GNSS reflection measurements in the form of delay-Doppler maps (DDM) from the CYGNSS constellation can be used to complement soil measurements from the SMAP Mission, which has a revisit rate too slow for some hydrological/meteorological studies. The standard approach, which only considers the peak value of the DDM, is subject to a significant amount of uncertainty due to the fact that the peak value of the DDM is not only affected by soil moisture, but also complex topography, inundation, and overlying vegetation. We hypothesize that information from the entire 2D DDM could help decrease uncertainty under these conditions. The application of deep learning based techniques has the potential to extract additional information from the entire DDM, while simultaneously allowing for incorporation of additional contextual information from external datasets. This work explored the data-driven approach of convolutional neural networks (CNNs) to determine complex relationships between the reflection measurement and surface parameters, providing a mechanism to achieve improved global soil moisture estimates. A CNN was trained on CYGNSS DDMs and contextual ancillary datasets as inputs, with aligned SMAP soil moisture values as the targets. Data was aggregated into training sets, and a CNN was developed to process them. Predictions from the CNN were studied using an unbiased subset of samples, showing strong correlation with the SMAP target values. With this network, a soil moisture product was generated using DDMs from 2018 which is generally comparable to existing global soil moisture products, but shows potential advantages in spatial resolution and coverage over regions where SMAP does not perform well.

Recent proof-of-concept experiments have demonstrated the potential utility of Signals of Opportunity (SoOp) in remote sensing. SoOp methods involve the re-use of existing satellite transmissions as sources in bistatic radar, applying fundamental physical principles to estimate surface and scattered medium properties from reflectivity and phase observables in the reflected signal. Through utilizing signals intended for communications, SoOp methods can make these observables using frequencies that are not allocated or protected for scientific use. Two promising applications in hydrology have been studied: Sub-canopy root-zone soil moisture (RZSM) using satellite communications signals below 500 MHz and snow water equivalent (SWE) retrieval from the observed phase different through propagation through the snow layer. Signals of Opportunity P-band Investigation (SNOOPI) is a NASA Cubesat technology demonstration mission to test forward scattering models and validate a prototype instrument for SoOp reflectometry in 250-380 MHz range. Contribution to the panel discussion will focus on the expected contributions of the SoOp techniques validated in the SNOOPI mission and the existing challenges in the full utilization of SoOp methods for both RZSM and SWE remote sensing. Multiple frequencies are required in order to solve the inverse problem and estimating a sub-surface profile. In the case of SoOp, this may require combining observations with diverse geometry due to the different orbits of the potential sources. This presents new challenges in the development of retrieval algorithms and may possibly require the integration of additional data sources. Another important challenge for SWE retrieval is the need for repetitive coverage to extract phase differences between subsequent passes, coupled with orbit determination for the non-cooperative sources. In contrast to GNSS reflectometry (in which high-precision orbits are publicly available for use in positioning), communication satellite orbits are not known to the required meter-level accuracy. Even geostationary sources frequently have a small inclination which results in motion relative to the surface of the Earth. Finally, antenna calibration is a substantial contribution to the error budget.