Replies: 1 comment
-
If peaks are of chief concern, then nearest is probably your best bet. ESMF may have some other options in their regridding tools for this; if that's the case I think xESMF might be helpful as an xarray-based Python library using htis. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hello MetPy people,
can you help me with a question on interpolating windspeed timeseries from a scattered grid to wgs84 ?
Power from windmills is highly sensitive to peak windspeeds.
Interpolating timeseries smooths, lowers peaks, which would bias windpower estimates:
wind power is 0 below ~ 3 m/s, so interpolating 0 6 0 6 ... to 3 3 3 ... would give 0 power, hmm.
Don't want that, so I just do nearest-neighbor --
This has pros and cons.
Pros: we see exact power at the known, nearby staggered gridpoints
Cons: nearest-neighbor is blocky
Do you have any experience with nearest-neighbor vs. say IDW ?
Thanks -- appreciate your comments
cheers
-- denis
(Some notes and plots on what I'm doing are under https://gist.github.com/denis-bz .)
Beta Was this translation helpful? Give feedback.
All reactions