Orientation Angle Calibration for Bare Soil Moisture Estimation Using Fully Polarimetric SAR Data

Xinyi Shen Xinyi Shen,T Grout,Shaohua Zhao Shaohua Zhao,Sheng Chen Sheng Chen,Qiming Qin Qiming Qin,Weilin Yuan Weilin Yuan,Yang Hong Yang Hong

doi:10.1109/tgrs.2011.2158583

Abstract

This paper focuses on assessing the effectiveness of applying orientation angle calibration to polarimetric synthetic aperture radar (PolSAR) data for soil moisture estimation. We employ Cloude-decomposition-based method to estimate the orientation angle because it can relate a scatter-distributed pixel to its major component of an equivalent "pure target," use the Jet Propulsion Laboratory/Airborne Synthetic Aperture Radar L-band fully polarimetric data to validate the proposed method, and observe results in good agreement after orientation angle compensation is employed. Specifically, root mean square errors of measured radar backscattering coefficients σ hh 0 and σ vv 0 and copolarization ratio versus advanced integral equation model predictions are reduced significantly from 1.95, 1.33, and 2.03 dB to 1.30, 1.15, and 1.43 dB, respectively. The compensated copolarized backscattering coefficients are also used as inputs to a novel inversion model to estimate the dielectric factor R hh and volumetric soil moisture m v . The results show that the estimation errors are reduced significantly from 0.075 to 0.054 and 0.056 to 0.041 for R hh and m v , respectively. This paper demonstrates the advantage of orientation angle calibration as a preprocessing for estimating bare soil moisture, particularly in agricultural areas, and the preponderance of fully PolSAR data on soil moisture estimation over dual and single polarizations.

Full Text