Wind speed interpolation #1
Conversation
|
Thanks, @ebbekyhl I'll accept this. I cannot see the jupyter notebook with the test on EC-Earth 3, can you share it? |
| from_name = "wnd{h:0d}m".format(h=int(from_height)) | ||
| from_name2 = "wnd{h:0d}m".format(h=int(from_height2)) # definition of second wind speed (for instance at 100m) | ||
|
|
||
| wnd_spd = ds[from_name] * (to_height / from_height)**((1 / np.log(from_height)) * np.log(ds[from_name2] / ds[from_name])) |
There was a problem hiding this comment.
Shouldn't the equation be:
wnd_spd = ds[from_name] * (to_height / from_height)**((np.log(ds[from_name2] / ds[from_name])/(np.log(from_height2) / np.log(from_height)) )
do you agree @ebbekyhl ?
There was a problem hiding this comment.
Yes, you are right. The former formulation only works when "from_height2" = 100 m and "from_height" = 10 m.
I would agree with Oskar's formulation (OskarSidor@bc8c622#r132404361):
wnd_spd = ds[from_name] * (to_height / from_height)**
(
np.log(ds[from_name2] / ds[from_name])
/
np.log(from_height2 / from_height)
)
|
@martavp You can find the comparison for EC-Earth-3 in the bottom of this jupyter notebook: https://github.com/martavp/atlite/blob/cmip6/sandbox_ebbe/download_cmip_cutout_and_process.ipynb |
Added the function from here that calculates the hub wind speed based on either 10-m wind speeds and roughness or based on an interpolation of wind speeds at two heights.
I also added the argument "model" to the "convert_wind" function which should be either "interpolation" or "extrapolation" such that both methods can be applied and compared.
Currently, this has been tested for the EC-Earth3, which leads to a slightly higher capacity factor with the model="interpolation" setting (see jupyter notebook).