On the imaging of exposed intertidal flats by single- and dual-co-polarization Synthetic Aperture Radar

Abstract We used 26 Radarsat-2 and TerraSAR-X single- and dual-co-polarization Synthetic Aperture Radar (SAR) images of a test site on the German North Sea coast to investigate the influence of imaging geometry and environmental conditions (wind speed, water level, and vegetation period) on the radar return from exposed intertidal flats. Multi-temporal analyses of single- (VV-) polarization SAR imagery indicate that the surface roughness is more variable at shorter scales (responsible for the X-band backscatter) than at longer scales (C-band). Less variation at both radar bands was found in sea-grass meadows. TerraSAR-X dual-co-polarization data were used for polarimetric analyses based on a decomposition of the Kennaugh matrix, whose elements provide information on the total intensity at both co-polarizations and on the relative strength of even- and odd-bounce backscattering. At steep incidence angles (around 30°) the radar backscatter from bare sand flats is similarly strong at both co-polarizations, while the vertically polarized radar return dominates at higher incidence angles (above 40°). At low water levels, resulting in lower moisture of the sandy sediments, strong single-bounce radar backscattering was observed, while higher water levels (and moisture) caused weak mixed (single- and double-bounce) backscattering. Apart from the absolute water level its history, e.g. the time and level of the closest low tide, must be considered. During the vegetation period, sea-grass meadows cause a stronger increase in horizontally (HH-) polarized radar backscatter, along with an increased double-bounce backscattering. We conclude that the Kennaugh element framework has potential to be used for classification purposes in intertidal areas, but also that, for a full interpretation of the SAR imagery, the exact topography and the surface roughness have to be known.