Simultaneously estimating surface soil moisture and roughness of bare soils by combining optical and radar data

Both radar and optical signals are sensitive to the change of surface soil moisture (SSM) and surface roughness properties (such as root mean squared height- RMSH), and the accuracy of retrieved SSM from single radar and optical remote sensing data is influenced by the spatiotemporal change of surface roughness. Here, we attempt to explore a method to simultaneously estimate SSM and RMSH of bare soil by combining optical and radar data, so as to weaken the effect of surface roughness on SSM inversion results. To achieve this goal, two satellite synchronous ground experiments were carried out, collecting 88 sampling plots each with an area of 50 m × 50 m. Radar backscattering coefficient and spectral reflectance are uniformly corrected to a fixed observation direction and solar incident direction respectively, which can eliminate the difference of satellite signal resulted from various sun-satellite geometry. Combining radar backscattering and optical reflectance model, Sentinel-1 and Sentinel-2 data are used to simultaneously retrieve SSM and RMSH of bared soils, and some conclusions are given as below: 1) a strong correlation is observed for (radar and optical) satellite signals and soil surface parameters (SSM and RMSH); 2) a higher accuracy was obtained by the combined use of optical and radar data, indicated by the decreased root mean squared error of retrieved SSM (~0.045 cm3/cm3) and RMSH (~0.8 cm); 3) the further improvement of retrieved SSM and RMSH was achieved by introducing their initial values, revealing that the prior knowledge of soil properties is also beneficial to improve the retrieval accuracy. This study proposed an framework for simultaneous estimation of SSM and RMSH by combining optical and radar data, and its feasibility is verified by experimental data.