[ncl-talk] regrid CCSM4 curvilinear grid to another curvilinear grid

Mary Haley haley at ucar.edu
Wed May 20 10:00:30 MDT 2015


Hi Brandon,

Would you be able to provide me with your script and data?  You can do this
offline if you like, using our ftp:

http://www.ncl.ucar.edu/report_bug.shtml#HowToFTP

I think the issue is simply that you will need to construct the lat/lon for
your destination grid, based on the information on the file.

I don't quite understand the metadata from the destination file, as it says
"inc = 1 degrees", but yet the size of the arrays seem rather large to be
just 1 degree spacing. I'm assuming, then, that the "inc = 1 degrees" is
not what I think it means.

--Mary



On Tue, May 19, 2015 at 3:43 PM, Brandon Fisel <bjfisel at gmail.com> wrote:

> Hello,
>
> I would like to regrid CCSM POP data onto another grid, but I am unsure
> where to begin as the destination grid has (i, j) coordinates that do not
> correspond to lat, lon. Running cdo sinfon:
>
> Source (ccsm) grid:
>    File format: netCDF2
>     -1 : Institut Source   Ttype    Levels Num  Gridsize Num Dtype :
> Parameter name
>      1 : unknown  CCSM     instant      1   1    122880   2  F32  : SST
>    Grid coordinates :
>      2 : curvilinear  > size      : dim = 122880  nx = 320  ny = 384
>                         TLONG     : min = 0.0147311016  max = 359.99602
>  degrees_east  circular
>                         TLAT      : min = -79.2205226  max = 89.7064096
>  degrees_north
>                         available : xvals yvals
>
> Destination grid:
>    File format: netCDF
>     -1 : Institut Source   Ttype    Levels Num  Gridsize Num Dtype :
> Parameter name
>      1 : unknown  unknown  constant      1   1   1684800   1  F32  : TEMP
>    Grid coordinates :
>      1 : generic      > size      : dim = 1684800  nx = 1560  ny = 1080
>                         I         : first = 1  last = 1560  inc = 1
>  degrees
>                         J         : first = 1  last = 1080  inc = 1
>  degrees
>
> I have tried using ESMF regridding, but I am unable to plot the regridded
> data due to allocation limits (workspace reallocation exceeding the maximum
> size 60000000), and do not trust the data has been correctly regridded.
>
> Any help or ideas where to begin would be appreciated. Thanks in advance.
>
> Cheers,
>
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20150520/c6665b41/attachment-0001.html 


More information about the ncl-talk mailing list