Impact of Polarimetric Dimensionality on Forest Parameter Estimation by Means of Polarimetric SAR Interferometry
Jun Su Kim(1), Seung-Kuk Lee(1), Konstantinos P. Papathanassiou(1) and Hajnsek Irena(1)
(1) German Aerospace Center, P.O. Box 1116, D-82230 Wessling, Germany
Polarimetric information is essential for the inversion of forest parameters (e.g. height and vertical structure) from single and/or multi-baseline interferometric SAR (Pol-InSAR) acquisitions ,. However, by means of system and mission design, increased polarimetric information has to be paid by increased system complexity and data volume and posed observation constrains. Quad-polarimetric systems provide the complete polarimetric information but are characterised by complex system architecture and maximum data rates. In contrast, conventional dual-polarimetric systems are characterised by a significantly reduced system complexity and allow the realisation of wider swathwidths (for the same specifications as an equivalent quad-pol system) and hence, shorter revisit times. However, the acquisition of a co- and a cross-polarimetric channel (HH and VH or VV and HV) limit the polarimetric information content especially when one considers the information gain arising from the formation of linear combinations of the two channels. The situation improves in terms of information content when the two co-polarised channels (HH and VV) are acquired, as in the case of TerraSAR-X, but this is compensated by increased T/R system complexity and swath constrains. In order to optimise the trade off between information content and system/observation complexity, the concept of compact polarimetry was proposed: It is based on the transmission of a single polarisation (linear, elliptical or circular) composed by two non-zero orthogonal polarisations that are then used for reception ,,.
In this paper we compare the different dual- compact- and quad-pol observation spaces with respect to the performance of Pol-InSAR forest parameter inversion techniques. We assess the impact of the different observation spaces with respect to different forest (sparce -> dense) and terrain (flat -> sloped) configurations. For this we use airborne quad-pol repeat-pass interferometric data acquired by DLR’s airborne Experimental SAR system at L- and P-band in the frame of different campaigns. Starting from the quad-pol case we generate the reduced observation cases and apply the inversion methodology adapting it when necessary. We compare and discuss the (model based) inversion results obtained at the different frequencies from the individual observation spaces. The differences at the two frequencies are emphasised and discussed with respect to the underlying (model and reconstruction) assumptions used.
 K.P. Papathanassiou and S.R. Cloude, ‘‘Single baseline Polarimetric SAR Interferometry’’, IEEE Transactions on Geoscience and Remote Sensing, vol, 39, no. 11, pp. 2352-2363, November 2001.
 S.R. Cloude, and K.P. Papathanassiou, ‘‘A 3-Stage Inversion Process for Polarimetric SAR Interferometry’’, accepted for publication in IEE Proceedings Radar Sonar & Navigation, vol. 150, issue 3, 2003.
 J.-C. Souyris, P. Imbo, R. Fjørtoft, S. Mingot, and J.-S. Lee, “Compact Polarimetry Based on Symmetry Properties of Geophysical Media: The π/4 Mode”, IEEE Transactions on Geoscience and Remote Sensing, vol, 43, no. 3, pp. 634-646, March 2005.
 R.K. Raney, “Hybrid-Polarity SAR Architecture”, IEEE Transactions on Geoscience and Remote Sensing, vol, 45, no. 11, pp. 3397-3404, November 2007.
 P.C.Dubois-Fernandez, J.-C. Souyris, S. Angelliaume, F. Garestier, “The Compact Polarimetry Alternative for Spaceborne SAR at Low Frequency”, IEEE Transactions on Geoscience and Remote Sensing, vol, 46, no. 10, pp. 3208-3222, October 2008.