24µm Profile Fit Photometry

Wm. A. Wheaton
5 Jan 2006

Summary Data

For all cases, results were significantly improved (as measured by flux repeatability) when source position (W,V) in each frame was fitted along with source and BKG rates. For the single DCE fits, the median differences between expected (corrected for distortion, based on the headers) and fitted source positions were 0.10 - 0.26 pixels for all AORs analyzed to date. In addition to improved repeatability, the flux estimates all increased, (typically by 10% - 15%) when fitting for position - as expected if the predicted positions were systematically away from the true PSF center. The reason for the significant differences between the predicted and fitted positions has not yet been established.

Fluxes in mJy, using standard calibration factors.

NB that data from the UA DAT have been analyzed by PSF fitting at the SSC for only two AORS, 11782656 and 9805824. Four additional AORs from MC21 have been analyzed for comparision using the SSC pipeline data, using the DCE-by-DCE approach only; results for these are shown as well.

Mosaic Data

Mosaic fitted errors have been derated by a factor obtained from the histograms of the normalized fit residuals, which typically had StdDev ~3 (rather than the theoretically expected 1.00) 


SSC
AOR      HD star#   FlxmJy   Flux   StdErr 
11782656 HD159330    540.0   508.7 +/- 0.7 
9805824  HD189276   2560.0  2567.6 +/- 4.5 

DAT
AOR      HD star#   FlxmJy   Flux   StdErr 
11782656 HD159330    540.0   483.9 +/- 1.9 
9805824  HD189276   2560.0  2417.0 +/- 6.6
The SSC pipeline flux estimates are 5.2% and 6.0% higher than the UA DAT flux estimates, for the weaker (HD159330) and stronger (HD189276) sources respectively.

Flux Averages of Single DCE Fits:

DCE fluxes have been corrected only approximately for pixel size effect due to distortion, by using a single pixel-size factor (namely the Jacobian of the distortion transformation) for each DCE, evaluated at the fitted source position. (It is apparent from some residual regularity in the time series of the residuals for the DCE fluxes that a little of the pixel size effect remains. This might possibly be removed by evaluating the size correction for each detector pixel and incorporating the results into the model, before fitting the model to the data.) 

SSC BCD
AOR      HD star#   FlxmJy   N UWFlux   StdErr  WtdFlux WAFlxEr 
13585408 HD159330    540.0  30   521.5 +/- 0.4    521.7 +/- 0.4 
13586176 HD159330    540.0  30   516.9 +/- 0.5    517.2 +/- 0.5 
13586944 HD159330    540.0  30   519.5 +/- 0.5    519.7 +/- 0.5 
13587712 HD159330    540.0  30   521.2 +/- 0.4    521.3 +/- 0.4 

11782656 HD159330    540.0  30   520.5 +/- 0.6    520.7 +/- 0.7 
9805824  HD189276   2560.0  30  2568.2 +/- 4.7   2568.2 +/- 4.6 

UA DAT (using CF=0.04391)
AOR      HD star#   FlxmJy   N UWFlux   StdErr  WtdFlux WAFlxEr 
11782656 HD159330    540.0  30   514.6 +/- 1.1    514.9 +/- 1.1 
9805824  HD189276   2560.0  30  2500.3 +/- 6.7   2500.0 +/- 6.8 

Columns:
FlxmJy:   Nominal flux, from UA IOC calibration table.
N:        Number of DCEs fitted in AOR.
UWFlux:   Unweighted average of fluxes from N DCEs.
StdErr:   Std Dev of DCE fluxes, * 1/sqrt(N).
WtdFlux:  Weighted average of N DCE fluxes.
WAFlxEr:  Standard Weighted error * sqrt(chisq/nu), where chisq is the chisquare
          to the hypothesis that the DCE fluxes are samples from a unique mean
          = WtdFlux, each DCE with its respective fitted uncertainty, and nu = N-1.

Repeatability of AOR flux estimates, based on the five SSC AORs for HD159330 analyzed by separate DCEs, was 1.86 mJy for the unweighted DCE flux estimates, and 1.80 mJy for the weighted estimates, or about 0.35%.

The SSC pipeline flux estimates are again high with respect to those from the DAT, but the difference is only 1.1% and 2.7%, respectively, for the weaker and stronger sources.

Previous results

For MC21 only, as reported Oct 2005, are given here, without pixel size correction. Fluxes have evidently increased by ~0.4% due to the effects of the distortion correction.

IPAC Home Page | Spitzer Science Center | JPL
e-mail: Bill Wheaton
URL: http://spider.ipac.caltech.edu/staff/waw/
Last modified: 28 February 2006.