Long and energetic surface waves radiating from distant storms may eventually cover a full ocean basin with a lifetime that can extend over a few weeks. Chances are that these swell fields will propagate over regions characterized by strong ambiant upper ocean currents. Resulting interactions can then trigger sea state variability possibly associated with rogue or freak waves. Theoretical frameworks can build on ordinary differential equations describing the kinematics and dynamics of wave train evolution in the presence of surface currents, explaining the cumulative impact of the current vorticity field on wave train kinematics. Ultimately, this can lead to significant ray deflection. However, accurate modeling and forecasting of sea state variability triggered by surface currents remains a challenge.

From a numerical modeling perspective using state of the art wave and ocean models , a recent study suggested that much of the modeled sea state variability, at scales < 80 km, is triggered by currents. However, comparisons with satellite altimeter measurements were not conclusive because the mesoscale geophysical variability is strongly buried within altimeter measurement noise at scales up to 100 km. To avoid such limitations and better exploit satellite altimeter capability to provide more robust estimates of the local significant wave height and sea surface height meso-scale variability, a novel approach is considered in the present study to filter and analyze altimeter measurements. It relies on the Empirical Mode Decomposition (EMD). Developed to cope with both nonlinear and non-stationary processes, EMD often overcomes limitations of other approaches, such as the Fourier and/or wavelet-based analysis. Here, this approach is extended to global satellite altimeter measurements, to further highlight striking evidences of large and persistent Hs variability triggered by surface currents. This self-consistent space-scale decomposition helps to map, over the global ocean, overlooked statistical properties at scales less than 100 km, where meso- and sub-mesoscale upper ocean circulation drives a significant part of the coupled ocean/atmosphere system variability.

This is further analyzed in the Agulhas current region where a dominant regime of incoming southwesterly waves imprints the local wave climatology. Persistent patterns clearly result from the cumulative impacts of the large-scale and persistent nature of the current vorticity field on wave train kinematics. Because of the significant ray deflection and trapping phenomena, localized sea state gradients can be anticipated and are well traced in altimeter signals. The ray tracing analysis performed over a three-year time period explains the link between observed climatological sea state anomaly patterns and the Agulhas current patterns.

The basin scale and synoptic monitoring of the marine currents in the Mediterranean Sea has geophysical, commercial and environmental purposes. At present, this monitoring is achievable via satellite altimetry, allowing to retrieve the large scale ( >= 100 km ) geostrophic component of the surface currents. Nevertheless, the Mediterranean is characterized by Rossby deformation radii around 10-20 km, hence, its typical mesoscale features (O(10-100 km)) are only partially captured by classical satellite altimetry. Moreover, this region (and in particular the Sicily Channel) represents the main ship route between the Indian Ocean and the main harbours of the European Union, hosting 25% of the world oil trade, with a consequent presence of illegal oil spills of the order of 600 Kt per year. This makes high-resolution surface currents a necessary tool for monitoring the Mediterranean Sea and improving the knowledge of its surface dynamics. To this scope, we present a method for the remote retrieval of the sea-surface currents based on the synergy of sea-surface height (SSH) and high-resolution sea-surface temperature (SST) data. During the period 2012-2016, we created daily gap-free high-resolution maps of surface currents at the basin scale. The quality of the new multi-sensor currents has been assessed through comparisons to other surface-currents estimates, as the ones obtained from drifting buoys (at the basin scale), HF-Radar platforms and ocean numerical model outputs in the Malta-Sicily Channel. The study evidenced that our synergetic approach can improve the present-day derivation of the surface currents in the Mediterranean basin, showing satisfying performances even in the reconstruction of the ageostrophic circulation. In the Malta-Sicily Channel, assuming the HF-Radar estimations as a reference, the merged SSH/SST currents exhibit a smaller root mean square error and bias than the altimeter-derived ones, mainly due to their enhanced spatial and temporal resolution. The improvements are also found with respect to the model-derived currents. The main perspectives of this work are to implement the reconstruction method using hourly SST data from geostationary satellites or even using the remotely sensed surface chlorophyll concentrations. This will bring the remote surface currents retrieval to even higher spatial-temporal resolutions.

The ocean circulation is responsible for a significant transport of heat and salt, and play a major role for climate and weather. Knowledge of surface currents is also crucial for oil spill and marine debris tracking, search and rescue operations, egg and larvae drift, and fisheries. Surface currents also affect the steepness of surface waves, and as such also highly needed for provision of reliable sea state forecasts for maritime operations and ship routing. Consequently, precise estimation and monitoring of the ocean surface currents is therefore essential for marine research and operations, especially in the coastal zones where most of the human activities are located.

Conventional observation methods include scattered in-situ point measurements from sensors located on buoys or moorings, and lagrangian surface drifters. Moreover, land-based remote sensing systems, such as High-Frequency radars (HFR) have also been proved to be an indispensable source for spatial mapping of ocean currents in the near-coastal zone. Unfortunately, pour spatial coverage and high deployment cost for the in-situ sensors, as well as limited spatial coverage of HFR (maximum up to 220 km from the sensor), along with other limitations disable systematic spatial monitoring of the surface currents.

High-resolution (~1x1 km) Doppler frequency shift observations from Sentinel-1 A/B Interferometric Wide (IW) mode is available on a regular basis (up to two times per day) in the coastal zone. Hence, the corresponding retrieval of the surface currents from SAR acquisitions can be used for filling observation gaps in areas not covered by by in-situ and land-based remote sensing measurements. The SAR measured radial velocities could eventually also be assimilated in numerical ocean models. However, a careful validation of the SAR derived surface current estimates through systematic collocation with other remote sensing and in-situ datasets is therefore required.

In this study results of the evaluation of these range surface current velocities derived from Sentinel-1A IW VV Level 2 Doppler frequency shifts acquired over the northern coast of Norway are presented and compared to HFR (CODAR SeaSonde) and lagrangian ocean surface drifters (CATHE) observations. In addition, the data are collocated with the 8-day average Sea Surface Temperature (SST) fields from MODIS-Aqua L3 products.


The geophysical range Doppler frequency shift has been calibrated with using land coverage within each scene. The mean bias (for all scenes) over the land after all calibrations is 0.74 Hz (0.04 m/s) while the root mean square error (RMSE) is about 5 Hz, in all sub-swaths. The wind/wave bias contribution to the Doppler frequency shift was estimated using the CDOP empirical geophysical model function. Near-surface wind speed and direction from the AROME-Arctic model (operated by MET.NO) was used as input to the CDOP.

Analysis of the scenes after the removal of the wind/wave bias revealed that significant signatures of the Norwegian Coastal Current (NCC) can be observed in most part of investigated SAR acquisitions. These features are also visible in the 8-days average SST maps.

Comparison of the SAR derived radial velocities to the CODAR SeaSonde observations ranging up to 175 km from shore shows significant correlation with mean bias about 0.04 m/s (equivalent to the mean bias over the land) and RMSE 0.37 m/s. In addition, also the collocation of the SAR radial velocities with the ocean surface velocities obtained from drifters shows good correlation. However, accurate retrieval of the surface current velocity from a single SAR image still has limitations due to the residual bias in Doppler centroid calibration and uncertainty due to the wind/wave bias contribution to the acquired signal. The latter is most likely arising from inaccurate model estimates of the wind speed and direction. The preliminary results suggest that the best agreement are derived for moderate winds in the range direction.


Acknowledgments: This work is performed under NERSC - 251348/F50 and CIRFA – 970422528 projects funded by Norwegian Research Council. We would like to thank the Norwegian Meteorological Institute for providing HFR, surface drifters and meteorological observations as well as AROME-Arctic data.

Through a joint effort between the Italian National Research Center (ISAC-CNR), the French National Space Agency (CNES) and the European Space Agency (ESA), a multiyear dataset of surface velocities have been calculated over the global ocean by combining altimeter derived geostrophic velocities and Sea Surface Temperature (SST) data. The method is based on the inversion for the velocity of the heat conservation equation using the altimeter velocities as background.
By accurately prescribing the error both on the background velocities and on the forcing term (heat fluxes), the surface velocities are successfully improved in areas characterized by strong SST gradients and remain unchanged in low SST gradient regions, where by construction no additional information may be brought by the SST field. This allows us to provide for the first time a global field of surface velocities from the merging of altimetry and SST.

This method is used to compute daily merged velocities from 1997 to 2017 using REMote Sensing System (REMSS) Microwave SST as input forcing. We use the version 5 MW SSTwhich is a low resolution product (25km).
Validation is performed on through comparison to in-situ drifting buoy velocities over the 1997 -2017 period. Major improvements (up to 15% locally) are obtained on the meridian component of the velocity and in the equatorial band. Both the spatial and temporal resolution of the altimeter derived velocities is enhanced by the SST information. Due to the large uncertainty on the heat fluxes and the spatio-temporal resolution of the input datasets, the high frequency ageostrophic components of the circulation is hardly resolved by the method and, outside the equatorial band, improvements are mainly on the geostrophic component. Moreover, some negative impacts of the method can be seen at high latitudes.
New forcing and background error estimates allow to partially resolve this problem and the use of higher resolution SST products (MWIR, MW + InfraRed data) also show some new perspectives to enhance the merge product resolution and accuracy.

This dataset is planned to be distributed in the near future through the CMEMS (Copernicus Marine and Environment Monitoring Services) MULTI OSERVATION (MULTIOBS) Thematic Assembly Center (TAC).

At scales between 1 km and 20 km, the ocean surface circulation interacts strongly with the atmosphere and exhibits surface current processes that can have relative vorticity and divergence (which vanishes for geostrophic flows) much greater than the Coriolis frequency. Investigation of these processes has been limited by the difficulty in obtaining vector surface current and wind data from which relative vorticity and divergence can be calculated. Limited data has been collected using pairs of ships traveling in parallel, as in the LatMix experiment. Present spaceborne capabilities are not sufficient for measuring vector surface currents, but a number of mission concepts, including WaCM from the US, and SKIM, STEROID, and SEASTAR from Europe, could provide this capability in the 2020’s.
In preparation for future spaceborne mission, NASA has developed the airborne Ka-band DopplerScatt system, which is capable of measuring simultaneous ocean vector currents and winds to produce synoptic maps (200 km x 100 km) at 200 m spatial resolution. From these estimates, surface current and wind stress curl and divergence can be calculated at scales smaller than 1 km. The system has been validated against coastal HF radar, optical surface current estimation, and surface drifters. These comparisons allow for estimating the effective depth for currents measured by microwave systems, and its dependence on wind speed and direction.
In this presentation, we report on DopplerScatt measurements of mesoscale and submesoscale surface currents, wind stress, and Sea Surface Temperture (SST) for two DopplerScatt campaigns. The first campaign, which coincided with the CARTHE SPLASH field campaign, took place over the Mississippi River plume and recirculation region in Barataria Bay, New Orleans. The data were collected both on and off the shelf, and included of large horizontal relative vorticity and divergent fronts caused by shear in the plume region and subduction of salty ocean water under fresh water carried by the plume or flowing from the Louisiana Bayou. A large number of surface drifters, drogued at about 50 cm, were released and in situ data were collected by ship-borne sensors. High-resolution optical SST data also were collected by various spaceborne sensors. The second campaign sampled a strong Gulf of Mexico mesoscale eddy generated by the loop current. During that deployment, DopplerScatt and the ROCIS optical surface current instrument flew simultaneously and sampled an area of about (150 km x 75 km) that included the velocity transition at the edges of the eddy, and surrounding regions. Additional glider data was available at different depths, plus ADCP and weather data collected at an oil station.
The results presented here center on the statistical characterization of the surface current, stress, their derivatives and interactions. We present the probability density functions, spectra and coherence for all these fields. We also examine the modulation of the surface wind stress curl and divergence, which drive vertical ocean and atmospheric circulation, due to the kinematic coupling between stress and currents, as well as changes in the marine boundary layer by SST. Finally, we examine the feasibility of measuring these features from spaceborne sensors of different space and time resolutions.

Marine litter is a major environmental problem. Quantifying the ecological and economic impacts of marine litter and assessing the efficacy of proposed solutions to the problem requires accurate knowledge of the distribution and transport of floating litter at the ocean surface. This can be done in the so-called Lagrangian framework, where trajectories of virtual particles are computed as they are transported through the ocean.

Floating litter is transported not only by ocean currents – including geostrophic, Ekman and submesoscale components – but also by waves and, if it has sufficient freeboard, also by winds. Accurately modelling the pathways of floating litter as a result of these phenomena requires high-resolution datasets for each field, both in space and time.

These high-resolution datasets exist, but they are often based on fields from Ocean General Circulation Models, which have known and unknown biases. These biases include their poor representation of processes such as Stokes drift and other wave effects. Hence, there is a need for transport models forced purely with remote sensed data.

In this presentation, we will review the currently available datasets of ocean surface flow from earth observations, and to what extent they can be used in Lagrangian simulations of marine litter, including microplastics and macroplastics, as well as other floating materials such as kelp. We will highlight how SKIM, one of the ESA EE9 candidate missions, is designed to improve the Lagrangian tracking of floating material by directly measuring the total surface currents.

Lagrangian particle tracking methods such as described here can be combined with optical, SAR and/or hyperspectral remote sensing techniques that identify and quantify the amount of floating debris. We will show here that such a combination of flow and imaging will lead to improved estimates of the two-dimensional distribution of floating marine litter. Lagrangian methods could potentially also be used to explore the provenance of litter observed in the ocean, by back-tracking virtual litter particles to their origin. For this latter application, it is particularly important to have self-consistent datasets; that is to have both surface currents and imaged marine litter from Earth Observation.