Multiyear independent validation of the water cloud model for retrieving maize leaf area index from SAR time series

Taking advantage of a large multiyear data set of synthetic aperture radar (SAR) and ground observations collected in Belgium, this research aims at improving the understanding of the SAR signal sensitivity to crop growth by means of water cloud model (WCM) inversion for retrieving maize leaf area index (LAI) from C-band and VV-polarized SAR data. The results show that at intermediate moisture levels, the contributions of both soil and plants to the SAR response are confused as, to the SAR sensor, the vegetation seems to behave as bare soil of about 21% water content. Moreover, as the WCM usually required a calibration every year, this research assessed the robustness of the calibrated WCM by model cross-validation between years for maize. Ten different calibrations and inversions of the WCM were completed based on three years of observations. Two other years of observation serve as independent data sets to calculate the LAI retrieval error. The results demonstrate the capability of transferring the model calibration to independent subsequent crop seasons with an acceptable performance reduction.