The Proposal Submission Process and Scheduling Webb

Amaya Moro-Martín,

For Webb, a single-stream process for proposal submission has been adopted because maximizing the scientific return from the telescope requires that the community has the ability to utilize it as efficiently as possible. One step in that process is minimizing the time between proposal deadlines1 and the start of an observing cycle. This optimizes the amount of Webb data available at the time when the subsequent round of proposals is written, thus guiding new Webb observations.2

We can achieve a faster turnaround if proposals can be scheduled for observing soon after they are recommended by the Webb Telescope Allocation Committee (TAC) and accepted by the Institute Director, without requiring a later Phase-II submission that includes the detailed observation plan. Under this single-stream process, most (but not all) proposals must include sufficient information at submission to define the scheduling constraints for all visits (each visit directed at a specific target). This includes specifying the science duration, which is the sum of the individual exposure times for all the targets. Also specified will be the direct overhead duration, defined to be the sum of the duration of the guide-star acquisitions; the target acquisitions; the time for any mechanism motions; the small-angle maneuvers (SAMs) due to specified dithers or offsets; and the initial slew duration from a previous visit.

The sooner that information is available, the sooner an initial Webb Long Range Plan (LRP) can be prepared, combining all visits in all the accepted proposals in a way that resolves conflicts in the schedule, maximizes the observing efficiency, minimizes the effect of detector persistence from prior exposures, and minimizes the build-up of angular momentum. In addition to the direct overheads mentioned above, all proposals will be charged an indirect Observatory overhead, which corresponds to 16% of the scheduling duration plus the slew duration, to support operations—primarily calibration programs and maneuvers to dissipate excess angular momentum (see Figure 1).

Figure 1: Total time charged to a Webb observing program (Exp. = exposure time; GS Acq. = guide star acquisition; SAM = small angle maneuvers).

Webb naturally lends itself to the single-stream approach, as was the case for Spitzer. For Webb, most observations will be specified via the Astronomer's Proposal Tool (APT), using pre-defined templates for individual instrument modes. Observing time will be allocated in hours, not orbits; consequently, visits need to be specified a priori in greater detail than for Hubble, regardless of the submission process. APT computes the direct and indirect overheads for an observing program in a deterministic way for the direct overheads that occur within the confines of a proposal’s visits, and in a statistical way for those that do not (namely the indirect Observatory overhead mentioned above and the initial slew duration3). For Cycle 1, these statistical estimates are guided by a simulated yearlong LRP constructed with the science programs in the Science Operations Design Reference Mission (SODRM), covering the range and depth of science and calibration programs that Webb might carry out in the Cycle. For Cycle 2, those parameters will be adjusted, based on the actual usage in Cycle 1.

Cycle 1 Preparation: Most Cycle 1 Webb GO programs will be required to include a full list of targets, specifications of the observations (instruments, filters, exposure times, dithers, observational sequence) and all user-requested scheduling constraints (including roll angle and other timing constraints); there will be exceptions for complex programs (see below). The proposer must also carry out guide star and visibility checks, verifying that the observations are schedulable given the specified constraints. However, the proposer may not choose specific guide stars or exact scheduling windows, as the schedulers need flexibility to prepare the most efficient LRP. (See the articles by Bill Blair and Christine Chen in this same issue of this Newsletter for a description of the planning tools that will be available for proposal preparation and the training opportunities that will be offered to the community).

In some cases, it may not be possible to fully determine whether a proposal is schedulable. This is the case, for example, of Target of Opportunity (ToO) proposals, linked to an event that may occur at an unknown time (e.g., novae, supernovae, gamma-ray bursts, newly discovered comets, newly discovered transiting planets, etc.).

Proposals flagged with warnings in APT may be submitted and the technical issues will not be shared with the TAC, as their recommendations are based on the scientific justification of the proposed observations. If accepted, the Institute will contact the Principal Investigator to address specific problems prior to scheduling the program. Technical reviews will be carried out for all Cycle 1 programs and, in some cases, the program coordinator and contact scientist may suggest changes to maximize efficiency. Modification requests initiated by the Principal Investigator after the proposal’s acceptance—to better achieve the approved science goals or to accommodate changes in instrument performance with respect to that expected at the time of the proposal submission—will have to be approved by the Telescope Time Review Board (for GO programs), and by the Institute Director (for all programs). All modifications need to be scientifically justified and have minimal impact on the LRP.

Exceptions to the single-stream approach in Cycle 1 will be made for programs for which this complete specification is not possible (e.g., microshutter array [MSA] spectroscopy of targets identified from Webb pre-imaging) or where this level of preparation places an exceptionally large burden on the proposers (e.g., large mosaic imaging programs). In those circumstances, the proposal process will follow a two-phase approach, similar to Hubble, where the full set of specifications will be provided quickly after acceptance. Regardless, the initial submission of those proposals must include a complete target list and an estimate of the total time requested to enable identification of scheduling windows as early as possible. This is particularly important for large programs since they impose significant constraints on the LRP.

Duplications checking: The finalized versions of all accepted programs will be reviewed to ensure that the overall observing plan is consistent with the TAC allocation. The programs will also be checked for duplications within the same cycle, or with scheduled or completed observations from previous cycles. A complete list of Cycle 1 GTO targets will be published by June 15, 2017 to allow ERS and GO proposers to carry out the required duplication checks well ahead of their proposal deadlines. APT versions of GTO Cycle 1 programs will be available when the Cycle 1 GO Call for Proposals is issued in November 2017. Duplications are not permitted, unless explicitly recommended by the TAC. Duplications within approved proposals will be resolved following Institute and NASA Webb policies.

Updates to the Webb LRP: The LRP will be updated periodically through the cycle to accommodate, for example, NIRSpec MSA configuration updates that occur as pre-imaging is obtained, DDT Observations and ToO Observations. Depending on the requested turnaround of the ToO observations, they might be considered disruptive of the LRP and a limited number of those will be allowed per cycle. Fast turnaround ToOs, requiring execution within a few days of triggering, will incur an additional overhead to account for the potential of having to interrupt an executing visit in order to accommodate the ToO visits. Additional overheads will also be imposed for time-critical observations (e.g., exoplanet transit observations) since those observations place severe constraints on the schedule and are likely to decrease the observing efficiency by introducing "dead time" prior to their execution.

Funding: As with previous NASA flagship/strategic missions, Webb will not only provide observations of superb and unprecedented quality to a worldwide community of diverse demographics, but for US-based observers, it will also be a source of significant funding to facilitate the resources needed for the analysis and publication of its valuable data. In Cycle 1, a Financial Review Committee (FRC) will set the funding for each proposal. As with Hubble, the FRC will review budgets and work plans submitted by the each proposer. In subsequent cycles, the aim is to assign a base funding level for each US-based proposal to be calculated automatically using a formula derived from the Cycle 1 allocations (with requests for additional funding subject to review and approval by the FRC). This would lead to more rapid budget approval after proposal acceptance than is the current case for Hubble.


1 I. Neill Reid describes the Webb timeline, including the deadlines of the different Webb observing opportunities, in an article in this same issue of this Newsletter.

2 The goal of the Director’s Discretionary Time Early Release Science (ERS) program, described in the article by Janice Lee in this same issue of this Newsletter, is precisely to maximize the amount of Webb data available at the time of the Cycle 2 deadline, when public non-proprietary Webb data will be extremely limited and its full scientific impact will have just begun to be understood.

3 Since visits are the primary scheduling unit, and since a given proposal’s visits may be interleaved with those from other proposals and activities, the actual slew prior to each visit cannot be known a priori and needs to be estimated statistically.