Knowledge of ocean surface geostrophic circulation has been greatly improved due to satellite observations. However, it is essential to use different observation sources (satellites and in situ data) in order to improve accuracy and resolution. This study combines along track Sea Level Anomalies (SLA) with geostrophic velocity estimated from surface drifters to map SLA and associated geostrophic current anomalies in the Gulf of Mexico. Firstly, substantial pre-processing is needed on the drifter data to extract the geostrophic component of the signal in order to be consistent with the physical content provided by altimetry. This step includes estimating and removing the Ekman current, Stokes drift and wind slippage. Three kinds of drifters are used: - Drifters belonging to Woods Hole Group, a CLS Group company that launches their own drifters in the Gulf of Mexico. - Drifters launched in the framework of the Lagrangian Submesoscale ExpeRiment (LASER) campaign (January-April 2016). - Drifters from Atlantic Oceanographic and Meteorological Laboratory (AOML). Secondly, drifters and along-track SLA from Jason2, HY2, SARAL and Cryosat-2 are combined through multivariate objective analysis to map a daily time series of SLA and associated geostrophic current anomalies from 01/09/2015 to 30/04/2016. Finally, comparisons with independent data reveal the improved agreement of maps combining both altimetry and drifter data, especially for the meridional component of geostrophic current. (C) 2020 COSPAR. Published by Elsevier Ltd. All rights reserved.

Vertical velocities at the ocean mesoscale are several orders of magnitude smaller than corresponding horizontal flows, making their direct monitoring a still unsolved challenge. Vertical motion is generally retrieved indirectly by applying diagnostic equations to observation-based fields. The most common approach relies on the solution of an adiabatic version of the Omega equation, neglecting the ageostrophic secondary circulation driven by frictional effects and turbulent mixing in the boundary layers. Here we apply a diabatic semigeostrophic diagnostic model to two different 3-D reconstructions covering the Southern Ocean during the period 2010-2012. We incorporate the effect of vertical mixing through a modified K-profile parameterization and using ERA-interim data, and perform an indirect validation of the ageostrophic circulation with independent drifter observations. Even if horizontal gradients and associated vertical flow are likely underestimated at 1/4 degrees x 1/4 degrees resolution, the exercise provides an unprecedented relative quantification of the contribution of vertical mixing and adiabatic internal dynamics on the vertical exchanges along the Antarctic Circumpolar Current. Kinematic estimates of subduction rates show the destruction of poleward flowing waters lighter than 26.6 kg/m(3) (14 divided by 15 Sv) and two main positive bands associated with the Antarctic Intermediate Water (7 divided by 11 Sv) and Sub-Antarctic Mode Waters (4 divided by 7 Sv) formation, while Circumpolar Deep Water upwelling attains around 3 divided by 6 Sv. Diabatic and adiabatic terms force distinct spatial responses and vertical velocity magnitudes along the water column and the restratifying effect of adiabatic internal dynamics due to mesoscale eddies is shown to at least partly compensate the contribution of wind-driven vertical exchanges to net subduction.

Under GODAE OceanView the operational ocean modelling community has developed a suite of global ocean forecast, reanalysis and analysis systems. Each system has a critical dependence on ocean observations – routinely assimilating observations of in-situ temperature and salinity, and satellite sea-level anomaly and sea surface temperature. This paper demonstrates the value and impact of ocean observations to three global eddy-permitting forecast systems, one global eddy-permitting model-independent analysis system, one eddy-resolving reanalysis system, and two seasonal prediction systems. All systems have been used to assess the impact of Argo profiles, including scenarios with no Argo data, and a degraded Argo array – unanimously concluding that Argo is a critical data set – the most critical for seasonal prediction, and as critical as satellite altimetry for eddy-permitting applications. Most systems show that TAO data are as important as Argo in the tropical Pacific, and that XBT data have an impact that is comparable to other data types in the vicinity of XBT transects. It is clear that no currently available data type is redundant. On the contrary, the components of the global ocean observing system complement each other remarkably well, providing sufficient information to monitor and forecast the global ocean.

Accurate estimate of ocean surface currents is both a challenging issue and a growing end-users requirement. In this paper ocean currents are calculated at two levels (surface and 15m depth) as the sum of the geostrophic and Ekman components. First, a new, global, 1 degrees/4 Mean Dynamic Topography, called the CNES-CLS13 MDT, has been calculated and is now available for use by the oceanographic community. By exploiting information from surface drifters and Argo floats, the new MDT resolves spatial scales beyond the resolution permitted by the recent Gravity and Ocean Circulation Experiment (GOCE) geoid models (125km). Associated mean geostrophic speeds in strong currents are increased by 200% on average compared to GOCE-based mean currents. In addition, for the first time, a two-level, monthly, empirical Ekman model that samples a spiral-like behavior is estimated. We show that combining both pieces of information leads to improved ocean currents compared to other existing observed products.

Abstract. Within the MyOcean R&D project MESCLA (MEsoSCale dynamical Analysis through combined model, satellite and in situ data), different estimates of the vertical velocities derived from observations have been compared. Two main approaches have been considered, one based on the retrieval of 3-D fields from the observations alone and one based on the analyses provided by MyOcean MERCATOR models. The motivation for this double approach is that, while data assimilation in numerical models is crucial to obtain more accurate analyses and forecasts, its results might be significantly influenced by specific model configurations (e.g. forcing, parameterization of smaller scale processes and spatial resolution). On the other hand, the purely observation-based approach is limited by the underlying assumptions of simplified dynamical models and by the relatively low resolution of present products. MESCLA tested innovative methods for the high resolution mapping of 3-D mesoscale dynamics from observations, developing new products that might be used to gradually build the next generations of operational observation-based products.

A new estimate of the Global Ocean 3D geostrophic circulation from the surface down to 1500m depth (Surcouf3D) has been computed for the 1993–2008 period using an observation-based approach that combines altimetry with temperature and salinity through the thermal wind equation. The validity of this simple approach was tested using a consistent dataset from a model reanalysis. Away from the boundary layers, errors are less than 10% in most places, which indicate that the thermal wind equation is a robust approximation to reconstruct the 3D oceanic circulation in the ocean interior. The Surcouf3D current field was validated in the Atlantic Ocean against in-situ observations. We considered the ANDRO current velocities deduced at 1000m depth from Argo float displacements as well as velocity measurements at 26.5°N from the RAPID-MOCHA current meter array. The Surcouf3D currents show similar skill to the 3D velocities from the GLORYS Mercator Ocean reanalysis in reproducing the amplitude and variability of the ANDRO currents. In the upper 1000m, high correlations are also found with in-situ velocities measured by the RAPID-MOCHA current meters. The Surcouf3D current field was then used to compute estimates of the Atlantic Meridional Overturning Circulation (AMOC) through the 25°N section, showing good comparisons with hydrographic sections from 1998 and 2004. Monthly averaged AMOC time series are also consistent with the RAPID-MOCHA array and with the GLORYS Mercator Ocean reanalysis over the April 2004–September 2007 period. Finally a 15 years long time series of monthly estimates of the AMOC was computed. The AMOC strength has a mean value of 16Sv with an annual (resp. monthly) standard deviation of 2.4Sv (resp. 7.1Sv) over the 1993–2008 period. The time series, characterized by a strong variability, shows no significant trend.

The MyOcean R&D project MESCLA (MEsoSCaLe dynamical Analysis through combined model, satellite and in situ data) was devoted to the high resolution 3-D retrieval of tracer and velocity fields in the oceans, based on the combination of in situ and satellite observations and quasi-geostrophic dynamical models. The retrieval techniques were also tested and compared with the output of a primitive equation model, with particular attention to the accuracy of the vertical velocity field as estimated through the Q vector formulation of the omega equation. The project focused on a test case, covering the region where the Gulf Stream separates from the US East Coast. This work demonstrated that innovative methods for the high resolution mapping of 3-D mesoscale dynamics from observations can be used to build the next generations of operational observation-based products.

In the framework of the ESA High Processing Facility (HPF), eight gravity models have been computed from the GOCE data since the beginning of the mission in 2009. The accuracy of these models for oceanographic application is assessed and compared with other geoid models (i.e., ITG-GRACE 2010S, EGM2008, EIGEN6C). For that purpose, ocean Mean Dynamic Topography (MDT) estimates are computed by subtracting the different geoids from an altimetric Mean Sea Surface. The MDT assessments were carried out by analyzing their associated geostrophic surface currents at different maximum harmonic degrees and by comparing them with independent in situ oceanographic data. Compared with GRACE-only geoid models, GOCE-only geoid models are in better agreement with independent data at scales smaller than 150 km while they show similar performances at scales larger than 200 km. Regional comparisons show that at 125 km the HPF releases compare well to independent observations with significant improvement with the last releases that use more GOCE data. At 100 km the western boundary currents are well resolved by the last releases, but the satellite-only geoid models need to be further improved to resolve smaller scales and reach the targeted 100 km resolution with higher accuracy.