Minutes for Telecon for June 04, 2002

Present: Churchwell, Watson, Meade, Stolovy, Jackson,
Babler, Clemens, Mathis, Kobulnicky

Next telecon: Tuesday June 25, 2002, 11 am CDT, 608-265-1000 #4112



SSC Review - Ed

  Reliability  

   Tom Soifer wants a reliability of 99.5% or higher for
   our catalogs.  We would have to change our minimum source flux to
   achieve this in our catalogs.  We could also have a database 
   (not a catalog) with a more complete source list with all the flags
   and warnings.  "catalog" needs to be high reliability.  "database"
   does not have to be as high. They suggest the GBSC have a 100%
   reliability and the GPSC 99.5% reliability. The "database"
   would be best effort to 5 sigma, background limit or confusion
   limit (whichever comes first) plus all the associated flags. They argued
   that people won't pay attention to all our flags. Should we trade
   reliability for area?  NO. Currently we have small number statistics.
   We need more runs. We need to "have our bowling pins in order" since
   this issue will be raised again.

  Delivery schedules

   The delivery schedule of the GBSC every 3 months might tax the team
   too much...this was the sentiment of Roc Cutrie, of 2MASS.  He says the
   step from source extraction to catalog entry is fraught with 
   difficulty.  Ed asked Roc to send us a list of issues they ran into
   with 2MASS.  We could change the GBSC deliveries to every 6 months.
   We do want to provide data to the community as early as possible.
   We should look at this over the summer and evaluate our difficulties
   in going from source extraction to catalog entry.

  Tightness between validation obs and our first scheduled observations.
   
   We emphasized that we needed IOC data for modelling the banding. 
   And we need to obtain our validation observation data early in the
   Legacy science verification observation window so we can be ready
   to observe our first segment in May.
   How far back can we push our first observation?  It is critical
   to get the data from the first segment to test the pipeline and further
   validate our observing program.
   Christer says we can do the 54.5 -67.0 deg segment at the end of
   May 2003 (~May 30-June 1). 

  Completion of the GLIMPSE survey.

   Tom Soifer wants to have all the Legacy Science observations done by
   L+12.  The current schedule has our survey finishing observing at
   L+16.  We would have to double up on the segments observed in
   August, September & Oct 2003 in order to finish by L+12.  We would
   actually be finished by L+9.  But in order to do this, we would
   be taking more than 50% of an IRAC campaign, which supposedly is
   not allowed. From the beginning of August 2003 to the beginning
   of October, we'd be taking all but one of our segments.  Can we get
   half of an IRAC campaign?  Ed will check on this.  After validation,
   we can change our scheduling strategy if need be.  We can try to 
   get all the data taking done by L+9 if we can get our May 2003 data
   and reduce it OK.  Otherwise we would not be willing to do that.

  Mosaicer 

   Susan sent us the source code and executables for the SSC mosaicer.
   We will also look at the IRSA mosaicer (MONTAGE) which should be
   available in July. The IRSA mosaicer is slower than SSC's.
   The SSC mosaicer works best if there are 20 pixel overlaps between frames
   for background matching.  That is more of a recommendation than a
   requirement. We are doing 12 pixel overlaps.