IPAC 2MASS Working Group Meeting #102 Minutes

IPAC 2MASS Working Group Meeting #102 Minutes 9/03/96

Attendees: R. Beck, R. Cutri, T. Evans, J. Fowler, T. Jarrett, B. Light, H. McCallon, S. Wheelock


  1. Status Reports
  2. File Name Templates


  1. Status Reports --

    R. Cutri: A telecon with NASA HQ was held; the NASA HQ 2MASS Program Manager, Paul Swanson, will be located at JPL under the new distributed management scheme, as will the NASA HQ Program Scientist, Mike Bicay.

    The Mt. Hopkins mirror has been polished to half-wave accuracy at 6000 Angstroms and will soon be sent to a firm in the East for coating. The camera alignment is almost finished for the northern hemisphere observatory.

    H. McCallon: The SLVALL code written to demonstrate the feasibility of the key POSFRM algorithm has been successfully run on all seven scans of the low density test field in Coma, and the results have been approved by the cognizant scientist. Projection to a 3-band situation indicates that we should be able to meet the requirement even in this near worst case situation. During the course of this test a great deal was learned which should be of benefit to POSFRM development. Work on the designated high density test field is well underway, and the results, so far, look very good.

    J. Fowler: The band filling subroutine BFILL, which will be used by GALWORKS to handle point source I/O and band filling, has been written, along with SIS's GAL03 and GAL08, the GALWORKS/BFILL interfaces and the band filled point source output list, respectively.

    Work has begun on modifying PFPREP to use dynamic allocation and to install several other modifications. The SIS POS05 describing the position reconstructed point source list has been written, along with a dummy POSPTS program that should support the version 0.1 2MAPPS pipeline; this POSPTS (version -0.8) will read the output from BandMerge (SIS BMG01) and apply the U-Scan/J2000 transformation from POSFRM (SIS POS01), generating the output file (SIS POS05) used as input by GALWORKS.

    SIS's MPC01 and MPC02 have been written, along with a draft SIS MPC03. MPC01 defines the interface between the asteroid/comet orbit computation program and the 2MAPPS shell that supplies it with scan information; MPC02 defines the output file of the orbit program. MPC03 defines the formats of files involved in asteroid/comet associations with 2MASS point sources.

    PIXCAL/RDFRAME/DFLAT/FREXAS was delivered with the following modifications: (a.) Obs/IPAC Rev J.1 compatibility was installed; (b.) hybrid sky offset algorithm was included, defaulted to pure RMDR (recursive median distance rejection) trimming 20 of 42 samples; (c.) Read1 flattened frames are corrected with an additive constant that forces the median to the median of the corresponding Read2-Read1 frame scaled by the ratio of exposure times.

    R. Beck: Writing the EXEC SDS and running pixcal tests.

    T. Evans: Coding MAPCOR.

    T. Jarrett: Coding GALWORKS.

    S.Wheelock: Worked with J. White on the CALMON/EXEC interface. The issue concerning where to obtain the final airmass value for each survey scan was discussed. FRX02 was suggested as a possible place for calmon to find airmass information. Another option mentioned was for EXEC to pull the information off for CALMON from the PAR file ( this was discouraged by JWhite). Using the StrpID field of the LGO file to house a unique field identifier for the calibration scans was also discussed. This would require some interfacing with UMASS, since they are the ones who create the LGO file. CALMON will use two dedicated subdirectories under a CAL subdirectory of the date hemisphere directory, one used to calculate the calibration parameter values, and the other used to apply the calibration to the survey scans. CALMON will be written in c. To get started, a version of CALMON that uses the protocam data has been coded. The relevant SIS's have been examined, and this week the focus will shift to working with the PCP data formats.

    B. Light: Debugging activities continue on the PROPHOT subsystem. The current code is producing photometry, but the consistency with prior KAMphot runs is not what it should be. That is, elevated chi-square and error values are seen, in addition to occasionally reaching the limiting number of iterations in the solution. While tracking down the peculiarities, two problems with the code were corrected.

    The first "bug" was a loophole in the local chi-square minimization routine that allowed a solution to proceed to the iteration limit while having already converged to a reasonable solution. That has been cured by putting in an additional "delta chi" test.

    The second, more important bug was introduced when making PROPHOT take an external coordinate list. As initially programmed, PROPHOT took the best estimate for the center of a point source (passed from PIXPHOT) and used that as the starting point. However, the estimate of the position of the point source being solved for needs to be to the (low X, low Y) side of the profile; this is an artifact of KAMPHOT and the matched filter grid. As the chi-square minimization routine steps across the center of the profile, it will find the best fit. Thus the PIXPHOT coordinates need to be moved slightly to the "low" side so that the code taken from KAMphot can find the proper solution. This fixup cured a large fraction of the photometric problems seen.

    The photometry is now better behaved, relative to KAMphot, but still has potential problems for a small subset of stars. These stars will be traced through their processing completely, both via KAMphot and PROPHOT, to learn why they are deviant.

    J. White: TAPELOAD is being modified and is working but needs some additional mods and testing to handle the new RSC compressed raw data tapes. A DISK2RSC c-shell script is written and is being tested to archive raw data previously written to disk by TAPELOAD. The RSC format meets the Rev J.1 mountain format with a few necessary exceptions, including: integral nights are stackable and the binary scan files are compressed. Raw data archive compression will be done by the tape drive hardware but requires preconditioning s/w which has not yet been worked on. The real 2MAPPS pipe has been used with a call from the real EXEC to process the 2nd UMASS "Fake Tape" through RDFRAME using a phony stub emulating the PIXCAL interface. A new RDFRAME upgraded to the Rev J.l (post Fake Tape #2) has been tested, delivered and incorporated into PIXCAL. No real data (Rev J.1) have yet been received from UMASS or the Observatory. Two "Fake Tapes" have been received and processed.

  2. File Name Templates -- It was pointed out that not all SIS's for disk files specify the file name templates. All subsystem cognizant engineers are requested to add such templates to any SIS's where they are currently missing. The following example of a file name template is taken from the SIS MAP01 and illustrates the basic idea of what is needed by a program that must construct a file name and open a file passed from another subsystem.

    " The name of the output file will be of the form: 
    [scan number].ph.mc example:  k027.ph.mc"
  3. QUALITY Q&A -- Next week L. Fullmer will answer questions regarding how subsystems interact with the QUALITY subsystem, such as what directories files should be written in, what information is needed for QUALITY to make use of the information, etc.