Date: Mon, 27 Oct 1997 16:52:49 -0800 (PST)
From: jwf@ipac.caltech.edu
To: 2mass@ipac.caltech.edu
Cc: chas@ipac.caltech.edu, sstrom@donald.phast.umass.edu,
    stiening@ipac.caltech.edu
Subject: 2MASS WG Mtg #134 Minutes

           IPAC 2MASS Working Group Meeting #134 Minutes
                            10/21/97

Attendees: R. Beck, R. Cutri, T. Evans, J. Fowler, L. Fullmer,
           D. Kirkpatrick, G. Kopan, J. Mazzarella, B. Nelson,
           S. Wheelock, J. White


AGENDA

1.) MAPCOR Tuning
2.) Optical Catalog Association Parameters
3.) Telescope and Camera Status
4.) Science Team Meeting Agenda
5.) Quality Assessment Issues
6.) Minor Planet Processing
7.) Sample Data Base


DISCUSSION

1.) MAPCOR Tuning

    R. Cutri reported that the top-priority parameter tuning
target at this time is MAPCOR. The current persistence rejection
is very good, but considerable work needs to be done on glints,
diffraction spikes, and fractured bright sources.


2.) Optical Catalog Association Parameters

    J. Fowler reported that a study of the IAU Colloqium 179
poster writeup "The OPTID Database: Deep Optical Identifications
to the IRAS Faint Source Survey" (Lonsdale et. al., August 1996)
suggests that the likelihood ratio parameter defined therein would
not be appropriate for use as one of the 2MASS optical-catalog
association parameters. The main reason is that the paper states
that this parameter follows very different distributions for
different classes of objects (e.g., bright stars vs. faint
galaxies), requiring some prior knowledge of the object type for
interpretation. Furthermore, the parameter was used in a way that
is dissimilar to the 2MASS context, namely multiple matches were
retained, with a full set of identification parameters for each
possible match, whereas the 2MASS associations will retain a
single match as an association. The two contexts are quite
dissimilar primarily because the IRAS objects had position
uncertainties large enough to permit multiple associations under
source-density conditions that would be considered sparse by
2MASS standards. Since the number of catalog objects within the
coarse window (nominally 3 arcsec radius) will be an association
parameter, and since the radial separation (and orientation
angle) are also included, the group decided that sufficient
match-quality and confusion indicators were included in the
current design.


3.) Telescope and Camera Status

    R. Cutri reported that the southern-hemisphere telescope had
been damaged in shipment. The mirror is OK, but the mounting
needs significant repair work. The possibility that the repairs
can be performed at a machine shop in La Serena is being pursued
as the most desirable course of action. There is a chance that
shipment back to the U.S. or replacement hardware will be needed.
All repair and shipping costs should be covered by insurance.
    [Note added in proof: a new part will have to be fabricated in
     the U.S. and air-freighted to Chile.]
    The lab data tape for the southern camera has arrived at IPAC
and will be run through the DARKS subsystem. There is concern
that the K array will need to be replaced, so the results of
timely IPAC analysis of these data are needed at UMASS.


4.) Science Team Meeting Agenda

    The Science Team meeting in Boston on November 5-6 (not 4-5,
as stated in last week's minutes) will include presentations by
all 2MAPPS cognizant scientists on the parameter tuning status of
their respective subsystems. A splinter group will discuss
photometric acceptance criteria. 


5.) Quality Assessment Issues

    R. Cutri reported that the desire for a boiled-down two-or-
three-numbers quality characterization for each scan still
exists, and work on designing this will continue. Offline
analysis will also pursue the relationship between the "pfrac"
parameter and PSF FWHM. The pfrac parameter is computed for each
band-scan by the FREXAS module; it characterizes the fraction of
light from a point source contained within a single pixel. The
smaller this number is, the larger the azimuthally averaged PSF
is, but the exact relationship has not been calibrated in detail.
Doing so is desirable in general and would be used in MAPCOR and
for interpreting GALWORKS results.


6.) Minor Planet Processing

    J. Fowler reported that the MPCAT (Minor Planet Catalog)
subsystem is now activated in the 2MAPPS pipeline. Several
asteroids have been matched, and resulting analysis has revealed
some residual bugs in the association program MPFIND. It also
became clear that additions to the MPC03 SIS were desirable,
specifically, the addition of the RA and Dec offsets between the
2MASS point source and the predicted minor planet position. This
change has been made.


7.) Sample Data Base

    T. Evans reported that most of the 11-night sample data base
has been generated. Scans with about 5000 point sources require
about 45 seconds to load into the data base, whereas scans with
about 30,000 sources require about 15 minutes. The complete data
base contains about 6 million sources, with duplicated entries
for multiply scanned regions (mostly calibration scans). There
are 15 indexes per source, and each index takes about 25 minutes
to be generated. Unindexed queries take about 30 minutes;
currently indexed queries take the same time because of a bug
that is not yet found.