From jwf@ipac.caltech.eduTue Mar 25 09:38:05 1997
Date: Wed, 12 Feb 1997 18:11:27 -0800
Subject: WG Mtg #115 Minutes

              IPAC 2MASS Working Group Meeting #115 Minutes

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


1.) Status Reports


1.) Status Reports

     A.) R. Beck

          Software deliveries were processed and pipeline tests were run.

     B.) R. Cutri

          The telescope optics are due to be installed about now. First
          light is expected 2/17. The 2MAPPS runtimes are about double their
          FDD guidelines.

     C.) T. Evans

          MAPCOR liens at this point include checking all output for bugs
          and input parameter optimization. Statistical and quality output
          are currently being added. Work is underway on tracking down bugs
          in the pipeline output. Work on DBMAN has begun.

     D.) J. Fowler

          The DARKS code and SDS have been modified to implement changes
          suggested by M. Skrutskie and E. Howard. PFPrep has been modified
          to limit the number of FREXAS detections per frame on the basis of
          brightness so that the brightest detections will be used and no
          panel bias will be introduced. BANDMERGE statistical processing
          design has been finished and coding is underway.

     E.) L. Fullmer

          CONMAN was redelivered with more match statistics. Plots of these
          statistics have brought to light a precision problem in another
          subsystem. Investigation is still underway concerning the validity
          of some extractions that look like they may be "split".

     F.) G. Kopan

          A new noise estimator was implemented in the PIXPHOT find step.
          Bugs were fixed in PIXPHOT and FREXAS. An investigation is
          underway of the PIXPHOT FIND completeness vs. FWHM. Work was done
          in support of the survey scheduler design and the survey database
          manager design, including creation of the first version of
          database manager SIS srv01.

     G.) H. McCallon

          I. PFPrep Status:

               1) Changes were made to eliminate system sort on very large 
                  files; the system sort uses scratch disk space and could 
                  cause problems in high-density fields if insufficient disk
                  space were available. PFPrep now uses an index sort which 
                  requires much less space and is done in core; this is both
                  safer and faster.
               2) Changes were made to limit number of Read2-Read1
                  extractions per frame based on magnitude sort.
               3) All stop statements have been replaced with exit calls.

          II. PosFrm Status:

               1) Capability to use the samples of real Tycho data has been
               2) Efficiency of position matching has been improved.
               3) All stop statements have been replaced with exit calls.
               4) A number of the routines borrowed from SlvAll (in the
                  haste to have something pipeline-ready for the Dec. 3rd 
                  delivery) have been rewritten for maximum robustness and 
                  maintainability; many more to go.

     H.) S. Wheelock


               Version 0.40 of CALMON was delivered Feb 5. This version has
               the major functionality of CALMON to determine the zero
               point via a nightly average or linear fit based on RMS
               values. The 4-hour fit is computed, but the logic concerning
               when to use it is TBD. Some major SIS changes occur with
               this delivery. Standard Star files now carry a version ID to
               assist in distinguishing which set of standards for a field
               were used in determining calibration. The calibration
               parameter file cal03 contains more information regarding the
               nightly calculations. Additional information includes which
               algorithm was used in determining the scan zero point and
               which standard star file was used.


               Utility SMERGE was modified to work with 3-channel data and
               used in the C&R analysis described below.


               The internal completeness and reliability analysis for k
               band only using calibration scans 022-027 is complete. The
               results are encouraging for the 2MAPPS processing but do
               highlight some processing problems which are stated below.
               In addition to the internal scan-to-scan C&R analysis, all
               6-times-confirmed sources from the 2MAPPS processing were
               matched with 6-times-confirmed sources from the protocam 
               data as well as the deep field coadd created by Roc. All
               non-matches were examined. All problems found in this
               analysis have been passed on to the responsible party.
               Problems found in performing this analysis are as follows.

                    Sources that fell at the end of descending scans and
                    within the short coadd were missed.

                    Some sources that were 6-times-confirmed in the
                    protocam processing had psfmag=99.9 in the 2MAPPS
                    processed data.

                    One source appears to have been inappropriately

                    Many R1 single detections occurred at faint R1

                    Some sources had multiple detections within a single

     I.) J. White

          Worked was done on compression and cascading. No final decision
          has been made regarding its use. The reduced total cost of tapes 
          may not be cost effective compared to implementation and
          maintenance cost. TAPELOAD was delivered as is; it is still not
          unit tested and needs observatory data. COPYCAT (raw disk to tape)
          was delivered as is; it is still not unit tested and also needs
          observatory data. RDFRAME was redelivered with a new center frame
          computation EQSPDX subroutine from H. McCallon. A minor change to
          Obs/IPAC I/F Document Rev J was requested: add Scan Direction to
          par file. A couple of suggestions regarding new fields added by
          the observatory were passed along to Mike Rudenko.