EMPTY plots when entire domain values = ZERO

Fri Sep 14 19:17:43 EDT 2007

FRI 14SEP07: 1615PDT

Dear Davide

Many thanks for your very simple yet effective solution to my problem.

I used your second suggestion of adding minimal noise, that is, merely
adding lon/10000000 to my defined values solved the problem of a constant =
0 field that GRADS does not like to plot.

Thank you again!

Take care

Regards

Kevin M Levey, MSc (University of Cape Town)
Director of Meteorological Operations
CustomWeather, Inc
San Francisco, CA, USA

http://www.1stweather.com
http://www.myforecast.com
http://www.customweather.com

> grads cannot create contours of a constant field (eg.: all values =0)
>
> I know only two solutions:
> 1) test by set gxout stat if the field is constant or not, then decide
> what to plot
> 2) if contour levels have been fixed (eg 'set clevs 1 5 10 ...') you
> could add a minimal noise to your field (eg: 'd cc+lon/10000000)
>
> hoping this is the answer you were looking for
> bye bye
> Davide
>
> On Wed, 2007-09-12 at 15:26 -0700, Kevin M Levey wrote:
> > WED 12SEP07: 1525PDT
> >
> > Hi All.
> >
> > I've come across a problem that I am a little stumped on and it's the
first
> > time I've encountered this error.
> >
> > I am plotting GFS (1x1 degree) convective cloud values and if I plot the
> > entire dataset, the plot comes out fine with convective cloud showing
up.
> > However, if I choose a smaller domain where ALL grid points are 0, I
expect
> > GRADS to plot the zero value, however, instead I get a blank plot. Has
> > anyone ever come across this problem before?
> >
> > Running GRADSC from the command line gives me the following output:
> >
> > ga-> test
> > No hardcopy metafile open
> > All files closed; all defined objects released
> > All GrADS attributes have been reinitialized
> > Cannot plot color bar: No shading information
> >
> >
> > If I make the domain larger where there is at least one grid box with a
> > value > 0 then the plot comes out OK.
> >
> > A copy of the test script:
> >
> > 'reinit'
> > 'set mpdset hires'
> > 'set mpt * off'
> > 'set mpt 0 0 1 6'
> > 'set mpt 1 0 1 1'
> > 'set grads off'
> > 'set timelab on'
> > 'set display color white'
> > 'clear'
> > 'open /prod/custom/Input/modelplots/GFS/temp/gfs_102.ctl'
> > 'clear'
> > 'set xsize 800 600'
> > 'set display color white'
> > 'set lat -36 -16'
> > 'set lon 10 36'
> > 'set map auto 1 1 6'
> > 'define cc = TCDCcvl'
> >
> > 'run /prod/custom/Input/modelplots/GFS/scripts/rgbset_cloud2.gs'
> > 'set gxout shaded'
> > 'set csmooth on'
> > 'd cc'
> >
> > 'run /prod/custom/Input/modelplots/GFS/scripts/SA_cities_clouds.gs'
> > 'run /prod/custom/Input/modelplots/GFS/scripts/cbarn'
> > 'draw title  CONVECTIVE CLOUD COVER (%) \INIT: 12UTC 12SEP07
> > VALID: 20H00 SUN 16 SEP 2007'
> >
> > The above domain for this hour has ALL grid points over South Africa =
ZERO,
> > but grads wont plot anything. The forecast hour before and after where
at
> > least one gridpoint has a value  > 0 then the plots came out just fine,
and
> > it only happens when ALL grid points for the domain are ZERO.
> >
> > Any help would be appreciated indeed.
> >
> > Thanks.
> >
> > GRADS information:
> >
> > Built Mon May 23 13:17:11 EDT 2005 for i686-pc-linux-gnu
> >
> > This version of GrADS has been configured with the following options:
> >
> >    o This is a 32-bit LITTLE ENDIAN machine version.
> >    o Command line editing (readline) ENABLED.
> >    o CIRES/CDC (http://www.cdc.noaa.gov) SDF/XDF interface DISABLED.
> >      This version is NOT configured to read NetCDF/HDF-SDS files.
> >    o DTYPE netcdf and DTYPE hdfsds are DISABLED.
> >    o OPeNDAP (a.k.a. DODS) gridded data interface DISABLED.
> >    o OPeNDAP (a.k.a. DODS) station data interface DISABLED.
> >    o PCMDI (http://www-pcmdi.llnl.gov) LATS interface ENABLED.
> >      This version is configured to write GRIB files (but no NetCDF/HDF
> > files).
> >    o DAO (http://dao.gsfc.nasa.gov) Athena Widget GUI DISABLED.
> >    o NRL/DAO/PCMDI XA or ImageMagick Image Output DISABLED.
> >    o printim command for direct png/gif output ENABLED.
> >      (via the GD Library -- http://www.boutell.com/gd)
> >
> > Regards
> >
> > Kevin M Levey, MSc (University of Cape Town)
> > Director of Meteorological Operations
> > CustomWeather, Inc
> > San Francisco, CA, USA
> >
> > http://www.1stweather.com
> > http://www.myforecast.com
> > http://www.customweather.com
> >
> --
> Sacchetti Davide
> ARPAL UO3 Centro Meteo Idrologico Regione Liguria - Dir. Scientifica
> V.le Brigare Partigiane 2 16121 Genova (I)
> tel: +39 010 6437535
> mail: davide.sacchetti at arpal.org   web: www.meteoliguria.it