Re: question for converting NCOM ocean model grid to CCSM3 POP model grid

A variant Of Keith's suggestion:

[1] "linint2_points" may [will] be a bit slow.

[2] Use "linint2_Wrap" to interpolate NCOM to a 1x1 grid

[3] then use the already generated SCRIP weight files
    to interpolate to the desired grid.
    
    map_1x1d_to_gx3v5_aave_da_060302.nc
    map_1x1d_to_gx3v5_bilin_da_060302.nc
    
    I have put the above file on ftp
    
    ftp ftp.cgd.ucar.edu
    anonymous
    email
    cd /ftp/pub/shea/POP
    prompt
    mget map_1x1d*
    quit
    

>Hi Keith,

>Yes, now I think about how to use SCRIP to generate remapping weights.
>It seems that I should create two grid file and then use SCRIP to
>generate remapping weights.

>Has somebody done this?

On Aug 7, 2006, at 6:08 PM, Keith Lindsay wrote:

> One option is to use SCRIP generated remapping weights.

**************************************
 
>Dennis is correct. NCOM grids are lat/lon. The latititude variable is
>typically not equally spaced. One option is to use SCRIP generated
>remapping weights. This is probably appropriate if you are remapping
>fluxes and want to conserve the integral of the flux. However, if you
>are not concerned with conserving the total integral of your field, you
>might consider using linint2_points, after calling ndtooned on the POP
>lat and lon variables.
>On Mon, 7 Aug 2006, chengs wrote:
>
>> Hi,
>>
>> Is there a way to convert NCOM ocean model grid to CCSM3:POP model grid
>> (gx3v5)?
>> I think NCOM ocean model grid is lat/lon grid. Is this right?
>******************************************************************

_______________________________________________
ncl-talk mailing list
ncl-talk_at_ucar.edu
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Tue Aug 08 2006 - 11:10:31 MDT