We have just experienced our first non-acquisition of a guide star during Frontier Fields observations. This occurred while in the midst of Abell 2744 observations.
Since HST is in constant motion, pointing is maintained by a set of three Fine Guidance Sensors (FGS) which find and lock on to a pair of guide stars, or a single guide star if pairs are not available. These guide stars are selected by software based on several criteria, including magnitude, relative position to other similar stars, position within the FGS “pickles” (Fields of View) and any pointing constraints on the observation such as ORIENT or POS TARGs within the Phase 2 program. Selected guide stars need to stay within the FGS pickles for the entire orbit, including all pointing changes due to POS TARGs or PATTERNs. If an observation spans more than one visibility interval, the guide stars are reacquired after each interruption either from occultation or SAA passages. A pair of guide stars provides the most accurate and stable pointing since they act as sort of handles for HST to focus on. If two stars are used in two separate FGS pickles, then HST is able to maintain almost perfect pointing throughout the observations. If only one star is used, HST may show some drift around the single star since there is not a second star to keep the telescope from rotating. More information about the accuracy of each type of guiding can be found online at www.stsci.edu/hst/acs/faqs/guide_star.html.
In some cases, a guide star may fail to acquire or it might successfully acquire but can not be maintained. Sometimes this is a result of a telescope problem, but more often, it turns out that a selected guide star fails to meet one of the criteria it initially appeared to pass. This can happen in the case of a variable star, a multi-star system that previously appeared as a single star, or with the presence of a similar star (called a spoiler) nearby that confuses the FGS. When PAIRs are used, it is possible to fail to acquire one star, but succeed with the other, resulting in observations taken with single star guiding which is often good enough for most science. There may also be situations when a star is acquired initially but fails to re-acquire in a subsequent orbit, or lock may be lost on one star during an orbit. This is usually due to the star itself being at the very edge of usability and violating one of the limits set by the telescope to help ensure HST knows where it is pointing. With guide star pairs, science can usually continue as long as one of the stars is acquired. If both stars fail (very unusual) or an observation using single star guiding fails to acquire its one star, the observations default to gyro control. This is often problematic to the science as the observations are likely to show significant drift and rotation, or may be far enough off that the target is completely missed.
During the first Frontier Fields visit observing Abell 2744 on May 14, one of the two selected guide stars failed to acquire, resulting in the observations continuing on single star guiding instead. As with all failures, the failed star was investigated and was found to be a bad star. It was flagged in the database within 24 hours of the failure, such that future observations would not attempt to use the same bad star. The second Frontier Fields visit of Abell 2744 on May 15 also failed, as it was already on the telescope and set to use the same guide star pair. Several other visits that were scheduled to execute on the telescope the following week, with the same guide star pair, were quickly reworked by the calendar-building team at STScI to use a different guide star pair. The remaining visits in the epoch not yet put on a calendar are unaffected, since the bad star is no longer an option for our software when selecting from available guide star pairs.
Figure 1: The HST Field of View of Abell 2744, with Fine Guidance Sensors Fields of View indicated by the large, gray arcs.
The green boxes in Figure 1 identify potential guide stars. To use guide star pairs, two stars must fall into separate FGS pickles and remain there throughout any shifts in pointing during the visit. If two similar guide stars are too close to each other, neither can be used since the FGS could lock onto the wrong star. Because of the multiple criteria involved and the need for precision, not all guide stars can be used for a given observation, even if the Field of View seems to show stars that could be used.
The Frontier Fields data products team carried out a detailed examination of all the data from the two visits that were affected by these guidestar issues. For the first visit (number 37), only one of the guidestars was lost, while the other star was successfully acquired and the observations were able to continue in single guide star mode. Analysis of the resulting images showed no measurable impact on the pointing or the PSF quality (consistent with our knowledge that HST is able to perform successfully with a single guide star, when necessary), and all the data from this visit were included in the mosaics.
For the second visit (number 81), the failure mode was somewhat different. The guide stars were fine during the first two orbits of this 4-orbit visit, but began to show problems during the third orbit and failed the reacquisition for the fourth orbit. Consequently, the ACS shutter was closed at the start of the fourth orbit and the fourth exposure for each filter was not obtained. As a result, we include only the first two exposures for each filter in our fast-turnaround v0.5 products, although we may include the third exposure in future versions. For WFC3/IR, all the exposures were obtained, and analysis revealed that the last exposure was offset by no more than a few tenths of an arcsecond compared to its expected location. Thus, there was no significant evidence of drift during the exposures, indicating that the telescope was able to track successfully in gyro mode during these exposures.
So, it makes no difference. Two, one, or zero guide stars – we can do great science in any case!
Patricia Royle – Frontier Fields Program Coordinator