Geophysical Doppler shifts measured by synthetic aperture radar (SAR) are influenced by the combination of ocean surface winds, waves, and currents. Accurate retrieval of radial surface current velocities therefore requires precise estimation and removal of the wave-induced Doppler contributions. To address this challenge, we developedWaveDop, a hybrid physics-guided and data-driven model based on the eXtreme Gradient Boosting (XGBoost) machine learning algorithm, designed to estimate Doppler shift contributions arising from wind and waves effects. The model was trained on a comprehensive dataset comprising Sentinel-1 SAR Doppler measurements acquired over global coastal regions from January 2015 to October 2022, in conjunction with WAVEWATCH III wave model outputs and drifter-observed surface currents. Our analysis reveals that wave-induced Doppler shifts are strongly modulated by the radar incidence angle, surface wind fields, and wave characteristics, with both wind waves and swell making significant contributions. By accounting for these wave effects, and correcting for non-geophysical artifacts such as antenna electronic mis-pointing, satellite attitude variations, and azimuthal scalloping, we derive radial current velocities from SAR observations. Validation against independent drifter measurements indicates strong agreement, with a correlation coefficient of 0.82, negligible bias, and a root-mean-square error (RMSE) of 0.16 m/s. These results demonstrate thatWaveDopeffectively estimates wave-induced Doppler shifts and enables reliable retrieval of ocean surface current velocities from SAR data.

This study examines the impact of wave breaking (WB) and Bragg scattering on the normalized radar cross section (NRCS) to reveal radar scattering features during high wind conditions. This is conducted by decomposing C-band quasi-synchronous wide-swath quad-polarization synthetic aperture radar (SAR) observations acquired from the RADARSAT Constellation Mission (RCM) and RADARSAT-2 (RS-2). The analysis results clearly demonstrate that the polarization difference (PD) associated with Bragg scattering saturates at high wind speeds, while still maintaining azimuthal modulation. Notably, the radar returns from breaking waves at cross-polarization (HV or VH) exhibit higher sensitivity to wind speeds but lower sensitivity to wind direction, compared to co-polarization (HH or VV). Moreover, our analyses show that WB contributes 40%, 80%, and 90% of VV-, HH-, and HV-polarized NRCS, respectively. This study highlights the unique capabilities provided by collocated RCM and RS-2 observations for investigating radar scattering features under high wind conditions. Results of this study can be further used to develop empirical models for estimating co- and cross-polarization radar backscatters induced by WB under high wind speeds.

The Doppler shift obtained from synthetic aperture radar (SAR) measurements comprises the combined contribution to the radial motion of the ocean surface induced by the sea state (wind waves and swell) and underlying surface currents. Hence, to obtain reliable estimates of the ocean surface current, the sea-state-induced Doppler shifts must be accurately estimated and eliminated. In this study, we use a semiempirical dual co-polarization Doppler velocity (DPDop) model, presented in the companion paper, to calculate sea-state-induced Doppler shifts using buoy-measured wind speed, wind direction, and ocean wave spectra. The DPDop model-simulated Doppler shifts are compared with the collocated Sentinel-1B SAR Wave (WV) mode observations at the 24° and 37° incidence angles, showing a bias of -0.24 Hz and a root-mean-square error (RMSE) of 5.55 Hz. This evaluation is also implemented on a simplified DPDop model at the same incidence angles. The model inputs include wind fields from the European Center for Medium-Range Weather Forecasts (ECMWF) and wave characteristic parameters (e.g., significant wave height, mean wave direction, and mean wavenumber) from WAVEWATCH III (WW3). The estimated Doppler shifts are validated using the ascending and descending observations of Sentinel-1B WV over the global ocean. Furthermore, the comparisons show that the bias and RMSE are -0.71 Hz and 9.25 Hz, respectively. Based on accurate wave bias correction, we obtain the radial current speeds of the ocean surface from the Doppler shift measurements. The estimated current speeds are compared with the collocated high-frequency radar measurements, with a bias of -0.04 m/s and an RMSE of 0.15 m/s. These results suggest that the original and simplified DPDop models can be used to estimate sea-state-induced Doppler shifts and, thus, derive accurate surface current retrievals.

This article proposes a Doppler velocity (DV) model based on dual co-polarized (co-pol) decomposition of a normalized radar cross section of an ocean surface on polarized Bragg scattering and nonpolarized (NP) radar returns from breaking wave components. The dual co-pol decomposition provides a quantitative description of resonant and NP scattering, as well as their dependence on the incident angle, azimuth, and wind speed. Subsequently, the contributions of the facet (resonant Bragg waves and breakers) velocities, tilt, and hydrodynamic modulations due to long waves to the resulting DV can be quantified. The tilt modulation contributions to DV are estimated using the measured/empirical tilt modulation transfer function (MTF). The hydrodynamic modulations are mostly dominated by wave breaking and are estimated using a semiempirical model based on in situ measurements. In addition to the VV and HH radar data, which are required for dual co-pol decomposition and tilt MTF estimates, the surface wave spectrum is required in the DV determination for a given radar observation geometry. In this article, qualitative and quantitative consistencies are presented between the model simulations and the empirical CDOP model. In a companion paper, a DV analysis is presented to analyze the Sentinel-1 synthetic aperture radar measurements and collocated in situ measurements of surface wind and wave spectra.

In this study, we use satellite and drifting-buoy synergistic observations to investigate the connections between tropical cyclone (TC) surface winds and currents, as well as their spatial distribution characteristics. The observed ageostrophic current speeds in ocean mixed layer (OML) increase linearly with wind speeds. The mean ratio (ageostrophic current speed to the wind speed) is around 2% in the left-front and left-rear quadrants with smaller variability, while it changes between 2% and 4% in the right-front and right-rear quadrants and shows much higher variations. Surface winds and currents both exhibit strong asymmetric features, showing larger wind speed and current velocity on the right side of the TC. The wind directions were approximately aligned with the current directions in the right-front quadrant, while a difference of about 90° was observed in the left-front and left-rear quadrants. The direction discrepency between wind and current in the right-rear quadrant was relatively small.

Under extreme weather conditions, the imprints of kilometer-scale marine atmospheric boundary layer roll vortices on the ocean surface are clearly visible in synthetic aperture radar (SAR) images of storms. Therefore, information about wind direction in storms can be obtained by analyzing SAR image features caused by boundary layer rolls. VH-polarized SAR imagery captures the structural features of storms well and shows prominent image gradients along the radial directions of the storm. The signal-to-noise ratios of VH-polarized images are small in low wind speed areas, but they are large in the same regions of VV-polarized images. Also, the capability of retrieving the atmospheric rolls orientation in VV-polarization is found to be sensitive to incidence angle, with better performances for larger incidence angles. Thus, there is the potential to retrieve the storm's wind directions using a combination of the VH- and VV-polarized SAR observations. In this article, we use the local gradient method to estimate tropical cyclone (TC) wind directions from C-band RADARSAT-2 and Sentinel-1A dual-polarization (VV + VH) SAR imagery. As a case study, wind directions with a spatial resolution of 25 km are derived by using both wide-swath VV- and VH-polarized SAR imagery over two hurricanes (Earl and Bertha) and one Typhoon (Meranti). We compare wind directions derived from ten dual-polarization SAR images with collocated wind directions from buoys, Global Positioning System (GPS) dropsondes, scatterometer, and radiometer. Statistical comparisons show that the wind direction bias and root-mean-square error are, respectively, -0.54° and 14.78° for VV-polarization, 0.38° and 14.25° for VH-polarization, 0.20° and 13.30° for VV- and VH-polarization, suggesting dual-polarization SAR is more suitable for the estimation of TC wind directions than VV- or VH-polarization SAR.

Co- and cross-polarized geophysical model function (GMFs) both have limitations to retrieve ocean surface wind speeds. The former underestimates high winds and the latter overstimates low winds. In this study, we both use co- and cross-polarized SAR observations to retrieve hurricane wind speeds and directions. The local gradient method is first used to estimate hurricane wind directions from C-band RADARSAT-2 and Sentinel-1A dual-polarization SAR imagery. We compare wind directions derived from 10 dual-polarization SAR images with collocated wind directions from buoys, GPS dropsondes, scatterometer, radiometer, and Hurricane Research Division Real-time Hurricane Wind Analysis System (H*Wind) data. Statistical comparisons show that the wind direction bias and root-mean-square error are 3.22° and 21.53°, respectively. Subsequently, as a case study for Hurricane Maria, the retrieved wind directions, along with co- and cross-polarized GMFs are used to derive wind speeds. The retrieved wind speeds are validated with collocated Stepped-Frequency Microwave Radiometer (SFMR) measurements. The results show that the combination of co- and cross-polarized GMFs has capability of obtaining more reasonable hurricane wind speeds and directions than by only using co- or cross-pol GMF.

RADARSAT-2 C-band quad-polarization ocean synthetic aperture radar (SAR) scenes are decomposed into resonant Bragg scattering from regular (no-breaking) surface and scattering from breaking waves. Analysis of the surface current signatures in dual co- and cross-pol SAR images revealed that governing imaging mechanism is modulations of wave breakings which are very sensitive to the presence of current non-uniformities. As found, due to small relaxation scale, short Bragg waves do not "feel" the current. Thus routinely observed current signatures in quad-pol SAR images originate essentially from wave breaking modulations, and modulation of Bragg waves does not matter this issue.

This study provides quantitative estimates of wind waves breaking contribution to dual co- and cross-polarized radar scattering. Extending approaches suggested by [1], [2], 1696 RADARSAT-2 quad-polarized Synthetic Aperture Radar measurements collocated with 65 in situ National Data Buoy Center buoys observations are analyzed. For VV polarization, wave breaking contribution decreases from 60% to 20% with increasing incidence angle, while for HH polarization and cross-polarization wave breaking contribution is about 60%-70% at all incidence angles. Empirical dependencies of wave breaking contributions to co- and cross-pol signals on wind speed, incidence angles and azimuth are then suggested.

This paper provides improved quantitative estimates of the wind-ruffled roughness contributions to dual co- and cross-polarized radar signals. Expanding previous approaches, 1696 RADARSAT-2 quad-polarized synthetic aperture radar (SAR) measurements, co-located with 65 in situ National Data Buoy Center (NDBC) buoy observations, are analyzed. Considering all wind conditions, the impact of breaking and near-breaking waves on dual co- and cross-polarized radar signals is robustly documented. For VV polarized measurements, the contribution of breaking waves decreased from 60% to 20% with increasing incidence angle, whereas for HH polarization and cross-polarization measurements, it can amount to about 60%-70% for all incidence angles. Building on the large analyzed data set, robust empirical dependencies between breaking waves and their impact on co- and cross-pol signals are then derived, as functions of wind speeds, incidence angles, and azimuth directions.