Date: Wed, 20 May 1998 17:18:59 -0700 (PDT)
To: 2mass
Subject: IPAC 2MASS WG Mtg #153 Minutes
Cc: chas, stiening, bgreen
Content-Length: 4490
Status: RO

           IPAC 2MASS Working Group Meeting #153 Minutes

Attendees: T. Chester, R. Cutri, S. van Dyk, D. Engler, T. Evans,
           J. Fowler, L. Fullmer, D. Kirkpatrick, G. Kopan,
           J. Mazzarella, H. McCallon, B. Nelson, B. Wheaton,
           S. Wheelock, J. White


1.) Galaxy Orientation
2.) Quality Assessment Calibration
3.) K Magnitude Anomaly
4.) Short-Form Data Access


1.) Galaxy Orientation

    T. Chester reported that the preferential orientation of
galaxies reported in last week's minutes has been traced to the
PSF behavior, as expected. Variations in the PSF derived on a
scan-by-scan basis from flattened frames (as opposed to the PSF
models used by PROPHOT, which are derived over many more scans
and modeled as functions of the seeing) show a convincing
correlation with the galaxy preferred-orientation effect. Details
of the analysis may be viewed on the web at the URL addresses:

What corrective action to take, if any, remains undecided. It was
pointed out that offline analysis should be done without any
presumption that new pipeline software would be the result,
although that possibility should not be ruled out. Among the
suggestions were (a.) smoothing of galaxy images with a
circularizing kernel; (b.) deconvolution (or other method of
removal) of the elliptical PSFs, possibly with post-processing
smoothing; (c.) simply reporting caveats derived from the
observed PSFs on a scan-by-scan basis. Investigation will
    T. Chester reported that a by-product of this study was the
observation of the fact that the scatter in the PSF shape
parameter increases considerably when a sub-optimal dither
pattern in the frame-to-frame offsets is in effect. The
deterioration is bad enough to call into question whether the
data taken early in the survey before the best dither pattern was
achieved should even be considered for use in the data products.
What effect the bad dithering should have on the quality score is
still under discussion.

2.) Quality Assessment Calibration

    D. Kirkpatrick and R. Cutri reported that although progress
has been continuing in calibrating the quality assessment
parameters, new issues have arisen, and the schedule for
finishing this activity reported in last week's minutes has
turned out to be optimistic. Some disagreement exists about the
extent to which the quality scoring should be automated. A
telecon with UMASS is scheduled for tomorrow (Wednesday, May 20).
As stated previously, the goal is to calibrate quality on a
nightly and scan-by-scan basis, and to be able to update the tile
data base to make optimum use of rescanning opportunities. The
scoring takes the form of multiplying a base value of 10 by the
product of about half a dozen factors normalized to the range
from zero to one. These factors include photometricity, seeing
variations, and background levels, among other things; factors
for extended source quality are needed. It was suggested that bad
dithering should be represented by setting the base value to 5.

3.) K Magnitude Anomaly

    S. Wheelock reported that an anomaly involving K photometry
has been found. In five well studied nights, two contained cal
scans in which the following strange behavior was observed: in
three-band sources with none of the usual problem diagnostics
turned on, the K magnitude on one of the six overlapping scans
was about two magnitudes fainter than on the other five scans.
This happened to four sources in the two nights that showed this
anomaly. It has not been seen in J or H. Any ideas on what might
be causing this are solicited, and B. Wheaton, S. Wheelock, and
R. Cutri will continue to investigate it.

4.) Short-Form Data Access

    R. Cutri requested T. Chester and T. Jarrett to develop a
short-form representation of the extended-source data record.
This is needed to go along with the point-source short-form
record. One immediate need for these data formats is for use when
the data base is locked because a loading operation is underway.
Roc also requested J. Mazzarella to arrange for an automatic (or
at least convenient) switchover to the short-form data products
for people attempting to access the data base when it is