KNOWN ISSUES/PROBLEMS WITH DS472.0

July, 2009

As a result of the changed dicussed below, MLD provided us with a new version of the data for the period Jan, 2006 - Jan, 2008. At the end of July, 2009, the old versions of the files were replaced with the new ones.

In October, 2008 the Meteorological Development Lab (MDL) performed a re-edit of the U.S. and Canadian Surface Hourly Data (ds472.0) for the period Jan, 2006 - Aug, 2008. The results of the process were :


December, 1996

The below message was received in December, 1996.

Dennis,
The machine I use to edit the hourlies (the Mainframe at NCEP) is
being removed in about 3 weeks, so I can guarantee that the editing
through November will be finished by then.  I should warn you that
some of the data may not be reliable.  In July 1996, the US went
to a new reporting format (METAR) and not all of the software that
coded up the data in the field (especially at FAA and DOD sites)
worked as expected.  This caused many problems that the decoders
did not handle very well here at headquarters.  Many of these problems
are in the temperatures, the precip values, and the present weather.
Our quality control program will catch some of the problems, but
our codes were not written with METAR in mind either.

When all of the processing is finally converted to the CRAY machines,
new codes which are being written now should give us reliable data again.
The CRAY data is not going to be packed up in the same format that it
is in now.  When we send you December's data and everything after that
we will also have to send you new software to read the data.

Kathy Hughes
TDL