The synthetic method consists in removing for each observation of the total ocean signal, the ocean variability measured by altimetry.

The temperature and salinity profiles are used to compute the synthetic heights. All measurements of Argo floats from 2002 to 2008 and all temperature and salinity profiles from CTD relative to the 1993-2007 period are taken into account.

Geostrophic velocities from drifting buoys are extracted to compute the synthetic velocities. Between 1993 and 2008, 2 800 000 measurements of in-situ geostrophic velocities across the globe, are taken into account.

All the profiles are used to compute by integration along the water column a set of synthetic estimates of the mean dynamic topography according to the equation:

<h>_1993-1999 = h_insitu - h'_alti

However, by integrating these temperature and salinity measurements along the water column, the sea surface height obtained is representative of the steric content until the baseline profile. To return to a synthetic estimate of the ocean dynamic topography (barotropic and baroclinic components from the surface to the bottom), we had to add an estimate of the mean dynamic topography to this baseline profile. The missing component is approximated by subtracting to a large-scale estimate of the mean dynamic topography, a large-scale estimate of the mean steric component referenced to the depth of the baseline profile.

In a similar way, geostrophic velocities deduced from surface drifters can be used together with geostrophic velocities (anomalies) computed with altimeter slopes. The synthetic observations of velocities are computed by subtracting, along the drifting buoys paths, the altimetric velocity anomaly to the geostrophic velocity extracted from the buoy velocity according to the equation:

In addition to the geostrophic component, the velocity extracted from the drifting buoy path includes the tide currents, the Ekman currents, the inertial currents and others high-frequency ageostrophic currents. To extracte the only geostrophic component from the drifting buoys velocities, we had to precisely model each of these ageostrophic components and subtract them to the buoy velocity.

Le retrait des courants de marées et barotropes haute fréquence ne sont, à l'heure actuelle, pas encore suffisamment précis pour corriger les vitesses de bouées dérivantes. The removal of tidal and high-frequency barotropic currents are, currently, not yet precise enough to correct the drifting buoys velocities. Only the component of the Ekman currents, computed along the path, is removing from the drifting buoys velocities. Then a low pass filtering is applied to remove the signals with a period less than 3 days. Finally,  the synthetic observations of mean geostrophic velocities are obtained after averaging in a grid box size of 0.25 ° by 0.25.