Prediction and observation of SAR clutter from vegetation canopies

A technique for the simulation of vegetation canopies is reported. Monte Carlo simulations are coupled to a first order, mean field coherent scattering model and used to predict the distribution of polarimetric microwave backscatter cross section. Scattering is calculated in the distorted Born approximation with Green´s functions for layered, anisotropic vegetation. The same layered Green´s functions are employed in the calculation of the direct backscatter term from the underlying surface. Monte-Carlo techniques are used to simulate the distribution of vegetation, as well as to generate an ensemble of surfaces with given correlations and height distributions. The model is applied both to a grassy canopy at X-band and a forest canopy at low frequency. The results are compared with SAR observations of both canopies. The utility of the models and the nature of clutter distributions are discussed