Re: regridding problem (sub-global to global lat-long grid)

[Ansley Manke <manke@xxxxxxxxxxx>]

Hi Lev,
We worked this out off-line, answering the Users' list:

This has happened because only one of the two grids 
is modulo and the two encodings of longitude are 
different (one -180:180).  Thus one can ask for the 
modulo-axis variable on the grid of the other but not 
visa versa. To regrid in the other direction, first 
create a modulo longitude axis that is encoded 0:360 
with the same longitude values as the -180:180 (second 
variable). Then regrid both variables to this axis.

Changes are in the works which will treat *all*
longitudes as modulo and this difficulty should go away.

Ansley Manke

Lev Tarasoff wrote:

> I'm trying to blend two data sets. One (variable hd3a) defined on 
>    GRID GDK1
> name       axis              # pts   start                end
> XLONB     LONGITUDE          156 r   163.5W               8.5W
> YLATB     LATITUDE            97 r   36.75N               84.75N
> The netcdf file was created in ferret.
> I want to blend this data into data from another netcdf file (not
> created in ferret) which has the global grid:
>    GRID GBA1
> name       axis              # pts   start                end
> LON       LONGITUDE          360mr   179.5W(-179.5)       179.5E
> LAT       LATITUDE           180 r   89.5S                89.5N
>
> however, when I try to regrid from GDK1 to GBA1, I get all data flagged
> as bad data when using 
> hd3b=hd3a[g=GBA1]
>
> If I try 
> hd3b=hd3a[g=GBA1@mod]
> I get the following error:
> **ERROR: regridding: @MOD regridding to a non-modulo axis: HD3A
>
> Regridding variables from the 2nd data set to GRID GDK1 does work
> however (non-modulo regridding).
>
> help
>
> thanks
>
> Lev
>
>
>
> Lev Tarasoff -  Dept of Physics, University of Toronto,
>                60 St. George St., Toronto, Ontario, CANADA, M5S 1A7
>                Tel (519)-821-3555  Fax (416)-978-8905
>                email: lev@atmosp.physics.utoronto.ca