Pol-SAR time series for soil moisture estimation under vegetation

Inhalt:
In this paper the problem of soil moisture estimation under vegetation is investigated. Pol-SAR inversion algorithms are developed and applied on the multi-temporal AGRISAR data set from 2006 to investigate their potential and limitations over time. Modifications of the model based Freeman decomposition are used to decompose the scattering contributions and to invert the surface and dihedral scattering components for soil moisture estimation at C- and L-band. The applied decomposition approaches for agricultural land surfaces are suitable but indicate insufficiencies in modeling the volume and the dihedral components, even though some improvements due to the modifications could be shown. Finally this results in a soil moisture estimation which does not perform sufficiently well especially over the whole acquisition period.