Integration of remote sensing data and GIS for accurate mapping of flooded areas

Abstract

This paper describes a synergetic use of satellite radar images and ancillary information to detect flooded areas at their peak and evaluates its potential with mapping. The procedure was tested on the catastrophic flood that occurred in Regione Piemonte in Italy in November 1994. Two ERS-1 synthetic aperture radar (SAR) images were processed, one acquired one month before the flood and the other acquired three days after the event. Visual interpretation and two different thresholding techniques were performed. The flood map derived shows only a small fraction (20%) of the actually flooded lands because of the time delay between the flood peak and the satellite overpass. To overcome this limitation, the authors developed a new procedure to estimate the flooded area at the peak time by integrating the flooded area from SAR imagery with digital topographic data from a GIS technique. This method allowed inundated areas covering 96.7% of the flooded area officially recorded by the local government to be mapped. The proposed procedure is suitable for mapping flooded areas even when satellite data are acquired some days after the event, thus overcoming the constraint of temporal resolution in the application of SAR imagery in hydrology.