3 Data

The CNES-CLS22 MDT is calculated from a combination of altimeter and space gravity data, in situ measurements and model winds. The method allows us to estimate the mean over the 1993-2012 reference period, but is not limited to observations from this period. For each in situ observation, we remove the altimetric variability referenced to the 1993-2012, thus obtaining an estimate of the mean dynamic topography corresponding to the reference period. The following datasets are used:
MSS. The CNES-CLS22 MSS derived for the 1993– 2012 reference time period by Schaeffer et al. (2023) is used.
Geoid model. The satellite-only geoid model GOCO06s (Kvas et al. 2021) is used with the CNES-CLS22 MSS in the computation of the MDT first guess.
Altimeter sea level anomalies (SLAs). The DUACS-2021 (Taburet et al. 2022) multimission gridded sea level and derived geostrophic velocity anomaly products distributed by the Sea Level Thematic Assembly Center (SL-TAC) from the CMEMS altimeter are used.
Dynamic heights. These are calculated from temperature and salinity (T/S) profiles from CTD and ARGO from CMEMS CORA Release November 2022 (period 1993-2020, Szekely et al. 2023), processed by the In Situ Thematic Assembly Center (INS-TAC) of the Copernicus Marine Environment and Monitoring Service (CMEMS). In order to carry out a cross-validation of the new CNES-CLS22 MDT with the latest CNES-CLS18 MDT (Mulet et al. 2021), about 5% of the T/S profiles which will not be used for the calculation of the MDT, is kept. These 5% of the T/S profiles are randomly selected since 2017, in order to have a set of validation data that is independent of the new MDT but also of the CNES-CLS18 MDT.
In situ velocities. Two types of in situ drifting buoy velocities are used, the 6-hourly SVP-type drifter distributed by the Surface Drifter Data Assembly Center (SD-DAC; Lumpkin and Johnson 2013) and the Argo floats surface velocities from the regularly updated YOMAHA07 dataset for the period 1997–2021 (Lebedev et al. 2007). SVP-type drifters consist of a spherical buoy with a drogue attached in order to minimize the direct wind slippage and follow the ocean currents at a nominal 15 m depth. When the drogue gets lost, the drifter is then advected by the surface currents and the direct action of the wind. In this study both the 15 m drogued and the surface undrogued drifter velocities over the 1993– 2021, period are considered for the MDT calculation. 10% of the AOML drifters dataset are randomly selected after 2017 and kept for validation of the results (independent data from CNES-CLS22 and CNES-CLS18). This new CNES-CLS22 MDT also uses velocity data estimated from HF radars located in the Mid Atlantic Bight area, East coast of the US from Cape Hatteras to Cape Cod. We used the cleaned, detited, high-frequency signal-filtered and mean currents over the period 2006-2016 (processed by Rutgers University, Roarty et al. 2020).
Wind data. Wind stress data are needed for the calculation of the wind-driven velocities (Sect. 2.3) that is used to remove part of the ageostrophic component from drifter velocities. We use the 3-hourly, 80 km resolution wind stress fields from ERA5 (Hersbach et al. 2018) for the period 1993–2021.