The JWST Advisory Committee (JSTAC): Maximizing the Scientific Productivity of JWST

Garth Illingworth,

The JSTAC’s role, distilling its charge down to a key phrase, is to advise the Institute Director on “maximizing JWST’s scientific productivity” during its operational life. While this enunciation is simple and focused, the challenges during science operations for a mission of the complexity of JWST facing the partner agencies (NASA, ESA, and CSA) and the Institute, and an advisory committee like JSTAC, are similarly wide-ranging and complex. JSTAC members were chosen across the three-agency partnership to have extensive background with space science missions. The Institute’s Director formed the JSTAC in 2009 after consulting the partner agencies. The Institute Director appoints the JSTAC, and its 18 members include representatives from the US, European, and Canadian astronomical communities and ex-officio observers from NASA, ESA, and CSA. JSTAC submits recommendations to the Institute’s Director. These letters are public: the JSTAC webpage is here.

JSTAC meets at the Institute twice per year. Over the years, a broad range of topics has been covered. Meetings typically include presentations from Eric Smith (JWST Program Director at NASA Headquarters), Bill Ochs (JWST Project Manager at GSFC), and John Mather (JWST Senior Project Scientist at GSFC) regarding the status of the JWST Project to set the stage for the committee. The focus then shifts to presentations by Institute personnel, organized by Massimo Stiavelli (JWST Mission Head), Jason Kalirai (STScI Multi-Mission Project Scientist), and Neill Reid (Science Mission Head). The Director and Deputy Director participate fully in all discussions (as their schedules allow). While presentations provide an essential component, the agenda typically contains extensive discussion time (∼50%) so that the committee members can engage with the presenters, with the Institute leadership, and with each other, bringing their experience to bear, so as to assemble recommendations that focus on “maximizing JWST’s scientific productivity.”

Figure 1: A typical JSTAC meeting seen from the Chair’s perspective.

Since 2009, the JSTAC has dealt with numerous topics including: (i) First Look observations (Early Release Science – ERS); (ii) Large/Treasury/Legacy program proprietary time; (iii) the grant funding levels for General Observers (GOs); (iv) parallel observations; (v) “community fields;” (vi) the length of the Proprietary Time 1 period; (vii) duplications; (viii) proposal submission policies, particularly for Cycle 1; (ix) observing overheads for JWST; and others. Every one of these topics involves tradeoffs and subtle challenges to finding an optimal solution for a limited-life mission (required life is 5 years with a goal of 10 years) with a very large science-user community. With a cost to launch of $8B, the JSTAC is also keenly aware of the responsibility that it has, and that we all have in the scientific community, of helping to ensure that the policy-makers and taxpayers of the partner countries get the scientific returns appropriate for such a huge investment of public funding.

In this first article about the JSTAC, I would like to give an overview of the topics that have occupied the committee’s attention over the years. I encourage those interested in more detail to read the letters from the JSTAC to the Institute’s Director on the STScI JSTAC website.

First Look/ERS Observations: Some of the earliest discussions with Institute leadership centered on the question of how to ensure that the science community became appraised of the instrumental and scientific capabilities of JWST, particularly before Cycle 2 proposals were due. There was joint concern by both JSTAC and the Institute that essentially all early data could be proprietary, and thus not quickly available to community members. The recommended solution was a set of First Look observations (now called the ERS) wherein the programs carried out would exercise the key modes of the JWST instruments for a range of science consistent with the four Key science objectives of JWST. The data would be made public immediately. Datasets like this have been made available before on Hubble and Spitzer (e.g., Spitzer’s first cycle). The importance of the ERS program for JWST is discussed in the JSTAC letter and a ∼500 hr ERS is now part of the baseline plan for Cycle 1. See ERS here.

Large/Treasury Programs: The JSTAC also recognized the value of programs that provide enhanced datasets to the community. Spitzer, Chandra and Hubble have all utilized such programs with zero proprietary time (called “Legacy” for Spitzer and Chandra). For JWST, the JSTAC has recommended, that the proprietary period continue to be zero for Large, Treasury, and Director's Discretionary time. This was one of the earliest recommendations made by the JSTAC in 2010, and is now part of the baseline plan for JWST proposals. See Large Programs.

GO Funding: The importance of funding the scientific research from NASA's space science missions has been widely accepted since it was instituted for Hubble (and Spitzer and Chandra and other missions). A key question has been the level of funding needed for JWST. In late 2014 the JSTAC instituted a working group to assess the GO funding level so that a recommendation could be formulated. This group evaluated the data and experience from the three Great Observatories (Hubble, Chandra, and Spitzer) and assessed the level of funding needed for GOs for JWST. A number of factors indicate that significantly more funding is required than is currently provided for GOs on Hubble (e.g., _the on-target time per year relative to_Hubbleis expected to be ∼1.6 times greater). The JSTAC recommendations, along with the report of the working group, were communicated to the Director in its May 22, 2015 letter. See GO-Funding here.

Parallel Observations: As Hubble has shown, obtaining data in parallel can greatly enhance the overall scientific productivity of a mission. However, JWST was not baselined to take parallel science data. This cost-saving measure was understandable in the early days of the mission, but the excellent progress on the mission has opened up the possibility of further consideration of parallel capability. This topic had been discussed for many years by JSTAC. By late 2014, the JSTAC decided that it was time to recommend adding parallel capability, while recognizing that, in the remaining time before launch, not all combinations of modes and instruments could be implemented (Parallels here). Some parallel capabilities are expected to be available in Cycle 1.

Community Fields: One of the topics that elicited much discussion in JSTAC during the early years concerned those regions of the sky where large non-proprietary datasets had been assembled from Hubble(and particularly from all three Great Observatories—Hubble, Spitzer and Chandra). Given the large investment of Great Observatory time (e.g., ∼20 Msec on the HUDF/GOODS-S region) on certain fields with non-proprietary datasets, the JSTAC expressed interest in establishing such regions as open access regions, i.e., no proposal could lock up such fields with proprietary datasets. See the Community Letter1 from 2010 here. This topic was the subject of continuing (and spirited) discussion within the JSTAC, and also with the SWG and the GTOs (who have 12-months proprietary time). Eventually the JSTAC decided that just one region stood out enough in the Great Observatories datasets—the HUDF/GOODS-S/CDF-S region—such that it warranted a recommendation that GO data acquired on that region normally have zero proprietary/exclusive access period. See the 2014 Community Letter2 here.

Proprietary Time: The JSTAC recognized very early in its deliberations that a long 12-month proprietary period greatly impacted the scientific productivity of a limited-life mission. The impact on JWST science was enunciated in JSTAC’s first letter to the Director in 2010, expanded upon in a later letter in 2010, and then extensively discussed in its March 2014 letter. The JSTAC recognized, as early as in 2010, that “with a 12-month proprietary period, Cycle 4 proposals are the first able to use all Cycle 1 data to do follow-up.” In a new, extraordinarily powerful mission like JWST, particularly one baselined as a 5 year, 5-cycle mission, follow-up of science discoveries is a key part of “maximizing the science return.” If the science community cannot quickly follow up discoveries with new approaches and new observations, the scientific returns from JWST will be greatly impacted. See Proprietary Time here.

After extensive deliberation and consideration of many different approaches (e.g., <1 data-preserve-html-node="true" data-preserve-html-node="true" data-preserve-html-node="true" data-preserve-html-node="true" year proposal cycles) and other input, as detailed in the 2014 letter, the JSTAC recognized that the only approach that would make a significant difference was a 6-month proprietary period. There are many aspects that played a role in JSTAC’s thinking, but one that was very surprising to JSTAC, concerns the long time (>2 years) to publication of science data (see Figure 2). This has interesting implications for discussions of the proprietary period. Now that Hubble is moving to 6-months proprietary time, it is anomalous that JWST continues with 12 months. A comprehensive summary from early 2016 regarding proprietary time, giving updates since the 2014 JSTAC letter is in the JSTAC presentation here.

Figure 2: There is a substantial delay (on average >2 years) in the time between the acquisition of Hubble data and its publication. This is true regardless of data type or proprietary period. It is clear that many of the concerns about shorter proprietary periods are not borne out by the data on publication timescales. These Hubble data were assembled by Jeff Valenti and Karen Levay. The >2-year timescale to publication is similar on Spitzer and Chandra.

Future Topics: The JSTAC’s recommendations regarding Cycle 1 proposals will be the subject of a future article, as will a more extensive discussion of proprietary time. This latter topic has occupied the JSTAC extensively over the years and deserves a more thorough exposition. JSTAC has also set up a Data Processing Working Group (JDPWG) to explore data processing questions and issues in more depth. See the 2016 letter re the Data Processing Working Group.

Two key factors have contributed greatly to the value of the JSTAC’s recommendations: (1) the experience and commitment of JSTAC’s members, and (2) the thoughtful and thorough effort by Institute staff to provide the needed background and analysis for the committee to be well informed on issues. JSTAC would not have been able to do what it has done without the excellent, productive, and very open dialog with the Institute’s leadership and staff.

1 Note that NASA also calls “Proprietary Time an “Exclusive Use Period.”