Soil Permittivity Estimation over Croplands Using Polsar Data

Abstract

Polarimetric Synthetic Aperture Radar (SAR) data has been extensively used to estimate soil permittivity because of its high sensitivity to the dielectric properties of the target. However, the presence of vegetation cover induces bias in the permittivity estimates. This work utilizes the scattering-type parameters: alpha <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\overline{\alpha})$</tex> and theta <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\theta_{\text{FP}})$</tex> to estimate soil permittivity using the X-Bragg as the dominant surface scattering model. A theoretical study ascertains that the recently proposed <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\theta_{\text{FP}}$</tex> is fairly robust towards the depolarizing component in the X-Bragg model. Hence, it is expected to produce better inversion accuracy. This study analyzes major phenology stages of Canola using the UAVSAR full-pol SAR data and the ground measurements acquired during the SMAPVEX12 campaign over Manitoba, Canada. The proposed method achieved an RMSE of 5.9 for soil permittivity with a Pearson coefficient, <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$r=0.83$</tex> . Further, the temporal trend of the soil permittivity estimates also agrees with in-situ measurements for the entire timeframe.