Sensitivity of Pol-InSAR Measurements to Vegetation Parameters

Abstract

Estimation of forest height from combined polarimetric and interferometric synthetic aperture radar (Pol-InSAR) measurements has been the focus of radar remote sensing studies in the past decade. The simplicity of the random-volume-overground (RVoG) model makes it one of the most widely used candidates for estimating canopy height. However, the polarization-independent extinction coefficient assumption in the RVoG model fails in some certain types of the canopies, as suggested by the oriented-volume-over-ground (OVoG) model. The sensitivity of coherence magnitude and phase to different parameters of the canopy is expressed in a closed-form formulation in this paper for the first time. In order to simplify our formulation, the forest is represented by a layer of discrete randomly distributed dielectric scatterers over ground, with azimuthal symmetry. The sensitivity analysis of this work quantifies the contribution of differential extinction due to polarization change in interferometric coherence. Therefore, we can quantitatively evaluate whether the RVoG model is accurate enough to be used for a specific kind of canopy or the OVoG model is needed for better estimation. A simple layer of leaves over ground is used to simulate the sensitivity of Pol-InSAR measurements to different parameters.

Estimation of RVoG Scene Parameters by Means of PolInSAR With TanDEM-X Data: Effect of the Double-Bounce Contribution

“…Numerous tests showed that all solutions fulfilling the condition of 5% of the RVoG line length provide the accuracy required to ensure satisfactory parameter estimations. When the resulting optimized solution does not converge to the established criterion, a different initial guess for the vegetation height (which in turn is known to be the most sensitive parameter [18], [19], [24], [28]) is generated randomly using a uniform distribution. The ranges in which the new guess is generated are the same as those we use to generate the simulated scenes (later specified in Section IV and Table II): h v,random ∈ [0, HoA /2] m for the crops and h v,random ∈ [2, HoA /2] m for the forests.…”

Forest Vertical Parameter Estimation Using PolInSAR Imagery Based on Radiometric Correction

“…In addition to the height of the vegetation, extinction coefficient and other vertical parameters can be derived. Based on the PolInSAR complex coherent coefficient model proposed by Oveisgharan [9], Papathanassiou transforms the inversion procedure into a classical non-liner parameter estimation problem, which is referred to as the six-dimension non-liner iteration method. It provides many valuable parameters simultaneously, however, calculation is time consuming, and it strongly depends on the initial value and tends to fall into the local optimal solution [10].…”