Coherent Modeling of Vegetation for Polarimetric SAR Interferometry Applications

Abstract

In order to improve the knowledge about individual scattering processes and their effects on the backscattered signal accurate modeling of SAR backscatter is required. In this paper we present a stochastic coherent polarimetric scattering model which can be used for interferometric simulation. Using simulated examples of several interferometric scenarios, including surface scattering and small particle volume scattering, we will outline new possibilities for the extraction of physical properties of the scatterers. In the case of surface scattering the scattering matrix [S] can be calculated either using the Physical Optics approximation (PO) or the Small Perturbation Model (SPM --> Bragg surface), depending on the surface roughness compared to the sensor wavelength. Volume scattering, e.g. from forest canopies, may be simulated by calculating [S] of small 3-dimensional ellipsoids and by arranging them to an elevated cloud of particles with given orientations and positions in height. Using a coherent Monte Carlo model we are able to apply both polarimetric and interferometric techniques in order to investigate the influence of the physical properties of the scatterers on polarimetric and interferometric observables. In this paper we present simulation results for surface and volume scattering and discuss them with special regard to known problems and the resulting future enhancements. Furthermore we investigate the influence of temporal changes on the interferometric coherence regarding repeat-pass mode applications.