Characterizing the C-Band Backscattering of Winter-Wheat Canopy with a Microwave Radiative Transfer Model

Accurate simulation of microwave scattering characteristics of wheat canopy is crucial for understanding the microwave scattering mechanism of wheat canopy and quantitative retrieval of biophysical parameters of wheat crop from radar remote sensing data. In this study, a wheat canopy scattering model (WCSM) was developed on a basis of 1st-order microwave radiative transfer equation. Multiple temporal C-band quad-polarized (Radarsat-2 and GaoFen-3) and dual-polarization (Sentinel-1) SAR data were utilized to validate the performance of the WCSM. Results showed that WCSM simulated C-band backscattering coefficients of wheat canopy is with a RMSE of < 1.8 dB for Radarsat-2 and GaoFen-3 SAR data and Sentinel-1 data at incident angle of 43.15°. Larger simulation error (>3.0 dB) mainly distributed in a few of wheat fields in which the row is perpendicular and parallel to the radar beam. Besides, the large deviations (RMSE of >3.4 dB) for simulated Sentinel-1 data with incident angle of 32.12° could be partly attributed to the rough resolution and radiometric errors, which is needed to be further investigated. This study demonstrated the proposed WCSM is effective in characterizing the C-band scattering features of wheat at various growth stages. By integrating with a robust inversion algorithm, it would become a useful tool in quantitative retrieval of wheat canopy parameters, and further for grain yield estimation at local or regional scales.