CoAuthors

Isabelle Pujol (CLS, France); Elodie Charles (CLS, France); David Sandwell (SIO, USA); Antoine Delepoulle (CLS, France); Yannice Faugere (CLS, France); Nicolas Picot (CNES, France); Gerald Dibarboure (CNES, France)

A new Mean Sea Surface CNES-CLS is under development; it must meet multiple requirements that concern both the exploitation of on-flight mission and the implementation of a better validation at shortest wavelengths for future SWOT data. The objectives are twofold: we need an accurate determination of the long and medium wavelengths and the shortest wavelengths those less than thirty km must be improved!
It has already been shown (Pujol 2018, Schaeffer 2016), that the medium wavelengths of MSS, and particularly those of the latest 2015 version of DTU and CNES/CLS, have been greatly refined. However, it is still necessary to improve the processing of the ocean variability particularly for scales lower than 200 km, especially for missions in geodetic phase.
In parallel, we must also focus on the most accurate determination of the finest structures of the MSS and also of the geoid models. One approach consist in using the new generation of J1GM, CryoSat-2, Sentinel-3A and SARAL. These are sampled at 40 Hz for AltiKa and 20 Hz for the others, and offer a great potential to improve the small topographic structures. However, it appears that at these frequencies the noise level is too high and can degrade the determination of the MSS. It is therefore a quest for filtering these data and finding an optimal compromise between the residual noise and the filtered signal.
Moreover, as these data come from different technologies and preprocessing (SAR and LRM) they are differently affected by the local slope of the ground, and create systematic effects on the range. Then the slope correction must be applied to all these data before the MSS determination.
We propose here a synthesis of these various analyses and their impact on the determination of the future MSS.