WIYN Observatory

NEID

Queue Policies

One use of priority levels is to reflect the timeliness of observations, allowing extremely time sensitive observations (such as Rossiter-McLaughlin measurements) to take priority over other observations. Because the queue is oversubscribed, observations granted priority 3 and priority 4 time are less likely to be executed. It is the intention of the queue design that a high percentage of observations done with priority 0 through priority 2 time will be completed.

What to Propose

NEID observations can address a variety of problems. The high radial velocity precision planned for NEID is its most unique aspect, but NEID proposals are reviewed with the same guidance as proposals to use other instruments at WIYN.

Because of queue scheduling, proposers should consider that observations are taken under a wide variety of conditions and consider the option of proposing projects that can take advantage of relatively poor seeing, transparency or sky brightness. It is likely that competition for observing time will be lessened under such conditions and projects planned with this in mind could be among the most successful. The high efficiency (HE) spectral resolution mode may be useful for certain programs expecting to acquire data under suboptimal conditions. Under particularly windy conditions, WIYN may only observe at certain azimuths, so targets located in other parts of the sky will be inaccessible. For example, if many queue proposals prefer southern targets (e.g., K2 or south TESS fields), a program having targets north of 32 degrees declination would be in favor whenever the telescope must point north. For this reason, proposers will be asked to supply some extra details on the targets for their programs during Phase 1. Upon receiving proposals for the NEID queue, NOIRLab staff will inspect the proposals to categorize the requests into subsets appropriate for the range of observing conditions expected. This information will be passed to the TAC so that the telescope time can be efficiently allocated.

During Phase 2 and 3, proposers will specify the limiting, poorest conditions under which their observations can be taken and will be able to specify relaxed constraints for observations that can take advantage of poor conditions.

For instance:

How to Propose for NEID time

Proposals for NEID queue time are done in 3 phases. The first of these, Phase 1, includes the ordinary NOIRLab proposal process completed by all proposers (due by the proposal deadline in the Call). The second, Phase 2, will be required for all successful proposals and its purpose is to fill in target and observation details to complete the needed information to schedule observations before the semester starts. Phase 3 of an observing program describes changes or requests for new targets made during the observing semester. Instructions for Phase 1 are given here. Instructions for completing the next phases of the proposal process will be provided in the future.

Proposers to use NEID should fill out the standard NOIRLab proposal forms for 2021B. This is one part of Phase 1. In addition to the standard text descriptions, more details are needed during Phase 1 that describe a program's targets and observation requests.

The TAC will very likely have to award some proposers time at lower (worse) priorities than requested, so proposers should explain the consequences of receiving low-priority time and specify the worst priorities their science can tolerate, to help guide the TAC in this decision. In general, the TAC should strive to award the highest-ranked proposals time at the requested priorities, provided it is well justified; lower-ranked proposals will likely receive worse priorities than they requested. While priority level is the principal means by which an observation is weighted in the queue scheduling decisions, low priority observations can be expected at times when the number of eligible targets in the overall queue dwindles (e.g., under poor seeing and transparency, when telescope azimuth is restricted due to strong wind, or possibly in bright conditions).

To guide these requests: A star that needs one observation on any night during the semester is generally easy to schedule. This is especially the case if it is bright and may be observed under poor conditions and bright skies. Such an observation shouldn't require high priority time. A star that is relatively far south for WIYN and that you wish to observe at specific times (e.g., observing twice at times of both quadratures relative to a planetary transit) is likely to require somewhat high priority time. A star that you wish to sample 12 times over a planet's orbital period may have a mixture of easily scheduled observations (starting out) and more difficult ones (later on to fill in missing orbital phases).

Finally, you have the option of specifying certain nights on the calendar that would be most advantageous to observe your proposed targets in the same way that you may specify the minimally acceptable observing conditions. These requests from accepted proposals will be used to guide the WIYN instrument schedule. Specific nights are probably relevant to a few science programs, but the schedule can only be partially flexible to meet a request for specific nights. A justification for requesting specific nights could be that you need to observe specific events in a planetary ephemeris that occur only rarely during the semester. The nominal scheduling plan for the queue is to switch often between queue nights and non-queue nights. Most of the time there will be a maximum of two “off” nights between queue nights, but longer shutdown periods will be scheduled occasionally. Many periodic events would then recur frequently throughout the semester (e.g., a particular phase of a planet with a short orbital period).

Scheduling Long Sequence Observations on NEID

The current NEID queue is designed with a scheduling system reliant upon each observation having a priority level which the Scheduler uses to select one observation over another, resolving conflicts. Based on anticipated NEID science cases and some experiences relayed to WIYN by HET users, the NEID queue time was distributed with relatively small fractions at the highest priority levels. This scheme was intended to minimize the number of conflicting observations whose epoch of observation was critical because it allowed users or TACs to attach high priority levels to such observations, increasing their chances of being scheduled. These observations would work best at a priority level of P0 and, to a lesser extent, P1.

A review of 2020B observing proposals reveals a significantly higher demand for time-critical observations using NEID than this original plan will easily accommodate. The bulk of the problem lies with a large number of programs seeking to observe host stars using a long sequence of consecutive exposures during planetary transits. Such observations are time critical while their duration requires a large time allocation. Various partner TACs are requesting programs with these types of observations, but were forced to allocate them lower priority time (P2 or lower). In addition to being time critical, such long observations are difficult to schedule without conflicting with other equal or higher priority observations, so merely putting more queue time at high priority level will not solve the problem very well.

The WIYN team has spent some time considering potential solutions that could be used to schedule observations of a specific type: long duration observations that have a stringent requirement in terms of their epochs of observation. One method to deal with these observations would be to modify a current algorithm that identifies and selects them for first-in-line scheduling (before the remaining time is filled in with easier-to-schedule observations). This algorithm had been slated to operate only on observations among the highest priority levels (P0 and maybe P1). However, it could be extended to lower priority levels to accommodate more observations of this type. This would effectively treat such observations in a special, privileged manner that may override some shorter, more easily scheduled observations of higher priority level. Note that "easy-to-schedule" is a property that is relatively straightforward to define on a single night, but not so straightforward to define for the remainder of a semester.

Other methods explored treat these types of observations as "classical" where the program assumes the risk of losing their time allocation due to weather or facility problems. Classical programs and observations would best be declared early on in the semester, ideally by the time the schedule is created. This may limit the type of programs that could use such a mode. NEID standard stars could be observed on classical nights like they are on queue nights, but otherwise would follow normal WIYN protocol. Alternatively, another method would be to define certain well-justified observations as classical, but schedule them in the midst of queue observations at pre-declared times.

NSF University of Wisconsin Indiana University Purdue University Pennsylvania State University University of California, Irvine

Last modified: 18-Jan-2022 10:58:52 MST