19990926: AVN Grids and McIDAS (cont.)

>From: weather <address@hidden>
>Organization: NMSU/NSBF
>Keywords: 199909242206.QAA23500 McIDAS-XCD

Robert,

>I might have been having bad data days when I pointed wxmcidas (I am in
>New Mexico at wxmcidas,. leaving for Palestine Monday) towards psnldm.
>Everything looks okay now, but I will will watch it to see if I need to
>increase the variable as you suggested.  On the other hand, wxmcidas
>seems to be really messed up.  I checked the .av1 file and it is the
>same as on psnldm.  If I do GRDLIST RTGRIDS/AVN NUM=50, this is the
>result:
>
>GRDLIST RTGRIDS/AVN NUM=50
>Dataset position 9      Directory Title= G213 12Z AVN 48 HR< VT<=72 HR
>PAR    LEVEL       DAY        TIME    SRC FHOUR     FDAY       FTIME  GRID  PRO
>---- --------- ------------ -------- ---- ----- ------------ -------- ---- ----
>T     850 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00    1 MERC
>T     850 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00    2 MERC
>T     850 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00    3 MERC
>T     850 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00    4 MERC
>T     850 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00    5 MERC
>T     850 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00    6 MERC
>T     850 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00    7 MERC
>T     700 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00    8 MERC
>T     850 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00    9 MERC
>T     700 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00   10 MERC
>T     500 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00   11 MERC
>T     700 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00   12 MERC
>T     700 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00   13 MERC
>T     700 MB   26 SEP 99269 00:00:00  AVN    12 26 SEP 99269 12:00:00   14 MERC

>As you can see there appears to be duplicate grids

The thinned AVN grids are global ones that come in in 8 octets that are
stiched together by DMGRID.  The listing above looks like the individual
pieces are not being stitched together.  I would guess that if you plotted
any one of the duplicates that you would find that they cover only
an eighth of the earth.  Yup, I just did this and verify the results
are only an eighth:

DATALOC ADD RTGRIDS WXMCIDAS.NSBF.NASA.GOV
GRDDISP RTGRIDS/AVN LEV=850 PAR=T DAY=1999269 FHOUR=12

>and it is labeled G213 12Z AVN..etc, when it is not.

Strange indeed.  The times that I have seen the stitching together fail
are typically indicative of more than one copy of DMGRID running.

>Even stranger when I do the same thing for the ETA:
>
>GRDLIST RTGRIDS/ETA NUM=50
>Dataset position 9      Directory Title= G213 12Z AVN 48 HR< VT<=72 HR
>PAR    LEVEL       DAY        TIME    SRC FHOUR     FDAY       FTIME  GRID  PRO
>---- --------- ------------ -------- ---- ----- ------------ -------- ---- ----
>W     200 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00    1 LAMB
>W     250 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00    2 LAMB
>W     300 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00    3 LAMB
>W     350 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00    4 LAMB
>W     400 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00    5 LAMB
>W     450 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00    6 LAMB
>W     500 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00    7 LAMB
>W     550 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00    8 LAMB
>W     150 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00    9 LAMB
>W     600 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00   10 LAMB
>W     100 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00   11 LAMB
>W     650 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00   12 LAMB
>W     800 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00   13 LAMB
>W     850 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00   14 LAMB
>W     750 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00   15 LAMB
>W     900 MB   26 SEP 99269 00:00:00  ETA     0 26 SEP 99269 00:00:00   16 LAMB

>I see it is labeled the same.

This could also happen if more than one copy of DMGRID is running.

>It seems that I am getting all of the world, but it is broken into
>small pieces by all these multiple grids.  When I plot one of the AVN T
>files, it will cover a small part of Asia, then another same time and
>variable will cover a small part of South America.

OK, so you have done the same thing as above.  The reason is because the
data comes in in pieces.

>I am completely confused.  RTMODELS.CFG is the same on both machines
>and the entries in dsserve.k LIST match RTMODELS.CFG but everything
>seems fine on psnldm (at least now) but all fouled up on wxmcidas.  I
>can still get all of the grids that McIDAS produces for the RUC, ETA,
>NGM, and MRF but the AVN is messed up and every directory label seems
>to be the same.

Check to see what processes are running.  If there is more than one copy
of DMGRID, shut the LDM down; make sure that all copies of DMGRID exit;
prehaps delete the bad files (they will fill up anyway); and restart
the LDM.

>Thanks for the help,

No problem.

Tom