On the physical modelling of polarimetric parameters of forests
With the increasing availability of polarimetric and inteferometic polarimetric data from airborne SARs as well as data from the forthcoming spaceborne systems such as ALOS/PALSAR , there is a clear need to improve our understanding of the physical content of polarimetric data.
In the recent years, polarimetric parameters have been proposed to classify surface covers or oto retrieve surface bio-geophysical characteristics. Many of them are derived from algebraic concepts and may in general not be adequate to describe the physical information content. More recently, parameters based on physical modelling have been developed, especially for polarimetric interferometry. However, the approach used which relies on simplified description of the physical mechanisms may not be applicable to a large range of observation conditions. Meantime, more complex backscatter models have been developed in the last decade to interpret backscatter intensity data. Although these models are able to simulate polarimetric data, in general, they do not include all the relevant electromagnetic interactions.
This paper will deal with the physical modelling of polarimetric features of forest covers. In a first step, modelling of the polarimetric phase is addressed. A higher order model coupled with a realistic description of the cover is developed and the model results are compared with polarimetric coherence data. In a second step, requirements for polarimetric interferometry modelling is given in the light of observations in interferometry and polarimetry on different forest covers.