This paper presents a learning-based, physics-aware soil moisture (SM) retrieval algorithm for NASA’s Cyclone Global Navigation Satellite System (CYGNSS) mission. The goal of the proposed novel method is to advance CYGNSS-based SM estimations, exploiting the spatio-temporal resolution of the GNSS reflectometry (GNSS-R) signals to its highest potential within a machine learning framework. The methodology employs a fully connected Artificial Neural Network (ANN) regression model to perform SM predictions through learning the nonlinear relations of SM and other land geophysical parameters to the CYGNSS observables. In situ SM measurements from several International SM Network (ISMN) sites are used as reference labels; CYGNSS incidence angles, derived reflectivity and trailing edge slope (TES) values, as well as ancillary data, are exploited as input features for training and validation of the ANN model. In particular, the utilized ancillary data consist of normalized difference vegetation index (NDVI), vegetation water content (VWC), terrain elevation, terrain slope, and h-parameter (surface roughness). Land cover classification and inland water body masks are also used for the intermediate derivations and quality control purposes. The proposed algorithm assumes uniform SM over a 0.0833 ${}^{\circ }\times$ 0.0833 ${}^{\circ }$ (approximately 9 km × 9 km around the equator) lat/lon grid for any CYGNSS observation that falls within this window. The proposed technique is capable of generating sub-daily and high-resolution SM predictions as it does not rely on time-series or spatial averaging of the CYGNSS observations. Once trained on the data from ISMN sites, the model is independent from other SM sources for retrieval. The estimation results obtained over unseen test data are promising: SM predictions with an unbiased root mean squared error of 0.0544 cm ${}^3$ /cm ${}^3$ and Pearson correlation coefficient of 0.9009 are reported for 2017 and 2018.

Soil moisture (SM) derived from satellite-based remote sensing measurements plays a vital role for understanding Earth’s land and near-surface atmosphere interactions. Bistatic Global Navigation Satellite System (GNSS) Reflectometry (GNSS-R) has emerged in recent years as a new domain of microwave remote sensing with great potential for SM retrievals, particularly at high spatio-temporal resolutions. In this work, a machine learning (ML)-based framework is presented for obtaining SM data products over the International Soil Moisture Network (ISMN) sites in the Continental United States (CONUS) by leveraging spaceborne GNSS-R observations provided by NASA’s Cyclone GNSS (CYGNSS) constellation alongside remotely sensed geophysical data products. Three widely-used ML approaches—artificial neural network (ANN), random forest (RF), and support vector machine (SVM)—are compared and analyzed for the SM retrieval through utilizing multiple validation strategies. Specifically, using a 5-fold cross-validation method, overall RMSE values of 0.052, 0.061, and 0.065 cm³/cm³ are achieved for the RF, ANN, and SVM techniques, respectively. In addition, both a site-independent and a year-based validation techniques demonstrate satisfactory accuracy of the proposed ML model, suggesting that this SM approach can be generalized in space and time domains. Moreover, the achieved accuracy can be further improved when the model is trained and tested over individual SM networks as opposed to combining all available SM networks. Additionally, factors including soil type and land cover are analyzed with respect to their impacts on the accuracy of SM retrievals. Overall, the results demonstrated here indicate that the proposed technique can confidently provide SM estimates over lightly-vegetated areas with vegetation water content (VWC) less than 5 kg/m² and relatively low spatial heterogeneity.

This paper presents a machine learning (ML) framework to derive a quasi-global soil moisture (SM) product by direct use of the Cyclone Global Navigation Satellite System (CYGNSS)’s high spatio-temporal resolution observations over the tropics (within $\pm 38°$ latitudes) at L-band. The learning model is trained by using in-situ SM data from the International Soil Moisture Network (ISMN) sites and various space-borne ancillary data. The approach produces daily SM retrievals that are gridded to 3 km and 9 km within the CYGNSS spatial coverage. The performance of the model is independently evaluated at various temporal scales (daily, 3-day, weekly, and monthly) against Soil Moisture Active Passive (SMAP) mission’s enhanced SM products at a resolution of 9 km × 9 km. The mean unbiased root-mean-square difference (ubRMSD) between concurrent (same calendar day) CYGNSS and SMAP SM retrievals for about three years (from 2017 to 2019) is 0.044 cm${}^3$ cm${}^{-3}$ with a correlation coefficient of 0.66 over SMAP recommended grids. The performance gradually improves with temporal averaging and degrades over regions regularly flagged by SMAP such as dense forest, high topography, and coastlines. Furthermore, CYGNSS and SMAP retrievals are evaluated against 170 ISMN in-situ observations that result in mean unbiased root-mean-square errors (ubRMSE) of 0.055 cm${}^3$ cm${}^{-3}$ and 0.054 cm${}^3$ cm${}^{-3}$, respectively, and a higher correlation coefficient with CYGNSS retrievals. It is important to note that the proposed approach is trained over limited in-situ observations and is independent of SMAP observations in its training. The retrieval performance indicates current applicability and future growth potential of GNSS-R-based, directly measured spaceborne SM products that can provide improved spatio-temporal resolution than currently available datasets.

SCoBi-Veg stands for Signals of opportunity Coherent Bistatic scattering model for Vegetated terrains. It simulates polarimetric reflectometry of vegetation canopy over a flat ground using a Monte Carlo scheme. The model is aimed at assessing the value of navigation and communication satellite Signals of Opportunity in a range of frequencies from P- to S-bands for remote sensing of a number of geophysical land parameters such as soil moisture and biomass. A fully polarimetric expression for bistatic scattering from a vegetation canopy is first formulated for a general case and is then specialized to the practical case of ground-based/low-altitude platforms with passive receivers overlooking vegetation using the signals transmitted from large distances. Using analytical wave theory in conjunction with distorted Born approximation, the transmit and receive antenna effects (i.e., polarization crosstalk/mismatch, orientation, and altitude) are explicitly accounted for. The forward model developed here enables the understanding of the effect of different geophysical parameters and system configurations on the coherent and incoherent components of the reflected signatures. It can thus help developing robust inverse algorithm for extraction of soil moisture and biomass. The model is applied to P-band signals of geostationary communication satellites to describe polarimetric reflections from tree canopies as observed from down-looking platforms at various altitudes. The relative contributions of diffuse and specular scattering on total reflected power and reflectivity are quantified for various observing scenarios.

National Aeronautics and Space Administration's Cyclone Global Navigation Satellite System (CYGNSS) mission has gained significant attention within the land remote sensing community for estimating soil moisture (SM) by using the Global Navigation System Reflectometry (GNSS-R) technique. CYGNSS constellation generates Delay-Doppler Maps (DDMs), containing important Earth surface information from GNSS reflection measurements. Previous studies considered only designed features from CYGNSS DDM, whereas the whole DDM image is affected by SM, inundation, and vegetation. This paper presents a deep learning (DL) based framework for estimating SM in the Continental United States by leveraging spaceborne GNSS-R DDM observations provided by the CYGNSS constellation along with remotely sensed geophysical data. A data-driven approach utilizing convolutional neural networks (CNNs) is developed to determine complex relationships between the reflected measurements and surface parameters which can provide improved SM estimation. The model is trained jointly with three types of processed DDM images of analog power, effective scattering area, and bistatic radar cross-section with other auxiliary geophysical information such as elevation, soil properties, and vegetation water content (VWC). The model is trained and evaluated using the Soil Moisture Active Passive (SMAP) mission's enhanced SM products at a 9 km resolution with VWC less than 5 kg/m ² . The mean unbiased root-mean-square difference between concurrent CYGNSS and SMAP SM retrievals from 2017 to 2020 is 0.0366 m ³ /m ³ with a correlation coefficient of 0.93 over fivefold cross-validation and 0.0333 m ³ /m ³ with a correlation coefficient of 0.94 over year-based cross-validation at spatial resolution of 9 km and temporal resolution similar to CYGNSS mission.

Global navigation satellite system reflectometry (GNSS-R) has the potential to offer a cost-effective solution for global land observations. In this study, we aim to understand GNSS-R sensitivity to changing land geophysical parameters. For this objective, we performed simulations of a ground-based receiver using a recently developed coherent bistatic vegetation scattering model (SCoBi-Veg) to detect GNSS-R signatures under varying soil moisture (SM), vegetation water content (VWC), and surface roughness during a full corn growing season. We modeled different corn growth stages by using in situ measurement data. We analyzed the simulated reflectivity and received power values based on the aforementioned variable input parameters. This study demonstrates that specular reflections dominate the diffusely scattered contribution in case of moderate roughness, regardless of the corn field row structure or the polarization. Significant correlations between VWC and cross-polarized reflectivity values are also shown. Furthermore, the study quantifies the effects of SM and surface roughness on GNSS-R deliverables.