recent decades. Studies have shown that this is most likely linked to climate change effects and anthropogenic activities that lead to pollution. Irrespective of the cause, floods incur massive economic and human losses. Synoptic data on flooding events help to support the planning and management efforts during this disaster event. Remotely sensed data, particularly from satellites is useful for mapping and monitoring large scale flooding events. More specifically, synthetic aperture radar (SAR) allows for data acquisition despite the interference of clouds and other atmospheric elements such as fog, light rain and mist. This study utilized SAR data from the Sentinel–1 satellite to map a major flooding event on the island of Trinidad which occurred during October 18–21, 2018. The peak of the flooding was estimated to have occurred on October 20, 2018. The SAR images were first calibrated then geometrically corrected and filtered. A threshold method was then applied to extract the inundated areas. A proprietary algorithm implemented by Geospatial Enabling Technologies (GET) and based on SNAP software, was used for processing Sentinel-1 imagery to separate the open water and non-water (land) areas from the images. Outputs were then integrated into ArcGIS 10.6 mapping software and the extents of the flooded areas were delineated based on the available data. By applying this method to a Sentinel–1 image captured on October 19, 2018 it was revealed that the total flooded area on that date was 9.94 square kilometres. This study provides a brief illustration of the value of SAR data for flood delineation and mapping but also highlights some of the limitations that can be involved when using such technology.