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

20000812: McIDAS scouring

>From: "Thomas L. Mote" <address@hidden>
>Organization: University of Georgia
>Keywords: 200008130142.e7D1gkN00099 McIDAS mcscour.sh

Tom,

>Here's a couple of dumb questions, but at least they should 
>be easy to answer.
>
>Why does the McIDAS mcscour.sh script list grids to 5500, 
>when I am receiving grids that are 56xx? 

The GRID file numbers used increased a couple of months ago
when the ETA model forecasts were extended.  Sites taking
advantage of this by filing the new forecasts grids in GRID
files need to worry about which files they will go into and
about scouring those files.  You should change the entry in your
mcscour.sh file to reflect that fact that XCD-decoded grids now
go out to GRID file number 5620.  You could even extend it more
if you like.  Trying to scour GRID files that are not there does
no harm.

>I cannot find any documentation for the qrtmdg.k program, 
>can you point me to it?

QRTMDG is an XCD command.  Since you are running XCD, you should be
able to do a:

HELP QRTMDG

in the McIDAS Text and Command window.  Also, if you use the Unidata
MCGUI, it is available under the 'McIDAS XCD Programs' entry under
the Help button on the right top of the GUI.  Finally, all of the
commands are documented online at:

McIDAS-XCD Administrative Commands 
http://www.unidata.ucar.edu/packages/mcidas/mcx/xcd_admincmd.html

>BTW, I am getting the run around on opening the 503 port 
>for ADDE. Our departmental networking person is trying to 
>get some action.

OK.  I am pretty sure that at one time they had to OK port 500.  If
push comes to shove, try bargaining with them: relinquish 500 in
favor of 503, but do this as a last resort.

>From address@hidden Sun Aug 13 20:16:29 2000
>Subject: McIDAS RTGRIDS.CFG and RESOLV.SRV

>Sorry to keep peppering you with questions. I think I may 
>have found why I am having problems displaying some of the 
>gridded data. I assume the grid numbers in the RTMODELS.CFG 
>should match those in RESOLV.SRV (see below for file 
>snippets).

Yes, absolutely.

>In my case, they do not. I don't know whether 
>this was something that was missed in your upgrade, my 
>installation or at some other time. Regardless...

OK.

>If this is a problem, as I suspect it should be, which of 
>these files is best to change. I thought you said that ETA 
>was running out of grid numbers in the 5000's,

Correct.  In order to file all of the forecast times, in
GRID files "chunked" by time, more GRID files were needed.  The
best way to do this is to put the ETA grids into GRID files
that are in the same GRID file number range.  More on this below.

>so I assume 
>I should change the RESOLV.SRV file.

After you start decoding GRID into new GRID file numbers, you have to
change RESOLV.SRV entries so that ADDE routines can access the GRID
files.  RESOLV.SRV doesn't have anything to do with filing the grids
in GRID files, however.

>If so, how do I decide 
>what to use for R2?

R2?  Please look over the RTMODELS.CFG entries that I recommend below.

>Once I make the changes in the file, is 
>there naything else that I need to do as a result of this 
>discrepancy.

RTMODELS.CFG controls the name space for the XCD-decoded GRID files.
RESOLV.SRV will then configure ADDE to be able to access these (and
all other) files.  mcscour.sh is then used to scour these (and other)
files.

>Also, you can ignore my previous question on where to find 
>help for QRTMDG. It finally occurred to me that a "HELP 
>QRTMDG" might be in order ;-)

OK.  But, there is still the web-based documentation that you may
want to checkout.

>I still think that I may be 
>inadvertantly scouring files that I need to save, but I'm 
>not sure why. Possibly it's just that the files aren't 
>being properly saved due to the problem described above.

Let's check your RTMODELS.CFG file and proceed from there.

>...from ~mcidas/workdata/RTMODELS.CFG...
>
>SCRATCH=5001
>NGM=  3 5051 120000 240000 480000
>MRF=  1 5101 120000 240000 960000
>MAPS= 0 5201 30000
>MDR=  2 5301
>AVN=  3 5401 120000 240000 1200000
>ETA=  1 5501  60000 240000 480000     
                                         
OK, this is almost like what I use.  I have:

SCRATCH=5001
NGM=  3 5051 120000 240000 480000
MRF=  1 5101 120000 240000 960000
MAPS= 0 5201 30000
MDR=  2 5301
AVN=  3 5401 120000 240000 1200000
ETA=  3 5501  60000 240000 720000

The only difference is the last line.  It tells XCD (DMGRID) to file
ETA forecasts out to 72 hour.  If you figure out the number of sets
of GRID files for the ETA from this entry, you will see that GRID file
numbers can extend out to 5620.

Due to GRID file numbers extending out to > 5600, you need to adjust
entries in RESOLV.SRV and mcscour:

>...from ~mcidas/workdata/RESOLV.SRV...
>
>N1=RTGRIDS,N2=ALL,TYPE=GRID,RT=Y,K=GRID,R1=5001,R2=5500,C=Real-Time Grids,

Change this to:

N1=RTGRIDS,N2=ALL,TYPE=GRID,RT=Y,K=GRID,R1=5001,R2=5620,C=Real-Time Grids,

>N1=RTGRIDS,N2=AVN,TYPE=GRID,RT=Y,K=GRID,R1=5401,R2=5480,C=Real-Time AVN Grids,
>N1=RTGRIDS,N2=ECMWF,TYPE=GRID,RT=Y,K=GRID,R1=5251,R2=5260,C=Real-Time ECMWF 
>Grid
>s,
>N1=RTGRIDS,N2=ETA,TYPE=GRID,RT=Y,K=GRID,R1=5011,R2=5050,C=Real-Time ETA Grids,

Change this to:

N1=RTGRIDS,N2=ETA,TYPE=GRID,RT=Y,K=GRID,R1=5501,R2=5620,C=Real-Time ETA Grids,

>N1=RTGRIDS,N2=MDR,TYPE=GRID,RT=Y,K=GRID,R1=5301,R2=5310,C=Real-Time MDR Grids,
>N1=RTGRIDS,N2=MISC,TYPE=GRID,RT=Y,K=GRID,R1=5001,R2=5010,C=Other Real-Time 
>Grids
>,
>N1=RTGRIDS,N2=MRF,TYPE=GRID,RT=Y,K=GRID,R1=5101,R2=5200,C=Real-Time MRF Grids,
>N1=RTGRIDS,N2=MRF-UW,TYPE=GRID,RT=N,K=GRID,R1=101,R2=102,C=MRF Grids in UW 
>strea
>m,
>N1=RTGRIDS,N2=NGM,TYPE=GRID,RT=Y,K=GRID,R1=5051,R2=5090,C=Real-Time NGM Grids,
>N1=RTGRIDS,N2=NGM-UW,TYPE=GRID,RT=N,K=GRID,R1=1,R2=2,C=NGM Grids in UW stream,
>N1=RTGRIDS,N2=RUC,TYPE=GRID,RT=Y,K=GRID,R1=5201,R2=5280,C=Real-Time RUC Grids, 

Now, onto mcscour.sh:

Change:

qrtmdg.k GRID 5001 5500 1

to:

qrtmdg.k GRID 5001 5620 1

or even:

qrtmdg.k GRID 5001 6000 1

Please let me know if this is doesn't help.

Tom