Due to the unprecedented long rainy season and typhoon that occurred on the Korean
Peninsula in this year, flood damage has appeared throughout the basin and urban areas. To
monitor water-related disasters, it is more important than anything else to have a preemptive
response system rather than follow-up measures, and to establish a national system based on
rapid acquisition of remote sensing data that can reduce disaster damage. In this study, the
flood damage on the Korean Peninsula was analyzed using SAR (Synthetic Aperture Radar) that
capable of observing water body in surface through clouds during a flood disaster.
The SAR remote sensing data used in this study are Sentinel-1, Cosmo-SkyMED, TerraSAR-X,
ICEYE, KOMPSAT-5, and etc. We applied both radiometric and geometric calibration for
calculating the georeferenced backscattering coefficient in SAR data. And it applied the speckle
filter and the Gamma-distribution fitting technique are used to extract the water body for
mapping flooded area. The gamma distribution of the water body and the land part was
extracted from the histogram of the image and the optimum threshold was set based on this,
and the accuracy of the water body was improved by about 90% or more. In addition, the
detection accuracy of the flooded area was improved by correcting the terrain distortion of
SAR image in consideration of the slope and curvature of the DEM. For rapid flood area
detection, we applied the automatic process during the heavy rain and the typhoon (Aug 7th
2020~sep 9th 2020).
Despite our effective process and results, there were limitations in analyzing and utilizing
rapid acquisition of satellite data as it was difficult to observe and secure satellites at the time
of the most severe flood damage. In the future, it is believed that it is possible to establish a
flood monitoring and response system across the watershed through the use of various
commercial satellites and the development of C-band SAR satellite (CAS500-5; Compact
Advanced Satellite 500-5) for water resources and water-related disasters to shorten the
observation period and improve precision.