20020830: gdplot wind problem with AVN thin grids

Bill,

I have not been able to duplicate your problem using gdplot with the
avn thinned grids. Is this still the case with your thinned files?
If so, it would help if I compared your decoded file to mine as
well as seeing what my plot output was.

One possible reason for getting different looking plots 
(assuming that the forecast time, level, and parameter were the same)
would be if you were using the GDFILE template, and the next 6 hour forecast
run had arrived so that you ended up with a diffeent grid file being accessed).

I'm assuming you were using GVECT=wnd?

Steve Chiswell
Unidata User Support






>From: address@hidden (William Gallus)
>Organization: UCAR/Unidata
>Keywords: 200208301734.g7UHYgZ10935

>Hello,
>
>We noticed yesterday that when using GDPLOT on data in the AVN thin grids
>(_thin.gem), the plotted wind vectors change as one simply redefines the GAREA
> .
>This is obviously not good.   When GAREA=10;-80;40;-10, the 500mb winds over
>a storm cluster at 15N/78W were from the southeast.  When GAREA=10;-90;40;-10
>(a simple extension of the west boundary westward), the winds showed up as
>westerly!   This problem seems to happen in all of our _thin.gem files.  It
>is not present in any Eta file, or the MRF201 or AVN211 grids.
>
>Have you ever run across this problem?  I can't figure out how it could
>happen -- bit it obviously messed up my course lab exercise yesterday, since
>students will have totally different answers on steering flow depending on
>what they had their GAREA set to.
>
>Bill
>**********************************************
>*  Bill Gallus                               *
>*  Associate Prof. of Meteorology            *
>*  Dept. of Geological and Atmospheric Sci.  *
>*  3025 Agronomy Hall                        *
>*  Iowa State University                     *
>*  Ames, IA 50011                            *
>*  (phone) 515-294-2270                      *
>*  (fax)   515-294-2619                      *
>**********************************************
>