[gradsusr] Problems converting grib2ctl

Tue Jul 3 08:59:25 EDT 2018

Tamara,

For preprojected GrADS grids, xdef and ydef define a latlon grid.  The
domain and grid spacing are arbitrary.  For really old versions of
grib2ctl.pl

1. grid spacing is related to the input grid  spacing
2. domain is a balance between maximizing area and minimizing
       number of points that outside of the original grid and have
undefined values.

For old versions of grib2ctl.pl

1. grid spacing is related to the input grid spacing (I don't think I
changed it)
2. I increased the size of the domain at the expense of more undefined grid
points.

The calculations for "2" are very crude and it was assumed that  people
could
change the xdef and ydef to their own liking.

As for new versions of grib2ctl.pl, there aren't any.  NCEP hit the
limitations of
grib1 years ago.  I find grib2 easier to use and grib3 is being proposed.

Issue 2.

The newer version used "pdef lccr" and the older version used "pdef lcc".
The
lccr keyword tells GrADS that the input winds are grid relative and need to
be rotated
to earth relative.

Wesley

On Tue, Jul 3, 2018 at 12:31 AM, Krishnakumar AP <krishnaap7 at gmail.com>
wrote:

> Hi Tamara,
>
> I believe your ydef initial point is different, please check these values
> on both files.
>
> ydef 1249 linear -54.387000 0.0363636363636364
> --------------------------^
>
> This is what I understood from both ctl files.
>
> On Mon, Jul 2, 2018 at 11:14 PM, tschonholz <tschonholz at smn.gov.ar> wrote:
>
>> Hi everyone! I´m posting this message because I´m finding a hard time
>> using grib2ctl
>>
>> I use this bash to generate my ctl and idx on a server, but the original
>> ctl and idx are generated on different server than mine.
>>
>>
>>
>> #!/bin/sh
>> #*****Script para procesar el analisis
>>
>> today=$(date +%Y%m%d)
>>
>> echo ${today}
>>
>> #chequeo que exista la salida del WRF
>>
>> if [ -d /nas-r5c/WRF/${today}_00/ ];
>> then
>> #si ocurre, genero el ctl
>> echo ***Proceso las salidas grib del modelo**
>> cd /home/mgatto/
>> #rm /home/mgatto/modelos/WRF/data/gribwrf.ctl
>> perl grib2ctl.pl -verf /nas-r5c/WRF/${today}_00/UPPOUT/wrfprs_d01.%f3
>> /home/mgatto/modelos/WRF/data/indexwrf00.idx
>> >/home/mgatto/modelos/WRF/data/gribwrf00.ctl
>> gribmap -i /home/mgatto/modelos/WRF/data/gribwrf00.ctl
>>
>> else
>> exit
>>
>> fi
>> --
>>
>> doing this, i get this ctl:
>>
>> dset /nas-r5c/WRF/20180702_00/UPPOUT/wrfprs_d01.%f3
>> index /home/mgatto/modelos/WRF/data/indexwrf00.idx
>> undef 9.999E+20
>> title /nas-r5c/WRF/20180702_00/UPPOUT/wrfprs_d01.000
>> * produced by grib2ctl v0.9.13
>> * command line options: -verf /nas-r5c/WRF/20180702_00/UPPOUT/wrfprs_d01.%f3
>> /home/mgatto/modelos/WRF/data/indexwrf00.idx
>> dtype grib 255
>> options template
>> pdef 999 1249 lccr -54.387000 -94.331000 1 1 -35.000000 -35.000000
>> -65.000000 4000 4000
>> xdef 999 linear -94.331000 0.0624474869364198
>> ydef 1249 linear -54.387000 0.0363636363636364
>> tdef 49 linear 00Z02jul2018 1hr
>> * z has 28 levels, for prs
>> zdef 28 levels
>> 1000 975 950 925 900 875 850 825 800 775 750 700 650 600 550 500 450 400
>> 350 300 275 250 225 200 175 150 100 50
>> vars 130
>>
>>
>>
>> and all the variables.
>>
>> The problem is that the original ctl says this:
>>
>> dset ^wrfprs_d01.%f3
>> index ^wrfprs_d01.idx
>> options template
>> undef 9.999E+20
>> title wrfprs_d01.00
>> * produced by grib2ctl v0.9.12.5p45
>> dtype grib 255
>> pdef 999 1249 lcc -54.387000 -94.331000 1 1 -35.000000 -35.000000
>> -65.000000 4000 4000
>> xdef 999 linear -94.331000 0.0624474869364198
>> ydef 1249 linear -56.853000 0.0363636363636364
>> tdef 49 linear 00Z02Jul2018 1hr
>> * z has 28 levels, for prs
>> zdef 28 levels
>> 1000 975 950 925 900 875 850 825 800 775 750 700 650 600 550 500 450 400
>> 350 300 275 250 225 200 175 150 100 50
>> vars 130
>>
>>
>>
>> which means that I lose the south of my area because I don´t know why
>> it's not doing the same than the original.
>>
>> I believe it has to do with my grib2ctl version, does anyone know what
>> happens or could help me please?
>>
>> Thanks a lot!
>>
>> Tamara Schonholz- Servicio Meteorologico Nacional-Argentina
>>
>> _______________________________________________
>> gradsusr mailing list
>> gradsusr at gradsusr.org
>> http://gradsusr.org/mailman/listinfo/gradsusr
>>
>>
>
>
> --
>
> Best regards
> Krishnakumar.AP
>
>
> _______________________________________________
> gradsusr mailing list
> gradsusr at gradsusr.org
> http://gradsusr.org/mailman/listinfo/gradsusr
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://gradsusr.org/pipermail/gradsusr/attachments/20180703/d105f795/attachment-0001.html>