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

20010424: WST wrong decoding in GEMPAK?



Christian,

I posted an update to the $GEMPAK/source/bridge/mt/mtdcod.f routine in
the gbuddy ftp account under: nawips-5.6/patches/dcmetr/mtdcod.f.

You can donload this file to the $GEMPAK/source/bridge/mt directory
and update your dcmetr decoder with:

cd $GEMPAK/source/bridge/mt
make clean
make all
make clean

cd $GEMPAK/source/programs/dc/dcmetr
make clean
make all
make install
make clean

This fix will restore the original behavior of checking for the
4 character METAR ID before stripping off the leading K or C.
this behavior was apparently lost in the transition from dchrly to dcmetr.

Steve Chiswell
Unidata User Support




>From: Unidata Support <address@hidden>
>Organization: UCAR/Unidata

>
>Christian,
>
>We see that @wst is getting the KWST observation, which is clearly not
>what you want for the CWST observation. The decoder is adding the station
>after stripping off the K- but apparently not looking for the KWST entry first
>which was the way dchrly was configured. I'll have to see what
>changes were made in dcmetr to change this behavior.
>
>Steve Chiswell
>Unidata User Support
>
>
>>From: Christian Page <address@hidden>
>>Organization: UCAR/Unidata
>>Keywords: 200104241340.f3ODeRL22917
>
>>Hi, I did change all entries of WST to either CWST or KWST correctly in all
>>tables file where I could find WST.
>>But it still doesn't help, it seems, by example, that area=@WST corresponds t
> o
>>KWST. Using area=@KWST or area=@CWST doesn't work. In my output (see
>>http://meteocentre.com/analyse/local/ne.gif )
>>you can clearly see the spurious High in the Northeast of the map. Any idea w
> h
>> y
>>it still doesn't work? BTW, I did stop and start ldm at 12Z after making the
>>changes.
>>
>>In sflist, I get:
>> GEMPAK-SFLIST>area=@KWST
>> GEMPAK-SFLIST>run
>> [SFLIST -3]  No stations reporting data.
>> Parameters requested: SFFILE,AREA,DATTIM,SFPARM,OUTPUT,IDNTYP.
>> GEMPAK-SFLIST>area=@CWST
>> GEMPAK-SFLIST>run
>> [SFLIST -3]  No stations reporting data.
>> Parameters requested: SFFILE,AREA,DATTIM,SFPARM,OUTPUT,IDNTYP.
>> GEMPAK-SFLIST>area=@WST
>> GEMPAK-SFLIST>run
>>KWST 241253Z AUTO 20008KT 7SM CLR 12/10 A2992 RMK AO2 SLP130 T01220100  
>>  
>> Parameters requested: SFFILE,AREA,DATTIM,SFPARM,OUTPUT,IDNTYP.
>> GEMPAK-SFLIST>exit
>>
>>
>>> 
>>> 
>>> Christian,
>>> 
>>> The confusion may be due to CWST and KWST both existing in the data
>>> stream.
>>> The station tables have an entry for WST (the canadian station).
>>> When KWST is decoded, the decoder first looks for KWST in the station
>>> table,
>>> and after not finding it, drop down to looking for WST.
>>> 
>>> To solve this, I would suggest changing WST to CWST in the station
>>> tables,
>>> and/or adding the KWST entry:
>>> KWST      99999 WESTERLY STATE                   RI US  4135  -7180   
>>> 25
>>> 
>>> Steve Chiswell
>>> Unidata User Support
>>> 
>>> 
>>> >From: Christian Page <address@hidden>
>>> >Organization: UCAR/Unidata
>>> >Keywords: 200104230757.f3N7v6L14956
>>> 
>>> >
>>> >Hi,
>>> >
>>> >I have some problem with a station in GEMPAK.
>>> >I am producing surface plots with barnes PMSL analysis near CYUL.
>>> >One of the station is WST (in Canada).
>>> >This station reports no pressure data according to the raw data.
>>> >This stations reports both in SA and METAR formats.
>>> >However, if I plot PMSL in GEMPAK, it shows PMS data for this station,
>>> and it
>>> >screws up my barnes analysis.
>>> >
>>> >Is it a bug in GEMPAK?
>>> >Or can I omit this station in the barnes analysis of PMSL?
>>> >
>>> >Thanks,
>>> >
>>> >Christian Page
>>> >address@hidden
>>> >
>>> 
>>> ***************************************************************************
> *
>>> Unidata User Support                                    UCAR Unidata
>>> (303)497-8644                                                  P.O. Box
>>> address@hidden                                   Boulder, CO
>>> ---------------------------------------------------------------------------
> -
>>> Unidata WWW Service                        http://www.unidata.ucar.edu/ 
>>> ***************************************************************************
> *
>>> 
>>
>>
>>
>>
>>Christian Page      
>>address@hidden    http://meteocentre.com/toulouse/
>>
>>Visiteur scientifique Tel:05 61 07 82 69
>>CNRM/GMEI/MNP et SCEM/PREVI/AERO/DP
>>
>
>