Rotating WRF u- and v-winds before and after (?) reprojection

[politeness]

Dear user community,
I have a code that requires u- and v-winds on a regular lat/lon grid. Therefore, I need to reproject WRF lambert conic conformal output to lat/lon. I plan to use wrf_uvmet() to rotate the u- and v-winds to Earth coordinates first. However, I wonder if I need to rotate them again after reprojection to lat/lon?

Your help would be appreciated.

Thank you.

 

[Ming Chen]

Hi,
You can rotate the grid-relative winds to earth-relative winds by the formulas below:
earth_u = u*cosa(ix,iy)-v*sina(ix,iy)
earth_v = v*cosa(ix,iy)+u*sina(ix,iy)
u, v, sina, cost are standard output in wrfout files.

 

[Nathaniel Winstead]

I am curious to know if the winds in the constant pressure level diagnostic output files (U_PL and V_PL) are grid relative or earth relative. I ask because those files do not contain sina or cosa in them and so doing the grid rotation becomes annoying if you have to store a regular wrfout file as well as all of the wrfout_PLEV files?