RE: problems with NEXTYPE script

Okay here's the problem.  If I use any smaller of an area for the radar
grids using gdradr and the NEXTYPE process than is used in the Unidata
example, in order to create a higher resolution regional image (still using
the 720;500 kxky) the NEXTYPE process does not work correctly.  The result
from gdradr
is correct but the ptype grid that results from NEXTYPE has * * *
(overflow?)
where it should have -9999 for snow and frzn1/2,  When gdcntr is run it
contours
those areas as having snow/frzn where there isn't any.  If I do a gdlist on
the file (gfunc=snow) I see * * * where it should be -9999.  The areas where
there are valid snow echoes are correct.  A ptype file created using the
Unidata
example shows -9999 where there are no echoes of snow/frzn.

Easiest way to demonstrate this is to run gdradr and use something like
-28;-98;-40;-80 for grdarea, then use that grid in the NEXTYPE process.

I have tried this on GEMPAK5.6k and GEMPAK5.7.2p2..on locally compiled but
unmodified GEMPAK distros.

Any ideas?

Thanks,
Robert


-----Original Message-----
From: Robert Mullenax 
Sent: Friday, July 09, 2004 2:24 PM
To: Robert Mullenax; 'address@hidden'
Subject: RE: problems with NEXTYPE script


Sorry I hit send by accident.  Let me do some more testing before I
finish this question.

Thanks,
Robert Mullenax

-----Original Message-----
From: Robert Mullenax 
Sent: Friday, July 09, 2004 2:23 PM
To: 'address@hidden'
Subject: problems with NEXTYPE script


Steve,

As you know I have been using gdradr and your nextype script to produce
radar composites for quite some time.  I have had an ongoing problem
in that if I use a smaller area than what you use