5.2 Optical CCD Imaging

The CCD imaging capability of STIS was designed primarily for target acquisitions, and therefore, only a small number of filters are available. Nevertheless, STIS CCD imaging has scientific utility of its own due to the high throughput and relatively low read noise of the CCD detector. STIS CCD imaging can be obtained as prime pointings or in parallel with other instruments.

The optical performance of the CCD in imaging mode is good, and the plate scale of the CCD is 0.05078 arcseconds per pixel, providing a good compromise between sampling of the PSF and field of view. There is some degradation of the image quality towards the edge of the field. Observers can assume that 15–20% of the light from a point source falls in a single STIS CCD pixel and that ~80% of the light from a point source is contained within a 5 × 5 pixel region. An image of a typical point source is shown in Figure 5.4. See Chapter 14 for encircled energies at the field center for the different imaging modes and information on the field dependence of the PSF. For a discussion of the coronagraphic mask, see Section 12.11.

Figure 5.4: A Co-added Image of SAO 255271 Taken Using 50CCD.


This figure is plotted with a logarithmic intensity scale and is about 10" across. It shows the structure in the STIS PSF. The ring seen below the center of the PSF is a ghost image. The position angle of this ghost varies as a function of location on the CCD (see Section 7.2.8 and Figure 7.6).
The throughputs used for the CCD imaging modes are for the most part based on measurements of on-orbit calibration data and are accurate to within 5%.

5.2.1 Effect of the Optical Baffles on STIS CCD Imaging

Calibration observations were taken to investigate whether the baffle structures around the CCD detector scatter light from the stars outside the CCD detector into its field of view, and if so, by how much. Images were taken with a bright star placed at 12 different positions, 10 arcseconds and 37 arcseconds away from the edge of the detector. The analysis of the images showed that there are no measurable scattering components by the baffle structures around the detector. Placing the bright star beyond the edge of the detector thus acts as an effective coronagraph for the STIS CCD detector (see Proffitt et al. 2004, ApJ, 612, 481 for more details).

5.2.2 Unfiltered (Clear) CCD Imaging: 50CCD and 50CORON

The 50CCD aperture is a clear, unvignetted aperture which provides maximum sensitivity over the full 52 × 52 arcsecond field of view. The shape of the bandpass is governed by the detector, which has sensitivity from ~2000 to 10,300 Å. Figure 5.1 shows the throughput as a function of wavelength for this imaging mode (see also Chapter 14 for sensitivities, signal-to-noise plots, and saturation plots).

The coronagraphic aperture 50CORON contains two wedges and a bar which can be used to occult a bright object. The unocculted parts of the aperture are clear, and so the throughput as a function of wavelength is the same as for the 50CCD aperture. Further details about this aperture as well as advice for conducting STIS coronagraphic observations are given in Section 12.11.

5.2.3 Optical Longpass: F28X50LP

The STIS longpass filter cuts off at λ < 5500 Å. It images a 28 × 52 arcsecond field of view. The F28X50LP filter is the principal target acquisition aperture (see Selecting the Imaging Aperture). The integrated system throughput for this filter is given in Figure 5.1 (see also CCD Long-Pass Imaging F28X50LP for sensitivities, signal-to-noise plots, and saturation plots).

5.2.4 [OIII]: F28X50OIII

This filter images a 28 × 50 arcsecond field of view and can be used in target acquisitions or for direct imaging in the light of [O III]. The [O III] filter integrated system throughput and a signal-to-noise comparison with the WFPC2 [O III] filter are shown in Figure 5.5 (see also F28X50OIII - CCD for sensitivities, signal-to-noise plots, and saturation plots). The STIS [O III] filter is very narrow: only 6.2 Å wide, compared to the WFC3 [O III] filter which is roughly 65 Å wide. The STIS [O III] filter has a substantial red leak that begins at 10,600 Å and continues to at least 12,000 Å. In the case of a very red star (K0 spectral type), the red leak will contribute approximately one-third of the detected counts. The red leak for this filter is included in the passbands used by the STIS ETC and synphot. Observers are encouraged to use these tools to predict source and background count rates carefully.

Figure 5.5: F28X50OIII Throughput and Flux vs. Exposure Time.


(Left) Integrated System Throughput and (right) Flux vs. Exposure Time to achieve a signal-to-noise=5 compared to WFPC2 for a FWHM=1 Å line, integrated over an area of one square arcsecond.

5.2.5 [O II]: F28X50OII


The [O II] filter images a 28 × 52 arcsecond field of view and can be used in target acquisitions or for direct imaging in the light of [O II]. The [O II] filter integrated system throughput and a signal-to-noise ratio comparison with WFPC2's [O II] filter are shown in Figure 5.6. See F28X50OII - CCD for sensitivities, signal-to-noise and saturation plots.

Figure 5.6: F28X50OII: Throughput and Flux vs. Exposure Time.


(Left) Integrated System Throughput and (right) Flux vs. Exposure Time to achieve a signal-to-noise=5 compared to WFPC2 for a FWHM=1 Å line, integrated over an area of one square arcsecond. There is no substantial red leak in this filter.