[IGS-DCWG-22] Summary of IGS DCWG Meeting in Berne

[Carey Noll]

******************************************************************************
IGS-DCWG Mail      18 Mar 06:52:18 PST 2004      Message Number 22
******************************************************************************

Author:  Carey Noll/CDDIS

We had a very busy 1.5 hour meeting in Berne on March 2.  I will attempt
to summarize our discussions and plans.  We covered two topics during
the meeting:

1.  Long-term archive of raw data.
Yehuda Bock reported that SIO keeps raw data on-line and users can
download these data for reprocessing anytime.  However, others pointed
out that in general raw data can be difficult for users to handle since
metadata (e.g., antenna height, etc.) need to be applied properly.  It
was therefore recommended that operational data centers keep the raw
data, rather than archiving them at a regional or global data center
level.  Archiving of raw data is useful for two major reasons.  Errors in
RINEX conversion have occured and require the raw data to correct.  Since a
majority of sites provide data to the IGS at a 30-second sampling rate, if
higher rate data are required (i.e., to study past or silent earthquakes),
raw data would need to be used.
Recommendation:  Operational data centers should archive raw data 
indefinately.  These data do not need to be on-line or accessible to
IGS users but should be available upon request.

2.  Handling of replacement data.
This topic addresses how to archive replacement, retracted, or data 
transmitted with a large (e.g., months) delay. We need to indentify
what type of problem/change requires a replacement.  When a replacement
is issued for a previously-released data set, the archive and the end
user need to be notified.  We also need to have a way to distinquish the
old version of the data from the new version (using a time tag/size of
file/md5sum/etc.). We came up with several possible solutions:
  a.  Establish a new email list and  have ODCs issue emails to this list
  using a prescribed template.  
  b.  When a replacement is made, archive the old file in an old
  subdirectory (could be within the daily directories) at the RDCs and
  GDCs.  Use an email mailing list as in a) as well.
  c.  Send a signal/flag file (e.g., SSSSDDDD0.msg) with a replacement data
  set that is archived with the data; presence of this file would indicate
  to the archive and the user that a new replacement file has been issued.  
  Use an email mailing list as in a) as well.
In all cases, an an on-line listing of data revisions with the replacement
information should be maintained. We all agreed that this was a difficult
topic.  UNAVCO will be looking into LDM more closely in support of PBO and
this effort may be of use to the IGS data flow in the future.
Recommendation:  Circulate possible solutions to the DCWG and decide
on a plan for handling the archive of replacement data.

I would like to get the opinion of working group members on these two
topics and generate some more "formal" recommendations for our group
that can be tested in the near future.

Regards,
Carey.

+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
Ms. Carey Noll  Manager, Crustal Dynamics Data Information System (CDDIS)
                                           Secretary, ILRS Central Bureau
Code 922                                   E-mail:  Carey.E.Noll at nasa.gov
NASA GSFC                                           Voice: (301) 614-6542
Greenbelt, MD 20771                                   Fax: (301) 614-6099
USA                  WWW:  http://cddisa.gsfc.nasa.gov/cddis_welcome.html
+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+