# Re: Curvilinear grids and subregions...

```Hi Gary,
The way to set a region with this kind of data is to use a mask.
Something like this:  (using the new continuation-lines syntax
from v5.40 for readability; the sub_var definition could all be
on one line)

let  sub_var =  if  lon[d=gridinfo.nc] ge `xmin` and \
lon[d=gridinfo.nc]  le `xmax` and \
lat[d=gridinfo.nc]  ge `ymin` and \
lat[d=gridinfo.nc]  le `ymax` then \

Gary Strand wrote:

> I've got some data on the PCM POP grid (see http://www.cgd.ucar.edu/pcm for
> more about it) and the necessary grid information for it.
>
> In what order should I do the following calls to get a subregion (say, the Gulf
> Stream) of that grid properly plotted? I've been stumped for a couple days on
> this!
> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
> use pop_data.nc
> use gridinfo.nc
> !
> let xmin = -90; let xmax =  10
> let ymin =  20; let ymax =  80
> let xdel =  10; let ydel =  10
> let mid_lon = `xmin+(xmax-xmin)/2`
> let mid_lat = `ymin+(ymax-ymin)/2`
> !
> go mp_lambert_cyl `mid_lon` `mid_lat`
> set region/x=`xmin`:`xmax`/y=`ymin`:`ymax`
> let mp_lambda = lon[d=gridinfo.nc] * deg2rad
> let mp_phi    = lat[d=gridinfo.nc] * deg2rad
> set grid `var[d=1],return=grid`
> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
>
> The above doesn't work - the "set region" command uses the i,j of the original
> data, which is only "indexical" space, not geographic space.
>
> I'm hoping I just need to rearrange the order of the calls to get it to work
> right!
>
> Thanks.
> --
> /\        Gary Strand (303) 497-1336                                NCAR ML262A
>   \_][    www.cgd.ucar.edu/ccr/strandwg                      1850 Table Mesa Dr
>       \___strandwg@ucar.edu                       Boulder, Colorado, USA  80305

--
Ansley Manke  Pacific Marine Environmental Laboratory  Seattle WA  (206)526-6246

```