Wikindx Resources

Communication incl. Poster: BibTeX citation key:  Caniaux
Caniaux, G., Guichard, F., Bourras, D., Key, E., Giordani, H. & Bourlès, B. 2009. Evaluation of sea surface flux fields from NWP models during the emplacement of the Atlantic Cold Tongue in 2006. Work presented at Third International AMMA Conference, July 20—24, at Ouagadougou, Burkina Faso.
Added by: roussot 2009-11-23 20:58:48
Categories: Ocean processes, Ocean-atmosphere interactions
Creators: Bourlès, Bourras, Caniaux, Giordani, Guichard, Key
Publisher: African Monsoon Multidisciplinary Analyses (Ouagadougou, Burkina Faso)
Collection: Third International AMMA Conference

Number of views:  996
Popularity index:  52.7%
Maturity index:  published

For forcing oceanic models in the Gulf of Guinea, “realistic” sea surface fluxes fields are needed. For this purpose, sea surface heat and salt budgets have been computed from different numerical weather prediction models (ECMWF, NCEP and the French operational ARPEGE model) and other fields (i.e. including satellite field products).
The period selected (May/June/July 2006) corresponds to the seasonal formation of the Atlantic Cold Tongue. Four distinct regions of the central and eastern portion of the Atlantic basin are investigated: offshore the western African coast between 4°N and 8°N, and the Gulf of Guinea, north of the Equator, from the Equator to 6°S and south of 6°S. Differences and errors affecting these fields are evaluated and discussed through comparison with ship (from the R/V ATALANTE) and buoy (including buoys from the PIRATA mooring array) data collected during the AMMA/EGEE-3 experiment.
Results indicate very different wind stress, surface heat and salt budgets, with strong uncertainties and non-systematic biases, especially between the Equator and the African continent where atmospheric convection prevails. In this region, discrepancies between models can be related to differences in cloud cover, humidity, winds and aerosols. When these data sets are used for forcing an oceanic model, large differences and drifts are observed. These results stress the necessity of introducing flux corrections or computing new flux data sets.
Added by: roussot