[data-issues] Mis-timed BUD data from EarthWorm feeds on the bud.iris.washington.edu machine

[Rick Benson]

  This email is intended to bookmark a known problem that exists
with BUD real-time data that was processed using a version of
ew2mseed software between July 20 and Sept 11, 2006, on one of our
real-time collection machines.

  What is known is that data is mis-timed by approximately 22 seconds,
making the data early. Therefore, phases that are picked from these  
stations are going
to be approx 22 seconds late.

  We are trying to determine the best mechanism to detect this from a  
list
of 569 stations that are known to have passed through this  
transformative
software.

  A complete description of the problem will be posted on the http:// 
www.iris.edu
webpage, and the http://www.iris.edu/data/data.htm (and anywhere else  
we can find
that protects innocent users) and a corresponding update to this list  
as well.

  As an illustration of the problem, the picture included here was  
created using the WILBER pseudo-
record section software which plots the predicted travel times for  
prominent phases.
What's important to note that ONLY SOME STATIONS exhibit this time  
problem in this plot, like
stations RSSD.US, HWUT.US, BW06.US, etc.

Other stations that appear correct went through different software on  
a different machine.

Please address any questions or observations to this list if you  
become aware of something that
contradicts this announcement.
Thank you,
  Rick Benson
  IRIS DMC