Feature #11771

Windspeed as new met field

Added by Philippe Le Sager 7 months ago. Updated 7 months ago.

In Progress
Target version:
Start date:
Due date:
% Done:



When the code requires the wind speed, we compute it at 1x1 from the u10m and v10m fields. We wonder how different it is from the wind speed calculated from the ERA-Interim native resolution. This issue is to investigate that difference. In other words, what is the difference between the average of the wind speed and the wind speed of the average wind ("average" meaning the remapping from ERA-I native resolution and the 1x1 grid used in TM5).

histo-diff-windpseed.png View (14.1 KB) Philippe Le Sager, 10/05/2018 03:23 PM

wind_speed_diff.png View (379 KB) Philippe Le Sager, 10/05/2018 03:23 PM


#1 Updated by Twan van Noije 7 months ago

Processes that will be affected are:
  • ocean DMS source
  • sea spray source
  • dry deposition
  • boundary layer height over land and associated diffusion
    The latter is part of the base (diffusion.F90)!

The 10 m wind components are only used for calculating the 10 m wind speed. The only reason to keep the separate wind components as input fields would be to able to keep on running the current model.

#2 Updated by Philippe Le Sager 7 months ago

Attached plots are:
- the comparison for one time step
- the histogram of the differences over one month of the 6-hourly subset over the all globe.

They are only a couple of points in the all dataset were the difference becomes large (~9 m/s). Most of it is pretty small. More stats over one month and all grid points:

mean difference: -0.02725006267428398
max difference: 4.0869646072387695
min difference: -9.80855941772461

Also available in: Atom PDF