Re: 19990127: grib to netcdf

• To: Melissa Richey <richeym@xxxxxxxxxxxx>
• Subject: Re: 19990127: grib to netcdf
• From: Robb Kambic <rkambic@xxxxxxxxxxxxxxxx>
• Date: Mon, 8 Feb 1999 12:57:31 -0700 (MST)

Just wanted to let you know that I'm looking at the problem. It's appears
to be a  gribtonc coding problem, but I'm looking at ruc2.cdl
modification solution.


On Fri, 5 Feb 1999, Melissa Richey wrote:

Hi Robb- Thanks for your help with the gribtonc command line. I was able to figure out what was wrong from the output and I have been getting pretty good results. I have however, another question (or two) for

you. We would like to degrib the variable "water vapor mixing ratio".
In the ruc2.cdl file there is only a variable for "cloud water
mixing ratio". In the official grib table, the grib code associated to
"water vapor mixing ratio" is 185,

Where's the official grib table?  Wonder why it would be different then
the 388 Grib manual?

Robb...

refer to in another one of your emails, it shows that associated
to code 185 is "Turbulence SIGMET/AIRMET", which is the reserved_hybr
variable in the ruc2.cdl file. All the other variables that I need

degrib successfully, but when the decoder gets to the reserved_hybr variable, in the log file it prints out 28 levels of

Feb 04 22:30:52 gribtonc[988]: oversize GRIB product

and then at the 29th level, it prints out the typical decoding
output

Feb 04 22:30:52 gribtonc[988]: 149: wrote reserved_hybr(0,28,*,*) to ruc2.nc

for the rest of the 40 levels.

Why is this happening, and is there a variable for "water vapor mixing
ratio"? I looked in param.c, but couldn't find anything like that.
I am using a hybrid-coodinate ruc2 grib file.

Thanks!!!

Missy

+ + On Thu, 28 Jan 1999, Melissa Richey wrote: + + > + > Hi Robb- + > + > I have run gribtonc on a few files using the ruc2.cdl file. When + > I use ncdump on the new netcdf file, no variables have been + > degribbed- the output is basically what was in the original cdl file. + + Missy, + + Is the raw grid file a ruc2 file? + + + Run gribtonc in verbose mode with the messages to the screen, ie + + % gribtonc -vl - ruc2.cdl ruc2.nc < ruc2.raw + + This should show the details. If you send a sample of the output, I can

+ help get you started.

+ + Also, you can use gribdump to show the contents of the raw grib file. + + %gribdump grid.raw | more + + + The grib documentation is located at: + + > The GRIB document (Office Note 388) dated

+ > March 10, 1998 in ftp://nic.fb4.noaa.gov/pub/doc/nws/nmc/gribed1.

+ > + > It also located on : www.unidata.ucar.edu/staff/chiz/grib + + + Robb... + > + > Do you have any suggestions as to what might need to be modified in

+ > the cdl file? I know this is a vague question but I am somewhat at a loss,
+ > which you have probably already figured from previous emails.

+ > + > + > Thanks-

+ > Missy

+ > + + ==============================================================================

+ Robb Kambic                              Unidata Program Center
+ Software Engineer III                    Univ. Corp for Atmospheric Research
+ rkambic@xxxxxxxxxxxxxxxx                 WWW: http://www.unidata.ucar.edu/
+ ==============================================================================

+ + + +

==============================================================================
Robb Kambic                                Unidata Program Center
Software Engineer III                      Univ. Corp for Atmospheric Research
rkambic@xxxxxxxxxxxxxxxx                   WWW: http://www.unidata.ucar.edu/
==============================================================================