[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[GEMPAK #WCL-790351]: Grid Diagnostic Problems

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: [GEMPAK #WCL-790351]: Grid Diagnostic Problems
  • Date: Fri, 17 Aug 2007 12:56:31 -0600
Kevin,

The problem you are having is within the PD_DRCT routine
which was translated from FORTRAN to C in 2005.
This routine computes the direction of the vector before
any scaling is done, and the components of the q-vector have
magnitudes on the order of 10**-11. The routine is doing
a check for component values of "0" using a check of
values less than 10**-5, so the direction of all your qvectors is being
set to zero.

If you built from source, I can send your the quick change for the PD_
routine.

Alternatively, a temporary fix for plots instead of specifying a SCALE
value is to do:
SCALE = 999
GDPFUN = smul(expi(10,11),qvec(thta,geo))

If you can upgrade to 5.10.3, I will repackage the current release with
a change to the PD_DRCT routine in both source and binary releases.

Steve Chiswell
Unidata User Support




> I have encountered a problem when plotting q-vectors using the 5.10.2
> release of gempak.  Below are three q-vector images;
> 5.9.3qvector500.gif is from June 5th, this was the operational map
> created on the last day before we went down for about two months.
> When we got our machine back we upgraded to 5.10.2 and now the q-
> vector map plots the qvectors only pointing from north to south as in
> image 5.10.2qvector500.gif which is from today 1200 UTC.  The
> counterpart to this is the 5.7.4qvector500.gif, which was made
> operationally on an older machine with nearly the same script as the
> one for 5.10.2.
> 
> I investigated this further and instead of using the qvec diagnostic
> function, I made my own as follows,
> vecr(mul(-1,DOT(DVDX(geo),GRAD(thta))),mul(-1,DOT(DVDY(geo),GRAD
> (thta))))
> 
> and whether I used this or the actual qvec function, I got the same
> result.  Similarly, if I change the grid file to the RUC instead of
> an OA of upperair data, the same thing would occur.
> 
> Any ideas as to what changed?
> 
> Regards,
> Kevin Goebbert
> 
> _________________________________
> 


Ticket Details
===================
Ticket ID: WCL-790351
Department: Support GEMPAK
Priority: Normal
Status: Closed