Interpolation of forcing data to ROMS grid

[shrikantmp]

Hi ROMS users

I know that ROMS can interpolate the surface forcing data from a lat-lon grid onto a ROMS application grid, but I am confused as to the lat-lon extent of the forcing data that we can provide. For example, in my grid file, the corner latitudes and longitudes are LowerLeft(-30S,30E), LowerRight(-30S,120E),UpperRight(30N,120E),UpperLeft(30N,30E). My question is can I provide the surface forcing data that spans -33S to 33N in latitudes and 27E to 123E in longitudes i.e. can I provide surface forcing data on a lat-lon grid that is larger than and encompasses the ROMS application grid. Or the corner point lat-lons in the grid file and surface forcing file should match exactly? Please help. Thanks in advance.

[kate]

The forcing extent has to be at least as big as the ROMS grid. I have my winds on a global grid, then I can use it on any ROMS grid.

[shrikantmp]

The forcing extent has to be at least as big as the ROMS grid. I have my winds on a global grid, then I can use it on any ROMS grid.

You mean I can have forcing fields on a grid that is larger than my ROMS application grid?

[kate]

Yes, you can, if they are on a regular latitude, longitude grid. Your two choices are to (1) let ROMS interpolate from a larger grid or (2) provide ROMS with fields on exactly your grid. It queries the grid dimensions and will attempt to interpolate if the dimensions don't match. Don't give it a global grid with exactly the same number of points as your smaller model domain!

[shrikantmp]

Thank you very much Kate. That's really helpful

[sara.oceanography]

Thank you very much Kate. That's really helpful

Dear Kate,
I did the same works, but The temperature near the coast is very high, the temperature in the shallow water is higher than in deep water in the winter.
I think Roms read the land value.
I would be grateful if you could help me with this matter.

Best Regards,
sara

[kate]

It's true that the atmospheric model won't have the same land mask as ROMS. You can end up interpolating land values of say Tair over the coastal inlets. That's why Raphael Dussin told me to run sosie as a preprocessing step. There's a way to get it to extrapolate ocean values into the land for these atmospheric fields.