Sea Surface Current Fields Research Articles

Prediction of Sea Surface Current Around the Korean Peninsula Using Artificial Neural Networks

AbstractThe prediction of sea surface currents is essential for various marine activities such as disaster monitoring, fishing industries, and search and rescue operations. Continuous improvements in numerical models have made it possible to predict more realistic oceans using data assimilation and fine spatial resolution. However, such complicated ocean models require high computational power and time, making them difficult to use for near‐real time forecasting. To compensate for these high computational costs, novel approaches with efficient computational costs combined with numerical model outputs need to be developed. Artificial neural networks could be one of the solutions as they require low computational power for prediction, owing to pre‐trained networks. Here, we present a prediction framework applicable to surface current prediction in the seas around the Korean Peninsula using three‐dimensional (3‐D) convolutional neural networks. The network is based on a 3‐D U‐shaped network structure and is modified to predict sea surface currents using oceanic and atmospheric variables. The forecast procedure is optimized to minimize the error of the next day's sea surface current field with a spatial resolution of 1/24° and its recursively predicting structure allows more days to be predicted. The network performance is evaluated by changing the input days and variables to find the optimal surface‐current‐prediction artificial neural network model. The optimized model shows a root mean squared error of 2.3 cm/s for the first forecast day, demonstrating its strong potential for practical use in the near future.

Sea surface circulation in the Baltic Sea: decomposed components and pattern recognition

By decomposing the total sea surface current into its geostrophic and ageostrophic components, we examined the contribution of each to the long-term variability of the total sea surface current. Our findings demonstrate the importance of geostrophic currents in Baltic Sea gyre formations. Additionally, ageostrophic currents contribute significantly to the flow across the region. Quantifying the difference between total sea surface current fields has revealed two dominant general sea surface circulation patterns in the Baltic Sea, whose characteristics correspond to the monthly mean climatology of sea surface current fields in May and December. Subsequently, a machine learning technique was employed to effectively detect the types of sea surface circulation patterns using wind vectors and sea level anomaly fields. This underscored the combined influence of sea level anomaly-driven and wind-driven components in shaping surface current vectors in the Baltic Sea, consistent with geostrophic and ageostrophic decompositions.

New diagnostic sea surface current fields to trace floating algae in the Yellow Sea

The new velocity fields based on the Generalized Ekman (GE) theory to trace floating algae were derived and verified by drifter observations and compared to reanalysis datasets in the Yellow Sea (YS). Two velocity fields using diagnostic approaches and two velocity fields from reanalysis datasets were examined. The results revealed that the diagnostic velocity fields had comparable accuracy to the reanalysis datasets, even locally better. Then, we applied each velocity field to trace green algae, Ulva prolifera, in July 2011 and brown algae, Sargassum horneri, in May 2017 using particle tracking experiments. In addition, drifter trajectories were simulated, and error accumulation speed was estimated for each velocity field. Simulation results using the diagnostic velocity fields consistently showed better agreement with satellite images and in situ observations than those using reanalysis datasets, demonstrating that the diagnostic velocity could be a superior tool for simulating surface-floating substances and organisms. The approach to derive diagnostic velocity fields can be easily applied instead of relying on heavy computing numerical models.

Extrapolation of electromagnetic pointing error corrections for Sentinel-1 Doppler currents from land areas to the open ocean

By utilizing single-look complex data from synthetic aperture radar (SAR), the Doppler centroid anomaly method (DCA) can be employed to determine the velocity of the sea surface current field along the radar line-of-sight direction. However, due to the influence of platform or antenna perturbations, SAR Doppler shift measurements are subject to electromagnetic pointing error (EPE). These errors can typically be reduced by using echo information from land areas (including islands) within the SAR images to correct the Doppler shift. This allows the successful estimation of Doppler shifts due to sea surface factors and, in turn, the retrieval of ocean currents in the radar line-of-sight. However, in cases where SAR images lack data over land, uncorrectable EPE can result in significant inaccuracies in the retrieval of ocean currents. This study utilizes data captured by Sentinel-1A on four orbits in the Florida Strait region of the Gulf Stream from July 2020 to October 2022 to investigate the variation characteristics of EPE. To address the issues of calculating these in SAR images lacking land data, an EPE fitting method is proposed. The correction significantly reduces the error of S1A retrieved current velocity compared with line-of-sight current velocity from the HYCOM model and in situ HF radar, indicating that the proposed method provides an effective way to reduce the impact of EPE in SAR images, even when land is not directly observed.

Performance analysis of global HYCOM flow field using Argo profiles

ABSTRACT Flow field data generated by ocean models are important for simulating ocean currents and circulation patterns, which are essential components in digital Earth construction. To evaluate the accuracy of model-simulated flow fields, Array for Real-time Geostrophic Oceanography (Argo) float observations can be considered benchmarks. In this study, a novel method for comparing Argo profiles with 3-dimensional trajectories obtained by simulating Argo floats in Hybrid Coordinate Ocean Model (HYCOM)-provided flow fields was proposed. Surface and subsurface trajectories were calculated, and their spatial matching characteristics were analyzed. The results demonstrated that (1) the HYCOM surface and subsurface flow fields generally conform to the basic characteristics and trends of ocean currents; (2) the HYCOM sea surface current field error pattern exhibits a symmetrical distribution centered on the equator in the Northern and Southern Hemispheres and increases with increasing latitude; and (3) the HYCOM subsurface flow field exhibits regional differences, with the largest differences in the Gulf Stream, North Atlantic Warm Current, and Westerly Wind Drift region. Through analysis of the disparities between HYCOM and Argo data, the effectiveness of using model simulation data can be enhanced, and the accuracy and dependability of ocean models can be improved.

The Ocean Surface Current in the East China Sea Computed by the Geostationary Ocean Color Imager Satellite

High-frequency observations of surface current field data over large areas and long time series are imperative for comprehending sea-air interaction and ocean dynamics. Nonetheless, neither in situ observations nor polar-orbiting satellites can fulfill the requirements necessary for such observations. In recent years, geostationary satellite data with ultra-high temporal resolution have been increasingly utilized for the computation of surface flow fields. In this paper, the surface flow field in the East China Sea is estimated using maximum cross-correlation, which is the most widely used flow field computation algorithm, based on the total suspended solids (TSS) data acquired from the Geostationary Ocean Color Imager satellite. The inversion results were compared with the modeled tidal current data and the measured tidal elevation data for verification. The results of the verification demonstrated that the mean deviation of the long semiaxis of the tidal ellipse of the inverted M2 tide is 0.0335 m/s, the mean deviation of the short semiaxis is 0.0276 m/s, and the mean deviation of the tilt angle is 6.89°. Moreover, the spatially averaged flow velocity corresponds with the observed pattern of tidal elevation changes, thus showcasing the field’s significant reliability. Afterward, we calculated the sea surface current fields in the East China Sea for the years 2013 to 2019 and created distribution maps for both climatology and seasonality. The resulting current charts provide an intuitive display of the spatial structure and seasonal variations in the East China Sea circulation. Lastly, we performed a diagnostic analysis on the surface TSS variation mechanism in the frontal zone along the Zhejiang coast, utilizing inverted flow data collected on 3 August 2013, which had a high spatial coverage and complete time series. Our analysis revealed that the intraday variation in TSS in the local surface layer was primarily influenced by tide-induced vertical mixing. The research findings of this article not only provide valuable data support for the study of local ocean dynamics but also verify the reliability of short-period surface flow inversion of high-turbidity waters near the coast using geostationary satellites.

Effects of Wave-Induced Doppler Velocity on the Sea Surface Current Measurements by Ka-Band Real-Aperture Radar with Small Incidence Angle

The Doppler shift of microwave radar sea surface echoes serves as the foundation for sea surface current field retrieval; it includes the shift caused by satellite platform motion, ocean waves, and sea surface currents. The Doppler shift caused by ocean waves is known as the wave-induced Doppler velocity (UWD), and its removal is critical for the accurate retrieval of sea surface current fields. The low-incidence Ka-band real-aperture radar rotary scan regime has the capability of directly observing wide-swath two-dimensional current fields, but as a new regime to be further explored and validated, simulation and analysis of UWD in this regime have a significant influence on the hardware design and currently observed applications of this satellite system in its conceptual stage. In this study, we simulated and investigated the impacts of radar parameters and sea-state conditions on the UWD obtained from small-incidence-angle Ka-band rotational scanning radar data and verified the simulation results with the classical analytical solution of average specular scattering point velocity. Simulation results indicate that the change in the azimuth direction of platform observation affects UWD accuracy. Accuracy is the lowest when the antenna is in a vertical side-view. The UWD increases slowly with the incidence angle. Ocean waves are insensitive to polarization in the case of small-incidence-angle specular scattering. The increase in wind speed and the development of wind waves result in a substantial increase in UWD. We classified swell by wavelength and wave height and found that UWD increases with swell size, especially the contribution of swell height to UWD, which is more significant. The contribution of the swell to UWD is smaller than that of wind waves to UWD. Furthermore, the existence of sea surface currents changes the contribution of ocean waves to UWD, and the contribution weakens with increasing wind speed and increases with wind wave development.

High-Resolution Sea Surface Current Vector Field Inversion Based on Circular Scanning SAR

High-Resolution Sea Surface Current Vector Field Inversion Based on Circular Scanning SAR

Study on the Elimination Method of Wind Field Influence in Retrieving a Sea Surface Current Field.

An along-the-track interferometric synthetic aperture radar (ATI-SAR) system can estimate the radial velocity of a moving target on the ground and on a sea surface current. This acquires the interference phase by combining two composite SAR images obtained by two antennas spatially separated along the direction of movement of the platform. The key to retrieving the sea surface current is to remove the interference of sea surface waves, wind-generated current, and Bragg phase velocity in the interference Doppler velocity. Previous methods removed the surface waves, Bragg phase velocity, and other interferences based on externally-assisted wind fields (e.g., ECMWF), using the M4S or other models. However, the wind fields obtained from ECMWF and other external information are often average results of a large temporal and spatial scale, while the images obtained from SAR are high-resolution images of sea surface transients, which are quite different in time and space. This paper takes the SAR image data of the Gaofen-3 satellite as the research object and employs an SAR-based wind field retrieval method to obtain an SAR-observed transient wind field. Combined with the CDOP model, the interference of Doppler velocities, such as the sea surface wave, wind-generated current, and Bragg wave phase velocity, was calculated and subtracted from the Doppler velocity, to obtain the sea surface velocity result. Then, the current field measured by the shore-based HF radar was compared with that obtained by correcting the ATI Doppler velocity based on the SAR retrieved wind field and the ECMWF wind field. The comparison of results indicated that the wind field correction result based on the SAR retrieved wind field was closer to the current field measured by the shore-based HF radar than the wind field correction result based on the ECMWF wind field.

Satellite Observation of the Long-Term Dynamics of Particulate Organic Carbon in the East China Sea Based on a Hybrid Algorithm

The distribution pattern and flux variation of POC in the continental shelf seas are essential for understanding the carbon cycle in marginal seas. The hydrodynamic environment and complicated estuarine processes in the East China Sea result in challenging estimates and substantial spatio-temporal variability in terms of POC concentrations. A hybrid retrieval model based on the mutual combination of the color index algorithm (CIPOC) and the empirical band ratio algorithm was applied in this study to effectively and dynamically monitor the surface POC concentration in the East China Sea in a long-term series for the first time using MODIS/Aqua remote sensing satellite data from 2003 to 2020. A hybrid retrieval model based on the mutual combination of the color index algorithm (CIPOC) and the empirical band ratio algorithm was applied in this study. The MODIS/Aqua remote sensing satellite data from 2003 to 2020 were employed for the first time to dynamically monitor the surface POC concentrations in the East China Sea for a long time series. The results demonstrated that the performance (R2 = 0.84, RMSE = 156.14 mg/m3, MAPE = 43.30%, bias = −64.79 mg/m3) exhibited by this hybrid retrieval algorithm confirms the usability of inversion studies of surface POC in the East China Sea. Different drivers such as river discharge, phytoplankton, wind, and the sea surface current field jointly influence the spatial and temporal distribution of POC concentrations in the East China Sea. This paper also verifies that the hybrid algorithm can be applied to retrieval tasks for POC in different seas with similar optical properties to the waters of the East China Sea. In conclusion, the long-term series East China Sea POC data record, which was established based on MODIS/Aqua, provides supplementary information for in-situ sampling, which will aid the long-term monitoring of POC fluxes in shelf seas. At the same time, it has also improved our understanding of the transport and spatio-temporal variability of POC in the East China Sea, enhancing our comprehension of the impact of POC on environmental changes and carbon cycling in marginal seas.

Direct Comparison of Sea Surface Velocity Estimated From Sentinel-1 and TanDEM-X SAR Data

This article presents the first direct comparison of the sea surface radial velocity (RVL) derived from the two satellite SAR systems Sentinel-1 and TanDEM-X, operating at different frequencies and imaging modes. The RVL is derived from the Doppler centroid (Dc) provided in the Sentinel-1 OCN product and from the along-track interferometric phase of the TanDEM-X. The comparison is carried out using unique opportunistic acquisitions, collocated in space and time, over three different sites located in the Iceland Sea, the Pentland Firth, and the Kattegat Sea. First, it is observed that the RVL derived from both satellites is biased, thus calibration is applied using the land as a reference. The comparison shows that the correlation and the mean bias between the two datasets depend on the differences in acquisition time, incidence angle, and azimuth angle, and on wind and surface velocities. It is found that, given a time difference of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\lesssim$</tex-math></inline-formula> 20 min, the spatial correlation coefficient is relatively high (between 0.70 and 0.93), which indicates that the two SAR systems observe similar sea surface current fields. The spatial correlation degrades primarily due to increasing time difference and decreasing velocity magnitudes. It is also found that the mean RVL bias increases primarily with the radial wind speed, which suggests that the bias is mainly due to the wave-induced Doppler shift. This article shows that under certain conditions, i.e., similar acquisition geometry and short time delay, a good agreement between the two independently derived RVL is achieved, both in the spatial variation and absolute mean value. This encourages a synergistic use of the sea surface velocity estimated from different C- and X-band SAR systems.

Observing Sea Surface Current by Gaofen-3 Satellite Along-Track Interferometric SAR Experimental Mode

Sea surface current is a research hotspot in oceanography. Space-borne along-track interferometric synthetic aperture radar (along-track InSAR, ATI) is a promising sensor for measuring high-resolution sea surface current field, and there is no operational system in orbit yet. To support future space-borne ATI systems, based on the ATI experimental mode of the Gaofen-3 (GF-3) satellite, the first sea surface current observing experiment was conducted in the Jiaozhou Gulf in China, in 2019. Meanwhile, SAR observations and in situ instrument measurements of the current are obtained in the experiments. The data is first preprocessed by a processor specially developed for the GF-3 ATI data. Then, the current is extracted based on the M4S mode. The retrieved current of the Jiaozhou Gulf is compared with ground-based high frequency surface wave radar data. The results show that the root mean square error of the surface current observed by the GF-3 satellite is less than 0.2 m/s.

Numerical simulation of the wave-induced Stokes drift effect on sea surface temperature in the North Pacific

Abstract The effect of the wave-induced Stokes drift is not taken into account in traditional ocean circulation models used for SST simulations. The spectral parameterization scheme is considered to be the most accurate of the wave-induced Stokes drift calculation schemes. The numerical simulation results of sea surface temperature (SST) with the Stokes drift and SST without the Stokes drift in the North Pacific in 2014 were analyzed. The Stokes drift plays a cooling role in the North Pacific, and the most affected areas are high-latitude sea areas. The following factors are responsible for cooling: the seawater divergence caused by Stokes transport, changes in the sea surface current field caused by the Coriolis-Stokes force and the effects of turbulence caused by Langmuir circulation. The simulation of the vertical temperature profile in the mixed layer is improved when the Stokes drift is accounted for. The simulated results of SST using the Stokes drift approximate parameterization schemes and the spectral parameterization scheme are compared. The results confirm that the spectral parameterization scheme can be used for accurate SST simulation, and the Phillips spectrum approximate parameterization scheme is the best among the approximate parameterization schemes.

Application of X-Band Wave Radar for Coastal Dynamic Analysis: Case Test of Bagnara Calabra (South Tyrrhenian Sea, Italy)

Sea state knowledge has a key role in evaluation of coastal erosion, the assessment of vulnerability and potential in coastal zone utilization, and development of numerical models to predict its evolution. X-band radar measurements were conducted to observe the spatial and temporal variation of the sea-state parameters along a 3 km long sandy-gravelly pocket beaches forming a littoral cell on Bagnara Calabra. We produced a sequence of 1000 images of the sea state extending offshore up to 1 mile. The survey has allowed monitoring the coastline, the directional wave spectra, the sea surface current fields, and the significant wave heights and detecting strong rip currents which cause scours around the open inlets and affect the stability of the submerged reef-type breakwaters. The possibility to validate the data acquired with other datasets (e.g., LaMMA Consortium) demonstrates the potential of the X-band radar technology as a monitoring tool to advance the understanding of the linkages between sea conditions, nearshore sediment dynamics, and coastal change. This work proves the possibility to obtain relevant information (e.g., wave number, period, and direction) for evaluation of local erosion phenomena and of morphological changes in the nearshore and surf zone.

Tsushima Warm Current paths in the southwestern part of the Japan Sea

We compiled a long-term record of monthly mean sea surface currents in the Japan Sea from satellite altimetry and satellite-tracked drifter data. The record covers the entire Japan Sea for the period from May 1995 through January 2009. Using these data, we examined the path of the Tsushima Warm Current in the southwestern part of the Japan Sea. The mean sea surface current field revealed a “triple-branch” pattern. The first branch through the eastern channel of the Tsushima Strait flows along the coast of Japan with a mean current velocity of 20cms−1. The second branch flows along the continental shelf edge at around the 200-m isobath with a mean current velocity of 12cms−1. The third branch flows along the east coast of the Korean peninsula with a mean current velocity of 25cms−1. We investigated seasonal variation in the path of each branch. The first branch flowing along the Japanese coast throughout the year is highly stable. The second branch is stable throughout the year except in November. The third branch flowing along the east coast of Korea is highly stable until 38°N. Downstream of the third branch around 38°N, the current bifurcates into northeastward and southeastward currents. The northeastward current exhibits stable flow from June to February except in January. This current joins an eastward current originating around 39°N, 128.5°E. The southeastward current flows stably in June, September, and December. There were two current paths east of 131°E over the continental shelf off the San’in coast of Japan. The nearshore current flows along the Japanese coast with a mean current velocity of 23cms−1, and the offshore current flows around the continental shelf area with a mean current velocity of 19cms−1. These two currents are highly stable throughout the year. A current flows to the west of Oki Island to around 38.5°N and is highly stable in all months except March to July. These results provide a schematic view of the paths of the Tsushima Warm Current in the southwestern part of the Japan Sea.

Imaging mesoscale upper ocean dynamics using synthetic aperture radar and optical data

A synergetic approach for quantitative analysis of high‐resolution ocean synthetic aperture radar (SAR) and imaging spectrometer data, including the infrared (IR) channels, is suggested. This approach first clearly demonstrates that sea surface roughness anomalies derived from Sun glitter imagery compare very well to SAR roughness anomalies. As further revealed using these fine‐resolution (∼1 km) observations, the derived roughness anomaly fields are spatially correlated with sharp gradients of the sea surface temperature (SST) field. To quantitatively interpret SAR and optical (in visible and IR ranges) images, equations are derived to relate the “surface roughness” signatures to the upper ocean flow characteristics. As developed, a direct link between surface observations and divergence of the sea surface current field is anticipated. From these satellite observations, intense cross‐frontal dynamics and vertical motions are then found to occur near sharp horizontal gradients of the SST field. As a plausible mechanism, it is suggested that interactions of the wind‐driven upper layer with the quasi‐geostrophic current field (via Ekman advective and mixing mechanisms) result in the generation of secondary ageostrophic circulation, producing convergence and divergence of the surface currents. The proposed synergetic approach combining SST, Sun glitter brightness, and radar backscatter anomalies, possibly augmented by other satellite data (e.g., altimetry, scatterometry, ocean color), can thus provide consistent and quantitative determination of the location and intensity of the surface current convergence/divergence (upwelling/downwelling). This, in turn, establishes an important step toward advances in the quantitative interpretation of the upper ocean dynamics from their two‐dimensional satellite surface expressions.

Multi-scale feature tracking in sequential satellite images with wavelet analysis to measure sea surface currents in the Gulf of St. Lawrence

Measuring sea surface currents is a technological challenge in oceanography. Feature tracking in time series of remote sensing imagery has been proposed as a way to address this problem. The most commonly used approach is the maximum cross-correlation (MCC) method, originally developed to track cloud motion. We propose a new technique that makes use of Daubechies wavelet analysis combined with the MCC method. In our approach, satellite images are decomposed into various spatial scales using the wavelet transform, and the location with the MCC coefficient among all the scales is selected as the most likely new position of the tracked feature. Results from the analysis of five pairs of sequential National Oceanic and Atmospheric Administration (NOAA) advanced very high resolution radiometer (AVHRR) images of the Gulf of St. Lawrence area show that wavelet analysis improves the estimated sea surface current field by increasing the number of current vectors about 20% under the same confidence level (0.9) as compared with that using the MCC method alone.

Transport of larval jack mackerel (Trachurus japonicus) estimated from trajectories of satellite‐tracked drifters and advective velocity fields obtained from sequential satellite thermal images in the eastern East China Sea

AbstractTransport processes of jack mackerel (Trachurus japonicus) larvae in the waters off the west coast of Kyushu in the eastern East China Sea, have been investigated using satellite‐tracked surface drifters and consecutive satellite thermal images. Trajectories of drifters describe northward flows over the continental shelf, eastward flows of the Kuroshio south‐west of Kyushu, and a weak clockwise gyre off the west coast of Kyushu. In particular, the clockwise gyre causes the entrainment of jack mackerel larvae into the waters off the west coast of Kyushu. Consecutive satellite thermal images help to elucidate the northward warm water intrusion from the Kuroshio front south‐west of Kyushu. Particle trajectories using sea surface current fields computed with the Maximum Cross Correlation (MCC) technique also reveal that the transport of jack mackerel larvae into the nursery ground off the west coast of Kyushu caused by the anti‐cyclonic gyre and the warm streamers are an important process for successful recruitment.

131 短波海洋レーダによる海面流速分布の測定

In order to elucidate the maintenance mechamism of abrupt large change of temperature and salinity across the Kuroshio front, measurements off sea surface current field and its temporal change were made in the sea south of the Yakushima Island from October 9 to December 7, 1992, using the HF ocean Radar system, developed at the Okinawa Radio Wave Observatory, CRL. From 10-minute mean line-of-sight component of currents measured by two independently scanning HF Radars located 10 km apart to each other at the coast of Yakushima Island, we have estimated horizontal current. vectors. hourly at grided points of 2 km interval in 50 × 20 km area. The horizontal divergence was found to have positive and negative maximum values alternately along the front rather than a narrow convergence zone along the front.

Evolution of the climatological near‐surface thermal structure of the tropical Indian Ocean: 1. Description of mean monthly mixed layer depth, and sea surface temperature, surface current, and surface meteorological fields

Mean monthly mixed layer depth (MLD), sea surface temperature (SST), and surface current climatologies are generated for the tropical Indian Ocean. In addition, surface meteorological climatologies are produced for those variables which could influence the evolution of the MLD and SST fields. Only the MLD climatology is described in detail, as climatologies for the other variables have appeared previously in the literature. The sum of the annual and semiannual harmonics account for greater than 75% of the energy in the MLD time series over most of the basin. The amplitude of the annual signal is greater than 20 m between 10°S and 25°S, with deepest MLDs observed during the southern hemisphere winter. The south central Arabian Sea, between the equator and 10°N, and the northern Arabian Sea are also regions of larger annual harmonic amplitude (&gt;15 m). The amplitude of the semiannual harmonic is largest in the central Arabian Sea (&gt;25 m). Deepest MLDs are observed there during the height of the two monsoon seasons. Correlation coefficients are computed between MLD and SST and several other oceanographic and meteorological variables to explore possible causal relationships. Net energy flux through the sea surface can account for 75% of the variance in the SST and MLD time series over most of the region south of the equator. Large coefficients are also observed in the northwestern Arabian Sea. Correlations between SST and MLD and surface currents are in general small throughout the region, with maxima observed in the central Arabian Sea, in the vicinity of the South Equatorial Current and in the extreme eastern equatorial Indian Ocean. These correlations will be examined in more detail in part 2 of this study in which simple models of mixed layer dynamics are employed.