POLARIMETRIC OPTIONS FOR SAR TOMOGRAPHY OFFORESTED AREAS

The aim of this paper is to discuss the potentialities of SAR Tomography to infer information about the vertical structure of forested areas, basing on the availability of either multi-polarimetric or single-polarimetric data. The core of the analysis is represented by a model of the second order statistics of the multi-baseline, multi-polarimetric data that describes the imaged scene as the superposition of two distributed scattering mechanisms. In this framework, the well posing of the Tomographic problem can be discussed in a simple fashion, by considering the number of equations that arise from the data-set with respect to the number of the unknowns required to solve the model. As a result, it follows that through the exploitation of strongly structured models the vertical structure can be retrieved basing on single-polarimetric data as well, provided that a sufficient number of baselines is available. Still, the availability of multi-polarimetric and multi-baseline data offers the possibility to face the problem of the retrieval of the vertical structure of each scattering mechanism from a very general point of view, allowing to relax many of the assumptions that are otherwise mandatory. In other words, the availability of multipolarimetric data makes it possible to exploit not only model based approaches, but also model free, and hybrid approaches. As a result of this analysis, a new general methodology will be presented for the processing of multi-polarimetric and multibaseline data, that is consistent the inversion procedures usually exploited in single-baseline PolInSAR in the case where only single baseline data are available. Experimental results will be shown basing on a data-set of 9 P-Band SAR images of the forest site of Remningstorp, Sweden, acquired under the framework of the ESA campaign BioSAR 2007.