20050607: Gempak - GDDIAG not copying/creating grids

Tom,

Your message didn't show your script, but you should check your
use of GPACK where you will have to specify the range by the number of
bits used to pack the data, or no packing. If you are using GRIB packing then
the default number of bits will be 16, but if that isn't enough for your data 
range
you can increase that up to 32 bits. See phelp GPACK. If in doubt, try
NONE for the packing.

Also, the current release of GEMPAK 5.8.2 has the capability to create the grid 
file
if it doesn't exist, as well as reproject to a new grid. The old version that 
you are 
running does not have those features. 

teve Chiswell
Unidata User Support



>From: "Tom Green" <address@hidden>
>Organization: UCAR/Unidata
>Keywords: 200506071432.j57EWxjF015731

>Institution: North Carolina State University
>Package Version: 5.7.2p2
>Operating System: Linux
>Hardware Information: unsure-N/A
>Inquiry: Hello,
>
>I\'m currently trying to create grids for temperature advection, vorticity, et
> c. as shown in the script to be used by a Fortran program.  I believe that so
> me, but not all, of the data files that I am using have bad data at some poin
> ts in the grid, because when using GDDIAG on the second time listed in the sc
> ript, I get the \"Warning: Range Too High\" message when GDDIAG is creating t
> he grids.  The temperature advection, vorticity, etc. fields do NOT exist in 
> the original file - I am asking GDDIAG to create them and put them in the out
> put file.
>
>If I use GDCNTR to plot these grids in the original files and use a small gare
> a (ex. RDU), there is no problem with the plots.  Likewise, GARP is also able
>  to display the quantites when using the original files.  In my script, I att
> empted to have GDDIAG only compute these quantites for a smaller geographical
>  area, but that didn\'t seem to work either.
>
>If I have bad data along the sides of the grid, is there any way for me to cre
> ate these new grids that I want?  I want GEMPAK to create these grids rather 
> than attempt to do it in Fortran because of map factor issues, but I also don
> \'t need the values of the entire grid - the garea \"RDU-\" is sufficient for
>  what I need.  Any help you can provide would be greatly appreciated!
>
>Tom
>
>
>
--
NOTE: All email exchanges with Unidata User Support are recorded in the
Unidata inquiry tracking system and then made publicly available
through the web.  If you do not want to have your interactions made
available in this way, you must let us know in each email you send to us.