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

19990131: hemispheric grids in GEMPAK/GARP

Gary,

I looked at the 500mb hght contours in Garp, and can see
that apparently there is a problem in the sm9s() and sm5s()
smoothers that are used in the Hght_m function.

Note that when using garp, and simply plotting HGHT, instead
of sm9s(sm9s(hght)) which is the default general scalar
function for Hght_m, the contour lines do not break at the
point where the hemispheric grids wrap.

The smoothing functions do not seem to create contour breaks in
gdcntr at the location wrap, so this appears to be a problem unique
to Garp, and possibly the way that the grid and/or grid size
is being interpreted and passed to the contouring routine.

I'll have to look into what is being sent to the routines
by Garp and see if I can tell where the discrepancy is occuring.

Steve Chiswell
Unidata User Support




>From: address@hidden (Gary Lackmann)
>Organization: .
>Keywords: 199902021231.FAA01862

>Steve,
>
>I do not get the same result using GDCNTR- there is no break in the
>contours.  I should have checked this before but I assumed it would
>give the same result as GARP.  That narrows it down I suppose.
>
>We are using SUN OS 5.6, garp v2.01, and our sysadmin Tom McDermott
>has applied all available patches.  Thanks,
>
>Gary
>
>
>> Gary,
>> 
>> I see what you are refering to in your plot, however, I
>> do not get the same results when I use your file here.
>> 
>> This might be peculiar to your platform, though I have
>> tested under SGI, HP, and Sun here. It could also be you are
>> missing a patch somewhere along the way.
>> 
>> Can you provide me with the following information:
>> What platform/OS are you using?
>> Do you see the same result when you use gdcntr?
>> What version of Garp are you using?
>> 
>> Steve Chiswell
>> Unidata User Support
>> 
>> 
>> >From: address@hidden (Gary Lackmann)
>> >Organization: .
>> >Keywords: 199902012236.PAA14562
>> 
>> >Steve,
>> >
>> >I placed two images (one for mrf, one for avn) and the
>> >mrf gempak grid file in your ftp directory under 
>> >the subdirectory ~/gbuddy/incoming/sunyb  
>> >
>> >Thanks for any help/ideas on how to solve this.
>> >
>> >Gary
>> >
>> >> 
>> >> 
>> >> Gary,
>> >> 
>> >> I am looking at the 500mb heights and surface pressure grids 
>> >> from the avn thinned grids in both gdcntr and Garp 2.01 and don't
>> >> see any line breaks in the contour field around 30W which is the
>> >> wrap point for the thinned grids. The thinned grids are in 8
>> >> pieces and get stitched together in the gempak grid file by dcgrib.
>> >> I also looked at the f084 500mb height for the MRF and that looks OK to.
>> >> 
>> >> If you can post a few gifs, that would help me see what you are looking a
> t.
>> >> I may need to look at the same grid file you are looking at too.
>> >> If you can post that to ~gbuddy/incoming that might be useful.
>> >> 
>> >> Steve Chiswell
>> >> Unidata User Support
>> >> 
>> >> >From: address@hidden (Gary Lackmann)
>> >> >Organization: .
>> >> >Keywords: 199902012039.NAA11198
>> >> 
>> >> >> 
>> >> >> Gary,
>> >> >> 
>> >> >> The wrapping point for the AVN thinned grids (eg 1.25x.125 degree grid
> s i
>> > s
>> >> >> 30W, not near the Greenwhich meridian. This is the global AVN data set
> .
>> >> >> 
>> >> >> Is this the AVN data set you are refering too?
>> >> >> 
>> >> >> Can you provide me with the program input and file data type you are u
> sin
>> > g
>> >> >> where you see this behavior?
>> >> >> 
>> >> >> Thanks,
>> >> >> 
>> >> >> Steve Chiswell
>> >> >> **********************************************************************
> ***
>> > ***
>> >> >> Unidata User Support                                    UCAR Unidata P
> rog
>> > ram
>> >> >> (303)497-8644                                                  P.O. Bo
> x 3
>> > 000
>> >> >> address@hidden                                   Boulder, CO
>  80
>> > 307
>> >> >> ----------------------------------------------------------------------
> ---
>> > ---
>> >> >> Unidata WWW Service                        http://www.unidata.ucar.edu
> /  
>> >    
>> >> >> **********************************************************************
> ***
>> > ***
>> >> >> 
>> >> >Hi Steve,
>> >> >
>> >> >The decoder program is dcgrib and the file type is gempak.  Both the
>> >> >MRF and AVN_thin grids exhibit this behavior when viewed using GARP.
>> >> >The MRF breaks at the meridian, and the AVN is broken at 30W.  If you
>> >> >like, I can save a few example gifs out of garp and post them on our
>> >> >web page, or for ftp.  Also, I could place an example grid file in the
>> >> >anonymous ftp directory if that would help.
>> >> >
>> >> >The pattern-action we use for the MRF is shown below:
>> >> >
>> >> >HRS     ^H[HPUV][EF][A-Z][0-9][0-9] KWB. ([0-3][0-9])([0-2][0-9])..1
>> >> >        PIPE    /weather/ldm/gempak/NAWIPS/bin/sol/dcgrib
>> >> >                -d data/gempak/logs/dcgrib.log
>> >> >                -g /weather/ldm/gempak/NAWIPS/gempak5.4/tables
>> >> >                PACK data/gempak/hds/YYMMDDHH_mrf.gem
>> >> >
>> >> >The AVN exhibits the same behavior (except breaks at 30 instead
>> >> >of the Greenwich meridian) when viewed in GARP.
>> >> >
>> >> >HRS     ^H.[I-P]... KWB. ([0-3][0-9])([0-2][0-9]).*
>> >> >        PIPE    /weather/ldm/gempak/NAWIPS/bin/sol/dcgrib
>> >> >                -d data/gempak/logs/dcgrib.log
>> >> >                -g /weather/ldm/gempak/NAWIPS/gempak5.4/tables
>> >> >                -q"lin,dlat=2.5,dlon=2.5"
>> >> >                PACK data/gempak/hds/YYMMDDHH_thin.gem
>> >> >
>> >> >
>> >> 
>> >> *************************************************************************
> ***
>> >> Unidata User Support                                    UCAR Unidata Prog
> ram
>> >> (303)497-8644                                                  P.O. Box 3
> 000
>> >> address@hidden                                   Boulder, CO 80
> 307
>> >> -------------------------------------------------------------------------
> ---
>> >> Unidata WWW Service                        http://www.unidata.ucar.edu/  
>    
>> >> *************************************************************************
> ***
>> >> 
>> >
>> 
>> ****************************************************************************
>> Unidata User Support                                    UCAR Unidata Program
>> (303)497-8644                                                  P.O. Box 3000
>> address@hidden                                   Boulder, CO 80307
>> ----------------------------------------------------------------------------
>> Unidata WWW Service                        http://www.unidata.ucar.edu/     
>> ****************************************************************************
>> 
>