[GEMPAK #UYH-961710]: GDGRIB issues

Kim,

I have a solution for the first part of this ticket.  Because the second scalar 
input uses a different level than the first scalar input (PRES@0%none), LEVEL1 
from both scalars are combined to be the new LEVEL1 and LEVEL2 outputs.  The 
fix is to use in-line notation when defining GRDNAM, such as GRDNAM = 
THTA@60:30%pdly.

The second part of your ticket I've managed to uncover a lack of support for 
GVCORD = PDLY in GDGRIB, which I'm still looking into and will update you when 
more is known, and hopefully with a solution shortly.

Michael


> Hi Michael,
> 
> Sorry, I seemed to have sent the wrong version of gdgrib to you! I had to 
> write different versions for whether I used NARR data or output from my WRF 
> simulations because gddiag defined the levels differently for each case. 
> Changing the level to 0:30 will output a grib file, but it is empty. This 
> happens even if I scale the parameter manually. Does this occur on your end?
> 
> 
> Thanks,
> 
> Kim
> 
> ----- Original Message -----
> From: "Unidata GEMPAK Support" <address@hidden>
> To: address@hidden
> Cc: address@hidden
> Sent: Tuesday, December 7, 2010 4:18:54 PM
> Subject: [GEMPAK #UYH-961710]: GDGRIB issues
> 
> Kim,
> 
> The incorrect LEVEL2 definition issue may be a bug that you discovered.  I'm 
> waiting to hear back from others about it, and will let you know what I find 
> out.
> 
> As for the error with GDGRIB, correct me if I'm wrong, but it seems that the 
> last grid which was added to your file was
> 
> GRDNAM   = combinedGGT%PDLY
> 
> with
> 
> GFUNC    = ADD(THTEGGT@0:60%PDLY,GGTpdly@30:60%PDLY)
> GDATTIM  = F000
> GLEVEL   = 0:60
> GVCORD   = PDLY
> 
> however, GDINFO shows combinedGGT saved as
> 
> NUM       TIME1              TIME2           LEVL1 LEVL2  VCORD PARM
> 347     080130/0000F000                          0  30     PDLY COMBINEDGGT
> 
> 
> Can you confirm this on your end? (in GDINFO set GLEVEL = all, GVCORD = all, 
> GDATTIM = all and GFUNC = COMBINEDGGT)
> 
> If I run GDGRIB as you have in your script, with GLEVEL = 30:60, I see an 
> error reporting no grid exists (true).  If I change it to GLEVEL = 0:30, I am 
> able to write out the grib file as expected.
> 
> I want to make sure that we're on the same page here, as there are a lot of 
> steps involved to get to this point, and have COMBINEDGGT written with the 
> correct time, v-coord and levels.  If we're good up to this point, then could 
> the problem you're experiencing simply be that GLEVEL is defined incorrectly 
> in GDGRIB?
> 
> Let me know what you find and hopefully I'll have some more info for you 
> about the GDDIAG level error.
> 
> Michael James
> 
> 
> Unidata
> 
> 
> 
> 
> 
> 
> Ticket Details
> ===================
> Ticket ID: UYH-961710
> Department: Support GEMPAK
> Priority: Normal
> Status: Open
> 
> 


Ticket Details
===================
Ticket ID: UYH-961710
Department: Support GEMPAK
Priority: Normal
Status: Open