From jwf@ipac.caltech.eduTue Mar 25 09:37:40 1997
Date: Thu, 30 Jan 1997 13:06:35 -0800
From: jwf@ipac.caltech.edu
To: 2mass@ipac.caltech.edu
Cc: chas@ipac.caltech.edu, sstrom@ipac.caltech.edu,
    stiening@ipac.caltech.edu
Subject: WG Mtg #113 Minutes

              IPAC 2MASS Working Group Meeting #113 Minutes                    
                                1/28/97

Attendees: R. Beck, T. Chester, R. Cutri, T. Evans, J. Fowler,
           L. Fullmer, T. Jarrett, G. Kopan, B. Light, H. McCallon,
           S. Wheelock, J. White

AGENDA

1.) 2MAPPS 1.0
2.) Camera & Telescope Status
3.) Association Parameters
4.) Tycho Proper Motions


DISCUSSION


1.) 2MAPPS 1.0

    R. Cutri reported that the date for delivery of 2MAPPS 1.0 is 5 February
(not 3 February, since normal deliveries are on Wednesday). The purposes of
this version of 2MAPPS are: (a.) to provide quick turnaround of processing of
three-channel data; (b.) to check runtimes; (c.) to develop quality assessment
capability.
    A poll of the subsystems was taken to determine significant liens expected
in this delivery of the system. This is summarized below, where the minor
tweaks needed by all subsystems are omitted.

     POSMAN: the position uncertainty model is unfinished; catalog
     associations (other than those involving astrometric catalogs) are not
     yet implemented; a problem with Tycho proper motions is unresolved (see
     item 4); a problem with high-density regions has just been discovered
     and has not been solved [note: this has since been determined to be a
     failure of the unix system sort routine; the file to be sorted was too
     large for the available workspace; PFPREP will be modified to do this
     sort internally].

     PROPHOT: the full set of PSFs has not yet been generated (this will
     require a substantial amount of work with real three-channel data).

     BANDMERGE: statistical processing has not yet been implemented (this may
     possibly be done by delivery time).

     QUALITY: not quite all CONMAN functions have been implemented; most
     other QUALITY functions are in a very preliminary stage, as are most of
     the input files generated by other subsystems.

     DBMAN: just starting development.

     MPCAT: coding not yet begun.

    R. Cutri requested all cognizant engineers to send him a lien list via
email by the Friday after delivery, February 7.


2.) Camera & Telescope Status

    R. Cutri reported that the camera is scheduled to be shipped to Tucson in
early February, and the telescope is expected to be online in mid February.
There will be a telescope shakeout period of unknown duration, but we should
expect the first real three-channel data as early as late February.


3.) Association Parameters

    T. Evans reported that the catalog association parameter in the SIS POS05
is inadequate. The association field is too small, and other parameters are
needed, such as the probability of the association being correct. She will
propose specific changes to the SIS when she has determined exactly what
changes are needed.


4.) Tycho Proper Motions

    H. McCallon reported that when proper motion was included in Tycho
positions used for pointing reconstruction, the excellent results reported
last week were degraded, especially in RA. Since the Tycho position epoch is
close to that of the protocam observations of M67 (i.e., about 4 years
difference), neglecting proper motion had little effect on the quality of the
solution for the test scan; the residual standard deviation was 0.08 and 0.10
arcsec in RA and Dec, respectively. Including the proper motion should have
made a very small improvement, but instead the residual standard deviations
degraded to 0.43 and 0.15 arcsec in RA and Dec, respectively.
    Considerable effort has been expended on reexamining the implementation of
the proper motion corrections, and no errors have been found. To investigate
further, 2020 entries in the sample Tycho data set were position-matched with
entries in the PPM catalog, and comparisons of the proper motions were made.
Although the difference in epochs is about 60 years, the proper motions
themselves were not expected to have changed very much in most cases. 
    Scatter plots of the proper motions in RA and Dec were generated, along
with correlation coefficients. The largest few percent of the proper motions
correlated relatively well, but the smaller ones matched rather poorly, with
an overall correlation coefficient of about 0.5 on both axes. The fact that
the largest proper motions matched well would seem to argue against a problem
with the implementation. The time baseline for Tycho proper motions was rather
small (about a year in orbit), so perhaps the motion of the slower moving
stars is not well determined (but by the same token, should not matter very
much). An indication of this is the fact that some entries have relatively
large uncertainties, with a mean uncertainty (per year) more than six times as
great as for the PPM. It is possible that the sample catalog has some
preliminary values which will be corrected when the full catalog is available.
Some discussion with the catalog authors will be pursued to see what course to
follow.