Height-dependent temporal decorrelation for POLINSAR and TOMOSAR forestry applications

Abstract:
This contribution deals with the temporal correlation of a vertical ditribution of scatterers observed by a repeat-pass interferometer. We model the temporal correlation of such a semitranparent medium by using a height-dependent function. The expression of the temporal correlation function is found by assuming differential standard deviation for the Browian motion of the scatterers. As a consequence, the function that characterizes the backscattering and structural properties of the medium results shaped along the vertical dimension depending on the temporal baseline of the interferometer. One major result is that the scattering phase center at a generic polarization is shifted due to the temporal artifacts. Estimating and compensating this phase shift bring benefits to both POLINSAR and TOMOSAR applications that aim at retrieving the vertical profile of forests. We expose the forward theoretical aspects of the temporal decorrelated random volume over ground model and briefly discuss the implications in the inversion procedure.