2MASS V3 Production Processing

Read1 Source Magnitude Sigmas

Summary

For the Read1 sources, the magnitude sigmas (?_msig) are computed in POSFRM during merging of the single frame exractions. The combined magnitudes are computed from the average of the selected magnitudes (preferentially the unsaturated aperture measurements) and the sigmas are computed from the standard deviations of the measurements used.

Rae Stiening has produced a histogram of the cmsigs for rd_flg=1 and ph_qual=A sources in the catalog. This plot shows a significant population of sources with cmsig values > 0.1 mag.

Since there is a considerable population of low snr problematic R1 sources that are single band, I looked at that subset of Read1 sources that are 3 band and have ?_msig> 0.1 in any band: 

  sources    criteria
  3365       j_msig > 0.1    
  9104       h_msig > 0.1    
 11060       k_msig > 0.1    
 16365       ?_msig > 0.1
Note there are a considerable number of sources with multiple bands exceeding the msig criteria.

Histograms of prox and magnitude for these sources:

  • 3 Band Sources with j_msig > 0.1
  • 3 Band Sources with h_msig > 0.1
  • 3 Band Sources with k_msig > 0.1

    It can be seen from the above plots that almost all of these 3 band R1 sources with msig > 0.1 are confused with another source within about 6". In addition, most of these sources are near or below (R2-R1) saturation. On inspection, nearly all of these sources are marked as photometrically confused as well as having high msig values. They are also heavily concentrated in the galactic plane, and inspection of the quicklook images shows them to be close doubles or otherwise confused.

    Also note in the h_msig plot, there is a small population near h_m of 5 due to the higher intrapixel response dispersion in H and low counts of unsaturated frames. The intrapixel dispersion also explains the fact that the msigs are higher in general in H band than those in J and Ks.

    Msig Flag Values

    During merging of the single frame detections and computation of the R1 msigs, certain values of msig were used as flags. The catalog entries containing these flags values are summarized in the following table: 
      msig   #sources    #j   #h   #k
   8.888    437      108  149  182  - R1 saturated in all but 2 frames
   9.995   2597       24  413 2220  - template flux only (mostly bad Ks column)
   9.996     39       14   12   19  - single frame R1 saturated (merge orphans)
   9.998   8931     1357 3318 4638  - single frame R1 unsaturated (merge orphans)
    The majority of these sources with msig>9.99 are also fainter than the (R2-R1) saturation threshold. The sources flagged 9.996 and 9.998 are single detections orphaned during the detection merging primarily due to confusion. The sources flagged 9.995 in Ks are almost all on the edge of a very noisey column in the Northern Ks array that was masked off in the processing. The aperture photometry failed due to masked pixels in the aperture. These are all of poor photometric quality.

    The sources flagged 8.888 are almost all on the R1 saturation threshold. They had unsaturated R1 measurements on 2 frames, but were flagged 8.888 due to having too little dispersion between the 2 measurements. These are probably all good photometric measurements. It should be noted that since the R1 sigmas are standard deviations, there will be statistically low values, especially likely in cases where low numbers of measurements were available.

    This page last updated on Dec 6, 2002.

    Gene Kopan - gene at ipac.caltech.edu