Sensitivity Study of Multiconstellation GNSS-R to Soil Moisture and Surface Roughness Using FY-3E GNOS-II Data

The potential of spaceborne global navigation satellite system reflectometry (GNSS-R) to retrieve a variety of geophysical parameters has already been demonstrated in numerous studies. In 2021, China successfully launched the Fengyun-3E (FY-3E) polar orbit satellite. It carries the GNSS occultation sounder-II (GNOS-II) that can simultaneously receive reflected signals from the global positioning system, BeiDou, and Galileo constellations. Multiconstellation measurement significantly reduces the revisit time of the spaceborne GNSS-R data. This offers the possibility to study the sensitivity of the reflected signals from multi-GNSS constellations to surface parameters at global scale. The main objective of this article is to analyze the sensitivity of FY-3E GNOS-II surface reflectivity (SR) to soil moisture (SM), vegetation, and surface roughness for different GNSS constellations, and different incidence angles. FY-3E data from May 2023 to October 2023 were collected along with SM data from the soil moisture active passive (SMAP). The SMAP static auxiliary surface roughness parameter SMAP-h was also collected for subsequent comparisons. Furthermore, the effect of vegetation in the reflected signal was accounted for using the SM and ocean salinity INRA-CESBIO L-band vegetation optical depth. Afterward, for each GNSS constellation, the SR was binned as a function of SM, SMAP-h, and incidence angle. The results indicate that the sensitivity for the different GNSS constellations is consistent, showing similar behavior. Moreover, this study also reports for the first time the experimentally computed sensitivity of FY-3E SR to SM under different SMAP-h values. The sensitivity of GNOS-II SR to SM is in general agreement with the values in the previous studies, demonstrating the feasibility of using single-pass multi-GNSS constellation GNSS-R data to retrieve surface parameters, such as SM, with shorter revisit times. Furthermore, the results of this study re-emphasize the nonlinear sensitivity of SR to SM. As compared with higher SM, SR is more sensitive at lower SM. This highlights the shortcomings of using linear models to retrieve SM. Meanwhile, another important finding is that, for spaceborne GNSS-R data, SMAP-h may have underestimated the effective surface roughness.