| Paper: | TH-A2.18 |
| Session: | Applications of Radiometry II |
| Time: | Thursday, March 29, 09:00 - 10:20 |
| Presentation: |
Poster
|
| Topic: |
Theory, physical principles and electromagnetic models: |
| Title: |
Modeling the backscattering coefficient of random rough surfaces to improve soil moisture retrieval from polarimetric SAR data |
| Authors: |
Julian Villa; Facultad de Ciencias Exactas y Naturales - UBA | | |
| | Mariano Franco; Instituto de Astronomía y Física del Espacio | | |
| | Matías Barber; Instituto de Astronomía y Física del Espacio | | |
| | Francisco Grings; Instituto de Astronomía y Física del Espacio | | |
| Abstract: |
The scattering properties of surfaces are of fundamental importance to generate retrieval algorithms in the context of remote sensing (a typical example is soil moisture retrieval from SAR data). Due to the inherent stochastic nature of natural surfaces, the scattering function of these kind of targets is usually computed through either perturbative or approximated methods. From the latter, one of most commonly used is the Kirchhoff -or tangent plane- approximation (KA). Kirchhoff approximation is widely used because its ease of implementation. This holds if the induced currents on the surface are computed using stationary phase approximation. However, under such approach, null cross-polarization is achieved in the backscattering condition, which is not supported by observations. This is particularly relevant in the context of SAR polarimetry, since a surface modeled as having zero cross-polarization is associated with a rank-reduced coherency matrix and zero entropy values. Therefore, in order to fully exploit polarimetric SAR data, it is paramount to propose a model for the scattering of natural surfaces characterized by a non-zero cross-polarization response.
In this paper, we propose a scattering model for random rough surfaces based on KA, that predicts non-zero cross-polarization in backscattering. This is achieved by computing the overall scattering through an ensemble of periodic surfaces with both random amplitudes and frequencies. The advantage of using this scheme is that for periodic surfaces KA can be implemented without any further approximation (such as phase stationary) which imply non zero cross-pol in backscattering. Therefore, the full-rank coherency matrix of the random surface can be modeled and surface’s polarimetric features can be studied from a theoretical point of view and compared with available observations. In this paper, we compared simulations based on the proposed model with observations from UAVSAR data, in order to evaluate to which degree the proposed approach is able to explain polarimetric features observed for real surfaces.
|
