Microwave scattering from forest canopies

In modeling the scattering for forest canopies at microwave frequencies, several scattering mechanism are usually taken into account: 1) crown volume scattering, 2) trunk volume scattering, 3) surface scattering from forest floor, 4) trunk-surface interaction, 5) trunk-crown interaction, 6) crown-ground interaction. Understanding the behavior of each scattering component for various forest canopies is dependent on the derivation of the models and the underlying assumptions. A majority of the models are discrete scattering models where the vegetation canopy can be viewed as an ensemble of randomly distributed scatterers (leaves, branches, and trunks) in one or many layers. In this study, three commonly used discrete scattering models, distorted Born approximation (DBA), Born approximation (BA) and radiative transfer theory (RT), are compared. The DBA and BA models are formulated according to a field theory point of view and the RT model as an intensity or transport theory. In all three models, each scattering mechanism is formulated explicitly so that its relative contribution to the total backscattering coefficient can be analyzed. The models are used to simulate polarimetric radar data over coniferous and deciduous forests by using the same input parameteres. The model simulations are then analyzed to check the effect of canopy parameters and the importance of scattering mechanisms. The results are also compared with radar images at P-, L-, and C-bands in to order to assess the validity of the model comparisons.<
>