Towards an improvement of wave forecasting in the Southern Ocean : Thanks to directional wave observations from CFOSAT

Lotfi Aouf (Division Marine et Océanographie Météo-France, France)

CoAuthors

Daniele Hauser (LATMOS/IPSL, France); Bertrand Chapron (IFRMER, France); Alessandro Toffoli (University of Melbourne, Australia); Alice Dalphinet (Meteo France, France); Cedric Tourain (CNES, France)

Event: 2020 Ocean Surface Topography Science Team Meeting (virtual)

Session: CFOSAT

Presentation type: Type Forum

The southern ocean is characterised by high waves generated by the fiercest westerly winds on earth in almost unlimited fetch conditions. The waves in Southern Ocean play a important role in controlling momentum and heat fluxes exchanges between the atmosphere and ocean. The CFOSAT mission with the SWIM instrument provides a unique directional wave spectra from beams which describe accurately wind-waves and swell ranging between 70 to 500m of wavelength. The goal of this study is to investigate the impact of the assimilation of partitions wavenumber commponents on the wave forecast in the Southern Ocean. The wave spectra used in this study are processed by the latest version of algorithms for level 2 products with significant improvement of the speckle noise on the along track direction. Assimilation runs have been performed from April 26 to 1 June 2019 during the beginning of austral Southern winter. Several assimilation runs have been implemented with wavenumber components only, SWH-nadir only and joint SWH-nadir and wavenumber components. These runs are compared to the control run without assimilation to examine the impact. SWH from model runs are compared to independent SWH from altimeters Jason-3, Saral and Sentinel-3.
The results show a significant reduction of SWH bias induced by the assimilation runs in the Southern Ocean. More precisely the best reduction is found when using the assimilation of wavenumber components from SWIM beam 10°. We also showed that the assimilation of partition wavenumber components seems more efficient to remove the SWH bias in comparison with the assimilation of SWH-nadir only, especially for SWH greater than 5m in the Pacific Southern Ocean.
The use of wavenumbers components from SWIM also induces a better dependence of dominant peak wavenumber and SWH in the wave growth phase in the southern ocean. This is in agreement with theoretical relationship from Elfouhaily et al. (1997). On the contrary the use of SWH-nadir only has a limited impact on the wave growth phase.
These findings related to the assimilation of SWIM wave partitions open a promising consequence to improve the wave climate description and the ocean/atmosphere coupling in the Southern Ocean.
 
Lotfi Aouf
Division Marine et Océanographie Météo-France
France
lotfi.aouf@meteo.fr