Modeling backscattering variations due to flooding over vegetated surfaces

The objective of this study is to evaluate the possibility to identify floods when soil is covered by vegetation, recognizing differences in backscattering between SAR measurements acquired before and after the event. To this end, the electromagnetic model developed at Tor Vergata University is run for an early stage and a fully grown wheat field, at L, C and X band. Theoretical results show that flooding can be detected at all frequencies in presence of short vegetation, while L band is the most suitable one for developed crops. Results obtained in the case of early stage are in reasonable agreement with ERS data collected over grass fields before and after the strong flooding occurred in Alessandria (North Italy) in 1994.