Consider a target to be imaged with WFC in a given filter in one orbit. Using the ETC
, we find that we need 2400 seconds of exposure time to reach the desired level of signal-to-noise ratio. Given that the observation must be split into a series of two exposures by CR-SPLIT (CR-SPLIT=2
), we map the overheads and the science exposure times onto the orbit as follows:
Thus, the two WFC exposures totaling 2400
seconds make full use of the typically available time in one orbit. The exposure times can be adjusted if the actual target visibility time differs from the derived total used time.
This example illustrates the impact of short WFC exposures on the useful time in
the orbit. We have one orbit to observe a target with WFC in two filters, so the observation consists of two series, each with two identical CR-SPLIT exposures. The ETC
has shown that at the minimally accepted signal-to-noise ratio the exposure time must be 540 seconds for each of the filters, so each of the CR-SPLITs
must be at least 270 seconds long. For the target declination, we find that the visibility time is 55 minutes. The time budget for the orbit is then as follows:
Comparing with the previous example, we see that although with the adopted
minimum exposure times we can squeeze the observation into one orbit, the efficiency of the orbit use is very low because of the large overheads associated with buffer dumps. However, if we increase each of the four exposure times so that they are larger than 339 seconds, we avoid these additional overheads. This would free ~17 minutes of the orbit time for science, which allows us to almost double the science exposure time (35 minutes instead of 18 minutes) and thus significantly improve signal-to-noise.
Similarly, a subarray can be used to readout only a fraction of the detector, allowing
more frames to be stored in the buffer before requiring a dump. In this example, using four WFC1-1K
subarrays for 4 short (t < 339 seconds) exposures would save 176 seconds in readout time and 1047 seconds in dump time. This frees up ~20 minutes of orbit time to be used for science.
This example illustrates the orbit calculation for a WFC observation with the ACS
box pattern, which implements imaging at four offset pointings. The goal of the observation is to obtain a dithered image of a field in such a way that would allow us to bridge the 50 pixel interchip gap between the WFC CCDs in the combined image. Given the WFC plate scale of 0.05 arcseconds/pixel, this requires that the offsets in the dithering pattern are larger than 2.5 arcseconds. Each offset will then take 0.5 minutes to move the spacecraft from one pointing in the pattern to another. We have determined that the exposure time necessary to reach the desired signal-to-noise ratio is 80 minutes. The visibility time at our target declination is 58 minutes. In this observation we do not want to rely on cosmic ray removal provided by the dithering data reduction package, and set CR-SPLIT=2
to be able to remove cosmic rays from the four individual images separately. As a result, the orbit calculation will involve a series of 8 exposures (two exposures at each of the four pointings in the dithering pattern) split across two orbits:
The total used time in the first orbit comes out a little bit larger than the visibility
time. However, given the conservative nature of the adopted overhead times as well as a bit of flexibility in the adopted signal-to-noise ratio, the difference is not significant. It should be remembered that the purpose of the above exercises is to estimate how many orbits to request for our science program rather than to exactly design the observation.
This example illustrates the orbit calculation for a simple 30
minutes WFC grism spectroscopic observation broken down by CR-SPLIT=2
into a series of two exposures.
Unlike similar imaging exposures, here we have to take into account an additional
imaging exposure before the sequence of spectroscopic exposures, which takes 10 minutes. off the available orbit time.
This example deals with the orbit calculation for an observation of a relatively faint
extended object using the SBC. The target has to be observed using two filters, F150LP and F165LP. The ETC shows that the required S/N for the observations are achieved in 3200 seconds and 2000 seconds for the F150LP and the F165LP filters, respectively. There is no readout noise associated with SBC exposures; therefore, the observations can be split into four equally long dither pointings. Since the average visibility time for the target is ~55 minutes, the images can be taken in two orbits, as shown in Table 8.7
. The standard ACS-SBC-DITHER-BOX pattern, which allows for the rejection of most artifacts, is suitable for these observations. Here are the details of the orbit calculation:
The total time is slightly shorter in the second orbit because of the shorter time
required for guide star re-acquisition and SBC overheads. There is also ~1min of visibility available in the first orbit that can be used for the observations. However, we recommend dithered observations in each filter using the same exposure time for each pointing.