Allan,
Here's some background info and an example from Jan 2007.
We receive our weather data via the Internet Data Distribution from the
National Center from Atmospheric Research. The Family of Services
products are ingested into that feed and we "capture" data by specifying
WMO product id's. That is how we get the standard airways observations
(e.g. ASOS data). I've noticed that many of the Canadian station are
reporting no precipitation. One example is station CXTO. That
station's data is "captured" by the bulletin headers "SACN98 CWAO"
(METAR format), and "SACN98 KWBC" (in the old standard airways
format(SA) format). It is my understanding that some countries are
still using SA rather than METAR format.
If you look at Jan 15, 2007 there are some specific examples where the
SA data show rainfall but the values are missing in the METAR format of
the data. Below are data extractions from that date. The string at the
beginning of each record up to the ":" mark is the ASCII data file from
which I've extracted the data. The file name also contains the specific
year, month, day, and hour of the observation. For example,
"2007011509_sao.wmo" refers to data for year 2007, month 01, day 15, and
hour 09. Here are the extractions in the 2 formats.
SA FORMAT
---------------
2007011509_sao.wmo:XTO SA 0900 AUTO8 M M M 185/-01/-04/MMMM/M/0009 PCPN
0.9MM PAST HR
PRESFR 8048 97MM=
2007011510_sao.wmo:XTO SA 1000 AUTO8 M M M 168/-01/-03/MMMM/M/0020 PCPN
1.1MM PAST HR
8051 31MM=
2007011511_sao.wmo:XTO SA 1100 AUTO8 M M M 155/-02/-03/MMMM/M/0036 PCPN
1.6MM PAST HR
/S+01/ 6055 57MM=
2007011512_sao.wmo:XTO SA 1200 AUTO8 M M M 143/-03/-04/MMMM/M/0053 PCPN
1.7MM PAST HR
SOG 02 6042 43MM=
2007011513_sao.wmo:XTO SA 1300 AUTO8 M M M 130/-04/-05/MMMM/M/0017 PCPN
1.7MM PAST HR
7038 76MM=
2007011514_sao.wmo:XTO SA 1400 AUTO8 M M M 131/-03/-04/MMMM/M/0030 PCPN
1.3MM PAST HR
/S+01/ 5024 22MM=
2007011515_sao.wmo:XTO SA 1500 AUTO8 M M M 120/-03/-05/MMMM/M/0050 PCPN
2.0MM PAST HR
6023 35MM=
2007011520_sao.wmo:XTO SA 2000 AUTO8 M M M 106/-03/-05/MMMM/M/0006 PCPN
0.6MM PAST HR
5023 38MM=
METAR format:
----------------
2007011509_sao.wmo:METAR CXTO 150900Z AUTO M01/M04 RMK AO1 9PAST HR
PRESFR 8048
SLP185 P0000 T10091037 50009=
2007011510_sao.wmo:METAR CXTO 151000Z AUTO M01/M03 RMK AO1 1PAST HR 8051
SLP168 P0000
T10131031 50020=
2007011511_sao.wmo:METAR CXTO 151100Z AUTO M02/M03 RMK AO1 6PAST HRS+01/
6055 SLP155
P0000 T10151027 50036=
2007011512_sao.wmo:METAR CXTO 151200Z AUTO M03/M04 RMK AO1 7PAST HR SOG
02 6042
SLP143 P0000 T10341043 50053=
2007011513_sao.wmo:METAR CXTO 151300Z AUTO M04/M05 RMK AO1 7PAST HR 7038
SLP130 P0000
T10371046 50017=
2007011514_sao.wmo:METAR CXTO 151400Z AUTO M03/M04 RMK AO1 3PAST HRS+01/
5024 SLP131
P0000 T10321042 50030=
2007011515_sao.wmo:METAR CXTO 151500Z AUTO M03/M05 RMK AO1 0PAST HR 6023
SLP120 P0001
T10331045 50050=
2007011520_sao.wmo:METAR CXTO 152000Z AUTO M03/M05 RMK AO1 6PAST HR 5023
SLP106 P0000
T10331048 50006=
As a further check on the daily precipitation totals, folks at the
Ontario Climate Center referred me to this web page to check the
observations:
http://climate.weatheroffice.ec.gc.ca/climateData/dailydata_e.html?timeframe=2&Prov=CA&StationID=31688&Year=2007&Month=1&Day=15
As you can see from that web page, the total for Jan 15 should be about
11.5mm,. The total from the SA data is 10.9mm, and 0.01 inches from the
METAR data. Clearly the METAR form of the data is inaccurate.
So this is an example of the problem I've been chasing down. It
actually has been that way since at least April of 2006 for some
stations. Hope my explanation and example is sufficiently clear. If
not, let me know what further information I can provide.
Sincerely,
Wayne
Wayne Gibson
Senior Research Assistant
Oregon State University
PRISM Group
Web Site: http://www.ocs.oregonstate.edu/prism
Office: 541.737.5696
----- Original Message ----- From: "Allan Darling" <address@hidden>
To: "Wayne Gibson" <address@hidden>
Cc: "Paul Noonan" <address@hidden>; "Walter Smith"
<address@hidden>; "Linda Miller" <address@hidden>;
"Fred Branski" <address@hidden>; "Unidata Support"
<address@hidden>
Sent: Friday, January 26, 2007 8:25 AM
Subject: Re: Contact at NWS Gateway
Wayne,
Yes, samples of products with suspected problems would be the starting
point for our investigation. Please send any examples you have. More
current data is better.
Regards,
Allan
Wayne Gibson wrote:
Fred,
Sorry to have bothered you when you're away.
Walter and Allan, would you like me to provide some specific examples?
Wayne Gibson
Senior Research Assistant
Oregon State University
PRISM Group
Web Site: http://www.ocs.oregonstate.edu/prism
Office: 541.737.5696
----- Original Message ----- From: "Fred Branski"
<address@hidden>
To: "Linda Miller" <address@hidden>; "Allan Darling"
<address@hidden>; "Walter Smith" <address@hidden>;
"Paul Noonan" <address@hidden>
Cc: "Wayne Gibson" <address@hidden>; "Unidata Support"
<address@hidden>
Sent: Thursday, January 25, 2007 11:54 AM
Subject: Re: Contact at NWS Gateway
Wayne and Linda,
Wayne is correct we receive Canadian SA data in both METAR and AIRWAYs
formats. both are made available in their native format and the
Canadian Airways reports along with a few remaining U.S. reports in
AIRWAYs code are converted to METAR here at the NWSTG. We are aware of
some translation problems which have been corrected. I'm not
completely
sure if all the problems we were aware of fixed. It is also possible
Wayne has identified a different problem.
Wayne should pass the details to Walter Smith and Allan Darling.
Allan's branch is responsible for the software and Walter for Data
Management.
Sorry I didn't get back to you sooner but I'm out of the office.
Cheers, Fred
Linda Miller wrote:
Hi Wayne,
FYI-Joe Irwin's email bounced. I believe he retired and I forgot
about that, so we'll hope that Fred will get back to you on this
question.
Thanks again,
Linda
Wayne Gibson wrote:
Linda,
That is a good question. If issued by the CWAO office, then the
format is standard airways (SA) and the precipitation is correct.
If issued by KWBC, then the format is METAR and the precipitation
is incorrect. As another check, the METAR format can't be correct
as the monthly precipitation totals for many of the Canadian sites
starting in April 2006 are very low (2-11 mm) and those values are
not realistic.
Thanks for tending to this quickly.
Wayne
--------------------------------------------------
- Wayne Gibson -
- Senior Research Assistant -
- Oregon State University -
- Spatial Climate Analysis Service -
- -
- Email: address@hidden -
- Web Site: http://www.ocs.oregonstate.edu/prism -
- Web Site: http://www.ocs.oregonstate.edu -
- Voice: 541.737.5696 -
--------------------------------------------------
On Wed, 10 Jan 2007, Linda Miller wrote:
Hi Wayne,
I've copied two people at the NWS, but when I read your email, I
wondered if this is happening on the Canadian side or at the
NWSTG. In any event, hopefully either Fred or Joe will get back
to you.
Thanks!
Linda
Wayne Gibson wrote:
Linda,
Good morning.
Recently I was looking at some Canadian surface data that we
receive via LDM. Apparently it is received in both SA (standard
airways) amd METAR format. Correct me if I'm wrong, but the SA
data is converted to METAR format at the NWSTG and
retransmitted. In that process, precipitation is not being
converted correctly. Do you know the appropriate contact at the
NWS that would be responsible for the conversion process? I'd
like to start a dialogue and see if the problem can be fixed.
Sincerely,
Wayne
--------------------------------------------------
- Wayne Gibson -
- Senior Research Assistant -
- Oregon State University -
- Spatial Climate Analysis Service -
- -
- Email: address@hidden -
- Web Site: http://www.ocs.oregonstate.edu/prism -
- Web Site: http://www.ocs.oregonstate.edu -
- Voice: 541.737.5696 -
--------------------------------------------------
--
Linda Miller - address@hidden
Community Services, Unidata
University Corporation for Atmospheric Research
P.O. Box 3000
Boulder, CO 80307-3000
303-497-8646 fax: 303-497-8690
--
===============================================================
Fred Branski, Operations Support, Performance Monitoring &
Data Management Branch, W/CIO11
RTH Washington
U.S. National Weather Service
Office of the Chief Information Officer
Telecommunication Operations Center
Phone: +301 713 0864 Ext. 146 Fax: +301 713 1409
Cell/Pager: +240 355 4468
E-mail: address@hidden
===============================================================
--
Allan Darling
Chief, Telecommunication Software Branch
NOAA's National Weather Service, OCIO/TOC
(301)713-0882 x114
http://www.nws.noaa.gov/tg/