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

New Ticket - [GEMPAK !CMJ-289235]: GEMPAK synop decoder

Giovanni,

The dclsfc decoder checks the i(r) indicator, and if it is a 3, it
assumes that P06M is 0.00:
-- rsdeco.f
C
C* If no rain observed, assume six hour dry period
C

The decoder does not assume a dry period of 0.00" for any other
time interval- that time interval would be given by the t(r) in 
the 6RRRt(r) group.

The SFLIST=dset will show the following periods in the default packing
file:

P03M;P06M;P09M;P12M;P18M;P24M

Only the P06M will be 0.00 if the 6RRRt(r) group is ommitted. The rest
are
filled with -9999.00. That is, the P24M parameter will only be defined
when t(r) is "4" as in 60004.

I do see synoptic obs with 60004 encoded, but don't know if it is legal
to have i(r) = 3 if there has been any observed precip in the previous
24 hours, but not in the most recent 6 hours- but the decoder does
not make that assumption.

Steve Chiswell
Unidata User Support



> Hello Steve,
> 
> I have one doubt about how exactly the "dclsfc" GEMPAK decoder
> interpret
> the synop message 111 Group precipitation indicator. I realized that
> when the precipitation indicator indicates that "
> 3 -- Precipitation omitted, no precipitation" the dclsfc decodes it as
a
> missing information (-9999.00) what does not seem to be apropriate.
The
> correct decodification should be "0.00". I believe that the missing
> value -9999.00 should be used only when the "111 Group precipitation
> indicator" is reporting "4 -- Precipitation omitted, no observation"
> Two examples are shown below:
> 
> 
> ### In Brazil
> 
> 111200 83997 32597 51408 10160 20129 30171 40216 51017 82201 333 20121
> 
> STN YYMMDD/HHMM STID TMPC P24M
> 839970 071011/1200 16.00 -9999.00
> 
> 
> ### In USA
> 
> 111200 72797 35966 60806 10094 20072 30121 40184 56005 91153 333 10161
> 555 91112
> 
> STN YYMMDD/HHMM STID TMPC P24M
> 727970 071011/1200 UIL 9.40 -9999.00
> 
> 
> 
> This interpretation of synop messages is causing a lot of data losses.
> 
> Cheers
> 
> 
> 
> --
> Giovanni Dolif
> Meteorologist, Msc. and Ph.D. Student
> National Institute for Space Research (INPE)
> Weather Forecast and Climate Studies Center (CPTEC)
> km 40, Rod. Pres. Dutra
> 12630-000, Cachoeira Paulista, SP, Brazil
> Phone: +55 (12) 3186-8400
> FAX: +55 (12) 3101-2835
> http://www.inpe.br/ingles/index.php
> http://www.cptec.inpe.br/
> 
> 



-- 
Steve Chiswell <address@hidden>
Unidata