Explanation of the DIRBE Point Source Photometry Tool

This interactive service permits measurement of photometry from DIRBE time-ordered data for sources much smaller than the 0.7 x 0.7 degree DIRBE beam. As described in section 5.6.6 of the DIRBE Explanatory Supplement, time-ordered data must be used to obtain the most accurate point source photometry. Measurement of point source photometry from a DIRBE skymap underestimates source fluxes. The error is on the order of 10%, and varies considerably from source to source and from band to band depending on factors such as the beam shape, the distribution of scan position angles at which the source was observed, and the location of the source relative to the DIRBE pixel boundaries. (The DIRBE spectral energy distribution browser can be used to obtain quick estimates of the flux of a point source and the dispersion of sky intensities at nearby reference positions from skymap data.) The point source photometry tool makes use of the DIRBE Calibrated Individual Observations (CIO) files, which contain the calibrated individual 1/8-second data samples taken in science-survey mode during each day of the cryogenic mission (11 December 1989 - 21 September 1990). The user specifies the source position, time interval in days, and DIRBE wavelength band. For each source scan, the software returns the source flux in Jy after baseline subtraction and beam profile correction, an error estimate for the source flux, and the background intensity determined from the baseline fit. The tool does not have the capability of tracking a moving object; fluxes for planets and asteroids detected in the DIRBE time-ordered data are available in the Solar System Object Dataset.

                      DIRBE Characteristics Relevant for Point Source Observations 
       

       Nominal    Delta_lambda/   Sensitivity    RMS Short-term  RMS Long-term   RMS Absolute     Beam
Band  Wavelength     lambda      per 1/8 second  Gain Variation  Gain Variation  Calibration   Solid Angle
      (microns)                       (Jy)             (%)             (%)         Error (%)       (sr)
____  __________  _____________  ______________  ______________  ______________  ____________  ___________         

 1A       1.25        0.26             2.5            0.14             0.3            3.1        1.208e-4
 2A       2.2         0.17             3.4            0.08             0.3            3.1        1.408e-4
 3A       3.5         0.27             2.9            0.09             0.4            3.0        1.290e-4
  4       4.9         0.14             4.2            0.08             0.4            3.0        1.466e-4
  5      12           0.67            17              0.16             0.6            3.1        1.422e-4
  6      25           0.36            32              0.20             0.5           14.6        1.478e-4
  7      60           0.46            56              0.44             0.8           10.0        1.506e-4
  8     100           0.31            60              0.54             1.3           12.8        1.442e-4
  9     140           0.25          4100              0.13             1.5           10.1        1.362e-4
 10     240           0.42          2300              0.18             1.8           10.1        1.332e-4


Notes: Detector sensitivity is given for a typical sky brightness at solar elongation of 90 degrees.

       Short-term gain variations pertain to timescales shorter than 1 day.

       Long-term gain variations were measured over timescales from ~5 to ~50 days; peak long-term gain
         variations are about 1% for bands 1A-4, and 2-3% for bands 5-10. 


        

Option 1 produces a visibility chart showing the weeks of the cryogenic mission during which the specified source position was observed by DIRBE. The DIRBE line of sight pointed 30 degrees from the spacecraft spin axis, and the combined spinning and orbital motion of the spacecraft caused DIRBE to trace out a helical pattern on the sky within a viewing swath between about 64 degrees and 124 degrees in solar elongation angle. Over the course of the COBE mission, sources near the ecliptic plane would be located within the viewing swath for 2 months, and then be unobservable for 4 months before coming back into view. Sources within about 25 degrees of one of the ecliptic poles were continuously in the viewing swath. Sample plots from option 1 are available for a source in the ecliptic plane and for a source at 45 degrees ecliptic latitude. Weeks during which the source was visible are shown shaded in green. A given week of the mission is shaded if the source was viewed at any time during that week.

Option 2 selects CIO data for scans that pass within about 0.3 degrees of the source position during a single day. The number of available scans can range from zero to more than 20, and generally increases with decreasing distance between the source and the edge of the DIRBE viewing swath. See, for example, the weekly sky coverage plot in Figure 3.1-1 of the DIRBE Explanatory Supplement. For each selected scan, a time string is extracted that consists of 15 consecutive data samples centered on the sample taken closest to the source position. Two plots are produced. The first plot shows the source position and the pointing positions for the time strings in ecliptic coordinates. An arrow at the end of each time string shows the scan direction. The second plot shows the observed intensity as a function of along-scan position for each of the time strings. (Along-scan position is the along-scan component of the angular distance of the source from the pointing direction, defined as positive toward the scan direction, so time increases from right to left for this plot.) Sample output from option 2 is available for Sirius at 3.5 microns. Sirius contributes to the observed signal for the three central data samples of each time string, and the appearance of these data samples in the photometry plot is determined by the source flux and by the DIRBE beam response at the along-scan, cross-scan position of each data sample. The DIRBE beam profiles are normalized to a peak response of unity. Occasional bad data samples (mostly due to cosmic ray hits or detector hysteresis immediately following the passage of a very bright source through the field of view) are sentinelized at large negative intensity values in the CIO data, and appear as nearly vertical lines extending off scale from the neighboring good data samples in the time string. The sample photometry plot for Sirius includes one such data sample.

Option 2 also produces a table of source photometry and other parameters for each source scan. Entries in the table are as follows.

1990 Day Number -- Average time of the central data sample (the sample taken closest to the source position) in the time string. 1990 Day Number = 1.0 at 1990 January 1 00:00:00 UTC, and day numbers less than 1 refer to dates in 1989.

DIRBE Pointing Direction -- Ecliptic J2000 coordinates of the DIRBE line of sight for the central data sample in the time string.

Source Flux -- The software fits a linear baseline to data samples 3-5 and 11-13 of each 15-element time string. The baseline-subtracted intensity for the central data sample is then divided by the DIRBE beam response at the appropriate along-scan, cross-scan position and multiplied by the beam solid angle to obtain the source flux density. ( DIRBE beam profile maps are available as an ancillary data product.) The flux density F_nu is quoted in Jy at the nominal wavelength of the DIRBE band under the assumption that the source spectrum is given by nu*F_nu = constant. A color correction must be applied to the quoted flux density if this assumption is not correct. 1 Jy = 10^-26 W m^-2 Hz^-1, or 10^-23 erg s^-1 cm^-2 Hz^-1.

Estimated Error -- Error in the source flux calculated as the quadrature sum of (1) the rms noise of the baseline-subtracted data in the regions used for fitting the baseline, (2) an error term accounting for signal-dependent detector noise (this term can be significant if the source peak is much brighter than the background), (3) the error due to rms attitude errors of 1 arcminute in the along-scan direction and 1 arcminute in the cross-scan direction, and (4) the error due to short-term detector gain variations. Attitude errors contribute by causing uncertainty in the correction for beam response. Additional error terms that have not been included, but may be relevant depending on the user's application, are long-term detector gain variations and absolute calibration error. Error term (1) includes a contribution due to sky confusion, but it is possible that confusion within the DIRBE field of view for the central data sample is much different than confusion in the baseline fit region. Thus, a better error estimate may be obtained from the variation of source flux values for multiple scans with different cross-scan offsets and different scan position angles, within a time period where source variability is not expected.

Background -- The baseline fit evaluated at the position of the central data sample is adopted as the background intensity.

Scan Position Angle -- The angle on the sky between the DIRBE scan direction and the direction to the North Ecliptic Pole, measured eastward from the NEP direction. For example, the position angle is 90 degrees for a scan moving at constant ecliptic latitude toward the east (increasing ecliptic longitude). Scan position angle is listed for the central data sample of each time string.

Option 3 allows batch mode processing for time periods longer than one day. It sends an email message to the user containing a table of source photometry for each source scan within the specified period, followed by a list of summary statistics. The content is the same as that of the option 2 output. Some sample light curves have been constructed from option 3 results. A light curve for Sirius at 3.5 microns illustrates the stability of DIRBE photometry for a nonvariable source that is much brighter than nearby sky confusion noise. Sample light curves can also be viewed for Mira, a long period variable with a period of 332 days. Light curves constructed from option 3 results occasionally show individual outlying data points with small error bars, such as the low data point on day 177 in the Mira 4.9 micron light curve. Such an outlier can be caused by a cosmic ray hit that was not identified by the glitch detection algorithm in the data processing software, or can be caused by an unusually large error in the spacecraft attitude determination. Thus, an occasional single outlying data point (either above or below neighboring points in the light curve) should not necessarily be attributed to intrinsic source variability.

Back to the DIRBE Point Source Photometry Tool