March 2017 STAN
In this STAN we provide updates on STIS to users preparing a program for the Cycle 25 HST Phase 1 deadline.
STIS Focus Changes
There appears to have been a modest change in the focus of the STIS instrument in recent years. The primary impact of this on observers is a decrease in throughput when using STIS apertures that are less than 0.2" wide. In most common cases these losses are relatively modest, and we discuss how these changes impact the planning of observations.
There is some evidence that in recent years the STIS instrument focus has changed. Because the STIS corrector mechanism has not been moved since 1997, it is unlikely that the focus will be adjusted unless the performance suffers from additional degradation.
The most obvious impact of these focus issues is a decrease in throughput when using apertures smaller than 0.2" in size. This will lower the S/N achieved by an observation of a given length and also reduces the accuracy of the absolute flux calibration. The exact throughput change for a given aperture will vary as the focus changes during the course of HST's normal orbital breathing. Analysis of recent echelle observations taken through the 0.2X0.06 and 0.2X0.09 apertures shows that the average throughput in recent years has been only about 80% of nominal, with occasional individual exposures with these apertures showing as much as a 40% throughput loss. Note that for these apertures, throughput variations of order 10% due to telescope breathing have always been commonly encountered throughout the lifetime of STIS. For the smallest aperture, the 0.1X0.03, average throughput in recent years has typically been only half of its nominal value. Since these throughput losses vary significantly from observation to observation, it is not possible to simply update the ETC throughputs, as the ETC must also warn against observations which are too bright or which may cause saturation, and must therefore adopt the highest throughput that might reasonably be encountered.
Focus offsets can also affect the relative flux calibration as a function of wavelength within a given observation. For modes covering a wide range of wavelengths, relative flux errors of 10% over the wavelength span of E140M and E230M observations done with the 0.2X0.06 aperture are now common. If combined with small aperture centering errors, these can sometimes increase to as much as 25%.
The resolution changes caused by these focus variations appear to be small, although they are measurable in high S/N observations and may affect programs which require extremely high levels of observation-to-observation repeatability.
Users requiring a particular signal-to-noise value when using a small aperture should take these effects into account when estimating the amount of exposure time required. Users requiring absolute or relative flux calibration better than 10% are advised to use apertures of 0.2" wide or larger. For first order observations, the 52X0.5 or 52X2 apertures are recommended to achieve the best photometric results.
A more detailed analysis of this issue is detailed in STIS ISR 2017-01: Status of the STIS Instrument Focus
Spatial Scanning with the STIS CCD
Spatial scanning with the STIS CCD can be used as an available-but-unsupported mode.
STIS observers should be aware that spatial scans, which are a supported mode for WFC3 observations, (see WFC3 ISR 2012-08: Considerations for using Spatial Scans with WFC3 ), can also be used with the STIS CCD as an available but unsupported mode.
In the past, very high S/N spectral observations with the STIS CCD have often been carried out by deliberately saturating the detector at a fixed pointing while using GAIN=4. At this gain setting, electrons collected above the ~130,000 full well limit of an individual pixel do not get lost, but bleed along the CCD columns which are perpendicular to the dispersion direction. This allows full spectral resolution to be preserved even for highly saturated images. For some applications, S/N in excess of 10000:1 have been demonstrated using this technique (see STIS ISR 1999-05, Gilliland, Goudfrooij & Kimble, 1999, PASP 111, 1009, and Bohlin & Gilliland, 2004, AJ 127, 3508). While for many science goals, this observational strategy will continue to be satisfactory, it has the disadvantage that since information about what row each photon landed on is lost, pixel level flat-fielding and the removal of IR fringing can become significantly more difficult. Because of this, the very high signal-to-noise values quoted for saturated STIS CCD images only apply to either very broad band or differential measurements where the uncertainties caused by small scale flat-fielding errors can be either averaged over or subtracted away. For more direct applications, such as measuring the absolute equivalent width of a weak absorption feature in a non-variable source, the actual S/N achievable will be much lower. For observations of this type, the use of trailed exposures with STIS may offer some unique opportunities.
STIS CCD observations are normally corrected using pixel-to-pixel flat fields, and, for observations at wavelengths greater than about 7500 angstroms, with an additional IR fringe-flat (see STIS IHB section 7.6.2). The pixel-to-pixel flats are relatively stable and wavelength independent and are normally constructed by combining several long slit tungsten lamp observations; however, the IR fringe flats are very sensitive to the exact distribution of wavelengths falling on each pixel and so contemporaneous observations are recommended to ensure the same instrumental alignment. However, for both types of flats, the illumination pattern of a long slit differs significantly from that of a point source, while use of a small slit to better simulate a point source illumination pattern results in limited S/N and restricts the choices for external target positioning on the detector.
The illumination pattern of a point source trailed in the spatial direction along the length of a long slit will much more closely resemble that of long slit pixel-to-pixel or IR fringe-flat lamp exposure. In addition, spreading the light over numerous rows of the detector will better average over both localized pixel-to-pixel and IR fringing variations, and allow for significantly improved S/N in any collapsed flat field constructed from the lamp observations.
The GO program 14705 (PI Martin Cordiner) has conducted some test observations using spatial trails with the STIS G750M grating at the 9336 central wavelength setting. The goal of this program is to achieve S/N of 500:1 or greater for the detection of weak diffuse ISM features near 9400 Angstroms. Preliminary results from the initial visit are still under analysis, but suggest that the observations worked well and the target was well aligned with the 52X0.1 aperture during the trail. The location on the detector for the start and end of the "forward" scans in this test were very close to the expected positions, while the "reverse" scans show somewhat larger variations.
Because this is an available-but-unsupported mode, observers should be aware that the documentation and support for this mode with STIS are very limited, although much of the WFC3 documentation on implementing spatial scans in APT will also apply to STIS observations. In addition, as is the case for all available-but-unsupported modes, use is on a "shared-risk" basis and observational failures directly related to the use of such a mode will not normally be eligible for a repeat. Further discussion of the policies related to the use of "available-but-unsupported" modes can be found in section 2.4 of the STIS IHB.
STIS CCD ETC Updates for Cycle 25
A new option for STIS CCD ETC calculations exist that allows the user to select the dark rate.
When performing ETC calculations involving the STIS CCD, the dark level previously used was appropriate for observations taken at the nominal position located in the middle of the detector. For observations taken at the E1 positions, however, this default dark rate is higher than the actual background rate at that position. To account for such differences, the STIS Team has added an additional option to Question 1 for the CCD detector. This option allows the user to select the level of additional detector background to incorporate in the ETC calculations based on the position of the spectrum on the detector: Low, Medium or High settings are now available and correspond to the Top, Middle or Bottom of the detector, respectively.