SSD Technical/Training Series - Photometry IV

Photometry with STIS
Harry Ferguson
10-10:30am, Thursday
December 10, 1998
Auditorium

• Why do photometry with STIS
• STIS CCD Peculiarities
• Simple test -- one star observed 4 times in succession
• Aperture Corrections/PSF variations
• CTE for the STIS CCD
• Multi-color photometry
• MAMA Peculiarities
• PSF/Aperture corrections

Why do photometry with STIS?

CCD Peculiarities

• Gain 4 fixed pattern noise
• Aperture corrections/PSF variations
• CTE
• Color terms

MAMA

• Background estimation
• Aperture corrections
• PSF variations

Filter Throughputs, redleak, and uncertainties

Why do photometry with STIS?

In the optical, STIS has more sensitivity and better resolution than WFPC2. (However, it does not have an extensive set of filters).

In the UV, STIS has better throughput than the FOC, a wider field of view, better redleak suppression, and much more easily understood systematics.

STIS can be used in parallel with WFPC2 and NICMOS. Lots of free data in the archive...

STIS CCD Peculiarities

Gain=4 fixed pattern noise.

---

Simple test -- one star observed 4 times in succession

The pattern noise affects the background which is important either for measurements of faint stars (in which case you should probably use gain=1), or for high-precision measurements of bright stars.

Aperture Corrections/PSF variations

• Aperture corrections are given in the STIS handbook for stars near the field center.
• A more detailed look at the PSF can be found in a report under "calibration" on the STIS web pages (www.stsci.edu/instruments/stis/performance/foibles/psf_robinson.ps).
• There are ghosts visible near bright stars. Total flux in the ghost is of order 1% of the flux in the star. The ghosts vary as a function of position and are different in the different filters.

For the PSF plots, please refer to Figure 18 and Figure 25 of the Robinson report.

CTE for the STIS CCD

The result from recent on-orbit testing is

• parallel CTE = 0.999982 +/- 0.000018

This corresponds to an expected 0.02 +/- 0.02 mag shift for stars at the very top of the chip relative to stars at the very bottom.

CTE can also affect centroiding of stellar images at the level of less than 0.1 pixel. Probably less important than PSF variations due to breathing.

So unless you are interested in super high precision, you can ignore CTE in STIS for now.

Multi-color photometry

Color transformations...try this in synphot...

---

MAMA Peculiarities

Background estimation -- This is a gotcha for faint sources!

• The background in a typical STIS MAMA pixel will be a small number, either 0, 1 or 2.
• Using "median" will not give a good estimate.
• Some programs use the median to compute the mode, so be careful.
• The iraf "ofilter" option works well.
• Or just check for bad pixels, and if there aren't any, take the mean .

PSF/Aperture corrections

For the PSF plots, please refer to Figure 14 and Figure 20 of the Robinson report.

For the PSF plots, please refer to Figure 15 and Figure 22 of the Robinson report.