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

[McIDAS #SPC-381530]: McIDAS 2006 bug and a question

Hi Robert,

re:
> In working with Steve Chiswell to diagnose an apparent problem with GEMPAK
> or problem with the AMPS GRIB data (the Antarctic MM5 model stuff you worked
> on with Matt), I discovered a bug with GRIDDISP in McIDAS v. 2006.

OK.

> If you plot wind barbs without specifying units, (e.g. PARAM=WINDB NAV=C 
> LEV=850)
> it will plot the barbs in knots, but will label the image as barbs (mps).

I have verified that if one specifies UNIT=KT (or UNIT=KTS), then the label 
changes to KT.
Even more strange, however, is a comparison between two wind barb plots when one
is explicit about what units to use.  Compare the following two GRDDISP plots:

SF 1
ERASE F;GRDDISP RTGRIDS/GFS PARAM=WINDB MAP=NACONF LEV=850 UNIT=MPS COL=2
SF 2
ERASE F;GRDDISP RTGRIDS/GFS PARAM=WINDB MAP=NACONF LEV=850 UNIT=KTS COL=3
LB 1 2
TERM L ON

Some wind barbs change; some do not (look carefully).  I would expect that a 
plot
in MPS would result in wind barbs depicting half the speed of a plot of the
same grids in KTS (0.5144 since 1 KT == 0.5144 mps).  I will report this to 
SSEC.

> GEMPAK is still much prefered for model data,

I completely agree.

> but do you think it is possible (SSEC says it's only good for NOAAport grids) 
> to
> use the new McIDAS gridded data MySQL database with the AMPS model output?

Yes, it should.  SSEC's disclaimer is because they only provide the GRIB tables
for the GRIB 1 grids coming across NOAAPORT.  If it doesn't work, it should mean
that one needs additional/different GRIB tables.

> Matt said that is something he has not been able to try out yet.

Same here.

> I read some about DMGRID, but it was not clear to be how to test this.

The easiest way to test if DMGRID might work is to grab and try the 
gribdec.tar.gz bundle
from our passworded McIDAS FTP site:

ftp ftp.unidata.ucar.edu
  <user> umcidas
  <pass> uni_mci!
  cd unix/gribdec
  binary
  get gribdec.tar.gz
  quit

Unpack and build the gribdec package and check out GRIB 1 decoding.

NB: I have not updated gribdec.tar.gz with recent mods from XCD, so your
milage may vary.

> I receive the AMPS output via LDM.

In this case, if you have already built XCD, you can test DMGRID decoding
of GRIB 1 messages pretty easily by adding running of xcd_run in your
pqact.conf file.

Here is my recommendation for users decoding NOAAPORT HRS model data in XCD:

#########################################################################
#
# McIDAS-XCD section
#
DDPLUS|IDS      .*      PIPE
        xcd_run DDS
HRS     .*      PIPE
        xcd_run HRS

Just add the feed type you are receiving AMPS grids to the HRS action, or
create a new action.  For instance, if you are receiving AMPS (and only) data
in an EXP feed, you could add the following action:

EXP     .*      PIPE
        xcd_run HRS

(mind the tabs!!!!!)

> Thanks,

No worries.  Please let me know the results of your test.  I havn't tried 
decoding
AMPS model data using McIDAS GRIB decoders in over two years, so I am interested
in your results.


Cheers,

Tom
****************************************************************************
Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
----------------------------------------------------------------------------
Unidata HomePage                       http://www.unidata.ucar.edu
****************************************************************************


Ticket Details
===================
Ticket ID: SPC-381530
Department: Support McIDAS
Priority: Normal
Status: Closed