grid points for defining FORCES

[chenchai]

Hi,

I think FORCES(ng)%Tair is defined at the 'rho' points. I am not sure about FORCES(ng)%Uwind and FORCES(ng)%Vwind. Are they defined at the 'rho' points as well? I noticed that in the forcing data (coamps_wc13_wind.nc), u,v have coordinates at the 'rho' points.

Thanks,

James

[kate]

It depends. If ROMS has to interpolate and rotate the winds, it wants them at rho points. If you have already interpolated and rotated them, then Uwind is at u-points and Vwind is at v-points.

[chenchai]

Kate,

Thank you for the reply and help to clarify.

I have a question about wind rotation.

The code set_data.F has the following. Appears that the wind is rotate from (e,n)ward to (x,y)ward. Is this correct?

Thanks,

James

===================
# ifdef CURVGRID
!
! If input point surface winds or interpolated from coarse data, rotate
! to curvilinear grid.
!
IF (.not.Linfo(1,idUair,ng).or. &
& (Iinfo(5,idUair,ng).ne.Lm(ng)+2).or. &
& (Iinfo(6,idUair,ng).ne.Mm(ng)+2)) THEN
DO j=JstrR,JendR
DO i=IstrR,IendR
cff1=FORCES(ng)%Uwind(i,j)*GRID(ng)%CosAngler(i,j)+ &
& FORCES(ng)%Vwind(i,j)*GRID(ng)%SinAngler(i,j)
cff2=FORCES(ng)%Vwind(i,j)*GRID(ng)%CosAngler(i,j)- &
& FORCES(ng)%Uwind(i,j)*GRID(ng)%SinAngler(i,j)
FORCES(ng)%Uwind(i,j)=cff1
FORCES(ng)%Vwind(i,j)=cff2
END DO
END DO

[kate]

Yes, that's right. For ROMS to do the interpolation/rotation, the winds have to be on a regular lat/lon grid.

[wilkin]

It depends. If ROMS has to interpolate and rotate the winds, it wants them at rho points. If you have already interpolated and rotated them, then Uwind is at u-points and Vwind is at v-points.

I don't think that's correct. Stresses sustr,svstr are at the corresponding u,v points, but the wind components Uwind,Vwind are always represented at rho-points, even after they are interpolated from a simple gridded input file that is not aligned with the ROMS grid.

This is apparent in code that uses Uwind,Vwind like Nonlinear_bulk_flux.F. There you see that the wind stress components Taux,Tauy are computed at rho-points, and only at the end are they mapped to the velocity points with a step like:

Code: Select all

          sustr(i,j)=cff*(Taux(i-1,j)+Taux(i,j))
...
          svstr(i,j)=cff*(Tauy(i,j-1)+Tauy(i,j))

[kate]

It just goes to show that you need to check the source code to know the answer for sure.