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

19991118: NWX, WMO Headings, and the AFOS PIL extension in LDM

David,

The "B" entry in master.tbl uses the full WMO (header and ID),
so FPUS55 KLKN  would be different than FPUS57 KLKN.

The "S" entry uses only the ID portion, so a change
from XXXXXX KLKN to YYYYYY KLKN would still be accepted.

The difference between "S" and "W" is that "S" is used for
nwx to display all the stations in the table (typically for
products that you expect the station to issue regularly),
and "W" (for warnings etc.) forces nwx to search for the 
bulletins actually being issued in the $TEXT_DATA first
and only present those diamonds as choices.

When the stations started converting from SFD to AFD, I
changed the table from B to W since the tables were changing frequently,
and instead for instance, I have:
AREA_DISC    stations.tbl W $TEXT_DATA/pubprod/area                  area

Where stations.tbl has the the issuing stations in it.


If you instead want to match on the PIL rather than the station,
for instance, if a product which had been issued by KSEA changes
to KSEW, but the pil remails xxxWA, you could use the "R" designator
similar to the FF_GUID entry:
FF_GUID      ffg.bull     R $TEXT_DATA/fflood/guid                   ffg

The R designator would look for data inside the bulletin- the pil is inside.
This was useful for FFG bulletins where one issuing center would issue
bulletins for numerous areas.


So, you have several choices. Use either "S" or "W" for the WMO header
to use just the station ID. "S" will be faster since it just plots
diamonds at each location in the expected station table, then
searched the $TEXT_DATA if you select a station. "W" must search
for all stations first. If the stations are changing relatively infrequently,
I would suggest "S" except for  watches/warnings where you only want
to see those stations that actually have reports- vs "S" or "B" where
you expect every office to issue their bulletin each day.

You can use the "R" for the more stable pil identifier, but you should
match on the entire 5 or 6 character pil in those cases, eg NPWSEA
rather than just SEA since otherwise you would pull up every bulletin
that has the characters "SEA" in it...and some stations ids pop
up frequently in word contexts.

Today/Tomorrow the NWS is recoding a large number of WMO headers
on products....so I will have to make some table updates as well.

Steve Chiswell
Unidata User SUpport





>From: David Ovens <address@hidden>
>Organization: .
>Keywords: 199911181955.MAA15108

>To: Unidata Support
>Re: Ever-changing WMO Headers vs. AFOS PILs in the NWX context
>
>Hello,
>
>I am currently reworking our pqact.conf file for the LDM to use the
>AFOS PIL extensions as in this example:
>#
># State Weather Roundups
>WMO    ^...... .... ([0-3][0-9])([0-2][0-9])([0-5][0-9]) .*/pSWR.*
>       FILE    data/gempak/nwx/pubprod/roundups/(\1:yy)(\1:mm)\1\2.roundup
>
>I have figured out how to edit the $NAWIPS/tables/nwx/ guidata.tbl and
>master.tbl files to add Roundups to the NWX menus.  What I did for the
>master.tbl file is this:
>
>ROUNDUP      stations.tbl W $TEXT_DATA/pubprod/roundups              roundup
>
>This should work fine no matter what the WMO headings are for these
>reports, since I have used "stations.tbl W".  My question is, are the
>hard-coded WMO headings in the $NAWIPS/tables/nwx/*.bull files used
>for actually finding reports to display in NWX?  
>
>For example, should the NWS decide to change the WMO Heading for the
>zone forecast from Elko, say from FPUS55 KLKN to FPUS57 KLKN, would
>zone.bull need to be changed in order for NWX to place a red circle on
>the map for that product?  Of course, the standard pqact.conf entry
>for this
>
>WMO    ^FPUS5[1-6] .... ([0-3][0-9])([0-2][0-9])
>       PIPE    /usr/local/ldm/decoders/bin/ldmConnect -e .zone (\1:yy)(\1:mm) 
> data/gempak/nwx/pubprod/zone
>
>would also need changing.  If both of these would need changing,
>then I recommend that standard pqact.conf files use the AFOS PIL
>extension whenever possible and that the master.tbl for NWX use the
>"stations.tbl W" for the LOC_TBL and STN/BULL FLAG fields.  This
>should be less changeable since AFOS PILs change less frequently than
>WMO Headers (as far as I can tell).  I am willing to rework all of our
>pqact.conf file entries for NWX (for now) to use the AFOS PILs and
>send this back to you folks.
>
>I welcome comments about any oversights I've made or any flaws in this
>approach.
>
>Thanks,
>
>David
>-- 
>
>David Ovens            e-mail: address@hidden
>(206) 685-8108
>Dept of Atmospheric Sciences, Box 351640
>University of Washington 
>Seattle, WA  98195
>