Coarse resolution atmospheric data
The data for the first exepriment seasonal cycles obtained by monthly mean forcings are based on the daily mean, spatially interpolated ECMWF fields, provided here at the resolution of the ocean models (1/4x1/4). The date are available as such (atmos.tar ), but please note that the data amount to be transferred is impressive (120MB).
In any case, for the experiment, you only need the monthly mean averaged wind stress fields (in m2/s2, .TX in west east direction, .TY in south nord direction. Data are in standard modb format, with each level corresponding to one month (k=1 for January ) and turbulent kinetic energy input (.TK). You can download a compressed tar file: (coarse.tar.Z) The flux of turbulent kinetic energy is proportional to the third power of the wind speed, with a proportionallity factor depending on your closure schemes. In the file .TK you find thus themonthly average of the || V10 ||
For temperature and salinity fields, a relaxation towards sea surface values should be done (Monthly Sea Surface Temperature and Salinity).
CHANGE OF SPECIFICATION 23/1/96 IMPORTANT: The relaxation should be done with a surface flux formulation where the flux of T or S is proportional to the difference between model and given sea surface values. The proportionallity factor should be the same for everybody in the project: 1.0 meter/day. If you refer a volume relaxation in your first grid box, this is equivalent to a relaxation time of 1*DZ days.
Please note also that all the surface or initial fields contain a mask, so that when you interpolate on YOUR grid, do not forget to implement a specific treatment of these points.