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

20020215: setting up non-decoding of certain grids in XCD (cont.)

>From: Anthony James Wimmers <address@hidden>
>Organization: UVa
>Keywords: 200202122315.g1CNFpx12009 McIDAS-XCD GRID

Tony,

I guess that I was _not_ mostly successful in stopping decoding of
grids past a certain forecast hour.  I have made another mod to
~mcidas/uvaworkdata/NOGRIB.CFG.  We'll see if the latest entries there
work or not later today.  Since I am in the office all day today, I
will keep an eye out and make adjustments as needed.  In the interim, I
deleted a number of GRID files in /p4/data/xcd to make space.

>The last time Jennie and I looked at it, we decided that we only needed
>the 1.25 x 1.25 degree Mercator projection of the MRF and AVN.

OK.  Did you know that the MRF model is going away?  The MRF products
(which only come out at 0Z) will be replaced by equivalent ones from
the AVN model, so there is no need to worry just yet.

>This is
>still the case. I don't think we knew at the time that you could control
>which projections come in. It would be a big bonus for us if we
>could keep those other projections out.

The setup of NOGRIB.CFG is to allow the user to _not_ decode grib
messages based on a number of circumstances: run times; valid times;
grid numbers; min/max pressure levels; and min/max parameter numbers.
What was unclear to me was how these were treated as a whole.  My first
assumption was wrong, so everything kept being decoded.  This go round,
I have assumed that all of the fields are used as a group, so I am
telling DMGRID to not decode for run times for all parameters at all
levels for run times greater than 13 hours and forecasts greater than
49 hours.

To get and idea of the coverage/density of the various grids in
NOAAPORT, check out this handy GEMPAK web page:

http://www.unidata.ucar.edu/packages/gempak/examples/models/grids/

The conduit feed has grid 3 fields that you will most likely
be interested in (global 1 x 1 degree).  This is one of the grids
that gribdec.k was unable to decode until my recent work.

>Also, I think that it might be wise to keep only the most recent
>model forecast for each grid: right after we get the 12Z run, we
>can remove the 0Z, etc. Is this possible?

>(Even if so, let's not do it just yet.)

OK.

>When I checked this morning, the /p4 partition was still totally full,
>so we apparently need to clear still more out. Hopefully the projection
>cuts will do it.

I cleaned up a bunch of GRID files in /p4/data/xcd, so there is breathing
room for the moment.

Tom