From jwf@ipac.caltech.eduTue Mar 25 09:37:53 1997
Date: Thu, 6 Feb 1997 13:24:49 -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 #114 Minutes

              IPAC 2MASS Working Group Meeting #114 Minutes                    
                                2/04/97

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

AGENDA

1.) 2MAPPS 1.0
2.) Pixel Linearity
3.) Chi-Square Improvement From Deblending
4.) Galaxy Photometry
5.) Association Parameters


DISCUSSION


1.) 2MAPPS 1.0

    Although the delivery for 2MAPPS 1.0 software is 5 February, R. Cutri
stated that validation will require more than a few days of testing and
requested suggestions for test scans. He listed the scans currently slated for
processing by the new pipeline, all from 950503: six M67 scans, one Abell 26D5
scan, and five MSX scans. Suggestions for additional scans should be reported
to R. Cutri.


2.) Pixel Linearity

    J. Fowler reported that some email discussions involving pixel linearity
have taken place. Participants include R. Cutri, J. Fowler, E. Howard, G.
Kopan, and M. Skrutskie. Past studies of nonlinearity were primarily concerned
with soft saturation at high illumination levels, and the use of Read1
photometry alleviated such concerns. The new issue concerns deviations from
linearity by several percent over the entire dynamic range. Possible
nonlinearity calibration, correction, and monitoring techniques are being
studied, primarily by E. Howard. The group concluded that until definite
software modifications can be proposed, 2MAPPS will proceed independently of
these considerations. It appears that the most likely impact would be in the
RDFRAME module (maintained as part of the TAPELOAD subsystem and used by DARKS
and DFLAT), where the process of converting data frames from 16-bit integer DN
to 4-byte reals could be made to include a nonlinear transfer function. This
would make the process transparent to the downstream software. There would
also be operational considerations at the observatory (e.g., periodic special
calibration procedures) and perhaps some offline analysis software at IPAC.
Details await further study.


3.) Chi-Square Improvement From Deblending

    B. Light reported that the PROPHOT deblending algorithm reduces the
chi-square figure of merit in the point-source extraction process, but
currently the original chi-square value is not tracked by the system, and the
point sources created by the deblending process are not associated
unambiguously. He suggested that a new file be generated which provides this
information, which might be useful for galaxy processing and should be
included in the point source data base. The group decided that a separate file
would be appropriate rather than additional columns in the main point-source
file structures, since the latter would increase substantially in size with
columns that are irrelevant for most sources. B. Light and T. Evans will work
out a SIS for a file interface between PROPHOT and DBMAN. This file will
contain information such as the original source ID numbers assigned by PROPHOT
and the chi-square values before deblending and after each deblending step.



4.) Galaxy Photometry

    T. Jarrett reported that he and J. Rosenberg have just completed an
extensive study of galaxy photometry. This study used SA57 scans in all three
bands. A memo describing the study will be distributed very soon.


5.) Association Parameters

    T. Evans, H. McCallon, and R. Cutri have finished iterating the design of
the updated catalog association parameters (see last week's minutes). H.
McCallon will issue a new version of the POS05 SIS after the 2MAPPS 1.0
delivery activities have settled down. Since several subsystems will be
affected by the SIS changes (GALWORKS/BFILL, QUALITY/CONMAN, CALMON, and
DBMAN), and since the current POSPTS processor is a substitute that just
transforms source coordinates from U-Scan to J2000, without making any catalog
associations, the SIS changes will not be implemented until the real POSPTS is
delivered.