Using ERS-2 and ALOS PALSAR images for soil moisture and inundation mapping in Cyprus

Abstract

Floods are among the most frequent and costly natural disasters in terms of human and economic loss and are considered to be a weather-related natural disaster. This study strives to highlight the potential of active remote sensing imagery in flood inundation monitoring and mapping in a catchment area in Cyprus (Yialias river). GeoEye-1 and ASTER images were employed to create updated Land use /Land cover maps of the study area. Following, the application of fully polarimetric (ALOS PALSAR) and dual polarimetric (ERS - 2) Synthetic Aperture Radar (SAR) data for soil moisture and inundation mapping is presented. For this purpose 2 ALOS PALSAR images and 3 ERS-2 images were acquired. This study offers an integrated methodology by the use of multi-angle radar images to estimate roughness and soil moisture without the use of ancillary field data such as field measurements. The relationship between soil moisture and backscattering coefficient was thoroughly studied and linear regression models were developed to predict future flood inundation events. Multi-temporal FCC images, classification, image fusion, moisture indices, texture and PCA analysis were employed to assist soil moisture mapping. Certain land cover classes were characterized as flood prone areas according to statistics of their signal response. The results will be incorporated in an integrated flood risk assessment model of Yialias catchment area.