9.2 The WFC3 Exposure Time Calculator - ETC

In most cases, you will find it convenient to use the online WFC3 Exposure Time Calculator (ETC) to make estimates of the required exposure times for your project.

The ETC calculates counts (e) and count rates (e/s) for input source and background parameters and assumed characteristics of the detectors. Once these are entered, the ETC then outputs signal-to-noise ratios (SNRs) achieved for a given exposure time, or the exposure times required to achieve a given SNR The ETC supports both direct-imaging and spectroscopic (grism) observations. Starting in 2016 (Cycle 24), the ETC supports spatial scanning for UVIS and IR imaging and IR spectroscopy (WFC3 STAN issue 22). A variety of circular and square extraction apertures are available in the ETC, allowing the user to select either a radius in arcseconds or a size in pixels. It is also possible to input a calibrated spectral-energy distribution (SED) of your source directly into the ETC. The ETC also outputs peak per-pixel count rates and total count rates, to aid in feasibility assessment. Warnings will appear if the source will saturate the detector, which would not only compromise CCD and IR observations, but might even affect subsequent exposures with the IR channel (Appendix D). The ETC has online help for its execution and interpretation of results.

There are some items worth noting:

  • For the UVIS channel, the ETC uses a CCD full-well value of 63,000 e, the minimum value for either CCD chip, to determine saturation; users wishing to strictly avoid this occurrence should allow a buffer of at least 10% below this. (Section 5.4.5 presents a discussion of saturation.)
  • For the UVIS channel, to the extent possible, one should select Detector chip 2 for UV filter exposures and place the target on chip 2 (e.g., aperture UVIS2 in Figure 6.1 or the subarray apertures in the lower left quadrant of Figure 6.2) since chip 2 has better quantum efficiency than chip 1 (Figure 5.2). Note, however, that the quad filters have quadrant-dependent passbands as specified in Table 13.5 in Section 13.4.1 of the Phase II Proposal Instructions. The user should be careful to select the Detector chip corresponding to the relevant quadrant for a given filter name, since the ETC will incorrectly allow either chip to be selected. (Quads A and B are on chip 1; quads C and D are on chip 2.)
  • To mitigate CTE losses, observers can add background to UVIS exposures by using the post-flash option (Section 6.9.2). See Section 9.6 for equations that show how the post-flash background affects the calculation of SNR and exposure time.
  • For the IR channel, when the # of Frames = 1, the ETC assumes a complete timing sequence (NSAMP = 15). For smaller NSAMP, the read noise is greater (Section 5.7.3). For most programs, NSAMP should be greater than 5, and ideally more if possible, to achieve the best readnoise and cosmic ray correction (Section 7.10.3).
  • The He I airglow line at 10,830 Angstroms is being added as a component of the IR sky background for Cycle 24 and beyond (Section 7.9.5). It only contributes to the background observed in the F105W and F110W filters and the G102 and G141 grisms. The default ETC value of “None” will provide ETC estimates unchanged from those of previous cycles. The other available options---Average, High, Very High---are the 50%, 75% and 95% percentile values of the excess background, per exposure, observed over the value estimated for the zodiacal light alone (e.g., Figure 7.12). Determined from an analysis of archival exposures, these percentiles correspond to an additional 0.1, 0.5, and 1.5 e-/s, respectively, normalized in the F105W filter. Note that the “Average” value of 0.5 e-/s is comparable to the flux predicted for the zodiacal background (Figure 7.12). Furthermore, achieving the He I value of “None” is essentially only possible if the SHADOW special requirement is used, though this is not recommended as the available usable orbit duration is then dramatically reduced; it is almost always preferable to keep observing even in the presence of elevated backgrounds for part of the orbit (WFC3 ISR 2014-03).

It is also possible to use stsynphot to calculate count rates and the wavelength distribution of detected counts.

The remaining sections of this chapter give detailed information on the sensitivities of the UVIS and IR channels when combined with the various spectral elements, and the use of this information in the determination of count rates, for those who wish to understand the subject in depth.