This 10m wind parameter is the wind speed that has been used by the wave model, which is coupled to the atmospheric model.
For this reason:
- it is archived on the wave model's native grid, with the same land-sea mask as that model. Therefore, this parameter is not defined over land and wherever else the wave model is not defined, where it is encoded as missing data. Improper decoding of the missing value usually results in very large values being given for these land points.
- the wave model resets all values below 2 m/s to 2m/s. The reason for this is that as the winds become weak, the long waves (swell) try to drive the wind from below but this is not modelled in the IFS, as it assumes that the wind profile should be logarithmic (+- stability correction). To account for this effect, the whole of the boundary layer scheme would need to be revised. A simple trick to avoid the problem is to boost the weak winds to 2m/s, which is outside the range where the waves can potentially drive the wind.
- this parameter is actually the 10m neutral wind speed as determined from the atmospheric surface stress (see documentation on Ocean Wave model output parameters).
- If wave altimeter data were assimilated, the analysis of this parameter also contains wind speed updates that come directly out of the wave height updates.
This parameter should not be used for looking at the quality of reanalysis surface wind - the u and v components of the 10m wind (atmospheric parameters 165 and 166) should be used instead.