[gradsusr] Plotting GFS_HD apcpsfc every 3 hours problems

[Wesley Ebisuzaki]

Jeff,
> for f in (list of file from f03 to fNN)
You were suppose to replace "(list of file from f03 to fNN)" with a list 
of the
files that had the forecast data in grib2 format.  f03 is a common NCEP
convention to identify the file as containing the 3 hour forecast data.
For example, your list may be

for f in gfs.t00z.pgrbf03.grib2 gfs.t00z.pgrbf06.grib2 
gfs.t00z.pgrbf09.grib2 gfs.t00z.pgrbf12.grib2

Wesley
 





Jeff Chabot wrote:
> Hello Wesley,
>
> Thanks for your explanation of the 3 hour versus 6 hour accumulation 
> problem with gfs_hd.
>
> I tried your workaround, but it failed pretty quickly for me.  I have 
> never used wgrib or wgrib2 and I searched on the internet for some 
> documentation, but I couldn't find much.
>
> This is where if failed for me:
>
> >sh
> >[ -f IN.grb ] && rm IN.grb
> >for f in (list of file from f03 to fNN)
> sh: syntax error near unexpected token `('
>
> Also, I believe that you would need to open the data file first, but I 
> couldn't get that to work either:
>
> >wgrib 
> http://nomads.ncep.noaa.gov:9090/dods/gfs_hd/gfs_hd20101202/gfs_hd_00z
> could not open file: 
> http://nomads.ncep.noaa.gov:9090/dods/gfs_hd/gfs_hd20101202/gfs_hd_00z
>
> So, I am not sure how to use this workaround.  Any further help would 
> be greatly appreciated.
>
> Sincerely,
>
>
> Jeff
>
>
>
>     Message: 4
>     Date: Wed, 01 Dec 2010 09:21:02 -0500
>     From: Wesley Ebisuzaki <Wesley.Ebisuzaki at noaa.gov
>     <mailto:Wesley.Ebisuzaki at noaa.gov>>
>     Subject: Re: [gradsusr] Plotting GFS_HD apcpsfc every 3 hours problems
>     To: GrADS Users Forum <gradsusr at gradsusr.org
>     <mailto:gradsusr at gradsusr.org>>
>     Message-ID: <4CF659CE.3050106 at noaa.gov
>     <mailto:4CF659CE.3050106 at noaa.gov>>
>     Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
>     Jeff,
>
>       The "Joys of Compatibility" strikes again (old code expects 6 hour
>     accumulations).
>     Anyways there is a function in wgrib2 to make all the
>     accumulations into
>     3 hour chunks.
>
>     1) put all the APCP records into one file (IN.grb) in order (f03 to
>     fNN).  Here is how to do it in sh.
>
>         [ -f IN.grb ] && rm IN.grb
>         for f in (list of file from f03 to fNN)
>         do
>            wgrib2 f -match ":APCP:" -append -grib IN.grb
>         done
>
>        IN.grb should contain the APCP from one forecast run starting
>     from 3
>     hour forecast to N hour forecast.
>
>     2) Run wgrib2 to normalize the records.
>
>        wgrib2 IN.grb -match ":APCP:" -set_grib_type c1 -ncep_norm
>     precip.grb
>
>                Wesley Ebisuzaki
>
>
>     Jeff Chabot wrote:
>     > Hello GrADS Users,
>     >
>     > I am looking for help plotting 3 hr precip using gfs_hd (from
>     > nomads.ncep.noaa.gov <http://nomads.ncep.noaa.gov>
>     <http://nomads.ncep.noaa.gov>) every three
>     > hours.  I have it working fine using nam and gens.  However, when I
>     > plot apcpsfc using gfs or gfs_hd, every other hour it seems like
>     it is
>     > plotting 3 hour precip, then 6 hour precip, and back to 3 hour,
>     then 6
>     > hour precip all the way through the 65 points.  This makes the
>     precip
>     > appear to pulsate, every other frame.  I have attached two images to
>     > illustrate the problem.  Does anyone have a work around for this
>     > issue?  The problem is isolated to just gfs data (both gfs_hd and
>     > gfs), but doesn't happen with GFS Ensembles.
>     >
>     > Versions:
>     > Grid Analysis and Display System (GrADS) Version 2.0.a9
>     > Fedora-13-x86_64
>     >
>     > Data source example:
>     >
>     > sdfopen:
>     >
>     http://nomads.ncep.noaa.gov:9090/dods/gfs_hd/gfs_hd20101130/gfs_hd_12z
>     >
>     > Attached files:
>     > precip_US_6.png (appears to show 3 hours of precip)
>     > precip_US_7.png (appears to show 6 hours of precip)
>     >
>     > Here is my attempt to work around the issue for just gfs_hd / gfs,
>     > though it doesn't solve the problem:
>     > (Note: This is the section of my .gs file related to my work around
>     > attempt):
>     >
>     > *grads code
>     >
>     > t = 1
>     > t0 = 0
>     > alt = -1
>     > while ( t <= 65 )
>     >
>     > 'set t 't
>     > 'define precip = apcpsfc/25.4'
>     > if alt > 0 then
>     >  'define product = precip(t='t')'
>     >  'define over4= precip(t='t')'
>     > endif
>     > if alt < 0 then
>     >  'define product = (precip(t=t) - precip(t=t0))'
>     >  'define over4   = (precip(t=t) - precip(t=t0))'
>     > endif
>     >
>     > *more grads code
>     >
>     > t = t+1
>     > say ' = 't
>     > t0 = t0+1
>     > say ' = 't0
>     > alt = alt * -1
>     > endwhile
>     >
>     > Any help here would be greatly appreciated.  Thanks again.
>     >
>     > Sincerely,
>     >
>     >
>     > Jeff Chabot
>     > jsc219 at gmail.com <mailto:jsc219 at gmail.com>
>     <mailto:jsc219 at gmail.com <mailto:jsc219 at gmail.com>>
>     > http://jeffsweatherservice.com
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> gradsusr mailing list
> gradsusr at gradsusr.org
> http://gradsusr.org/mailman/listinfo/gradsusr
>