Typhoon Doksuri inflicted significant flooding across extensive regions in Vietnam and other countries in September 2017. An innovative satellite remote sensing method* has been developed and implemented for use with Sentinel synthetic aperture radar (SAR) satellite data at a resolution of 10 meters to identify, map, and monitor inundation including pre-existing/pre-flood water bodies (e.g., lakes, reservoirs, rivers, coastal ocean, etc.) and newly flooded areas after the typhoon made landfall. Because Sentinel SAR operates at C-band microwave frequency, it can be used for flood mapping regardless of could cover conditions typically associated with storms, and thus can provide immediate results without the need to wait for the clouds to clear out.
Attached is a high-resolution 4K-UHD flood map of a region around Hà Tĩnh (north central coast of Vietnam) showing flood inundated areas (in yellow) on 16 September 2017 together with pre-existing/pre-flood surface water (in blue) on 4 September 2017. This is just one example selected from a larger flood map covering an extensive region of about 250 km x 680 km along the central coast of Vietnam. Besides Typhoon Doksuri, flooding in Japan and China occurred as a consequence of Typhoon Talim.
*Note on flood mapping method: This flood mapping method has been developed for urgent support to the U.S. Federal Emergency Management Agency (FEMA) in a national emergency response to Hurricanes Harvey, Irma, and Maria that have caused catastrophic flooding that devastated a number of regions in U.S., Puerto Rico, and Virgin Islands. The flood mapping method that we've developed at JPL has been verified in five different ways: (1) with clear-sky high-resolution WorldView images,when zoomed-in WorldView images, you can even see waves in inundated areas detected by Sentinel SAR, (2) with known surface water such as lakes, reservoirs, rivers, coastal ocean, etc., (3) with most-recently updated Google Earth images, (4) with in-situ photographs and surface observations at several locations, and (5) with other flood maps derived from Landsat and MODIS data at lower resolutions. Note that we are making these flood mapping products in urgency for immediate release regarding emergency responses to hurricane/typhoon flood impacts, and we have not filtered out miscellaneous things like speckles, noisy pixels, isolated misidentified pixels, etc. These can be removed/accounted for when having more time in the future.
**Acknowledgments – This research on satellite Sentinel SAR observations of inundation to obtain flood maps for assisting FEMA in emergency responses to hurricane flooding impacts, carried out at the Jet Propulsion Laboratory, California Institute of Technology, has been supported by the NASA Land-Cover and Land-Use Change Program and by the NASA Terrestrial Hydrology Program. These results contained modified Copernicus Sentinel data (2017).