1.1.

Concepts Before the 2000 Decadal Survey

Years before the Hubble Space Telescope (HST) was even launched and became a celebrated space observatory, the scientific and engineering communities were already discussing a much larger-aperture follow-on mission that would cover the ultraviolet, optical, and infrared (UVOIR) wavelength regime and be able to continue breakthrough science when HST was no longer available. To a remarkable degree, many of the basic design requirements and some priority scientific objectives of those flagship concepts of three decades ago, which were advocated as the natural eventual successors to HST, are reflected in the more advanced recent concepts described in this special section of the Journal of Astronomical Telescopes, Instruments, and Systems (JATIS) (Smith and McCray¹ described this time period in depth, although included several topics beyond the scope of this paper.).

Probably the earliest substantive identification of a large UVOIR space observatory specifically intended to follow HST was the National Research Council (NRC) Space Science Board (SSB) report, “Space Science in the Twenty-First Century: Imperatives for the Decades 1995—2015” (begun in 1984 and published in 1988; see summary by Field²). Volume II of this report presciently observed, “A large-aperture space telescope for the (UVOIR) regions has immense scientific potential. The need for such a telescope will be very high after 10 to 20 years of use of HST. Even now, we see that some of the most fundamental of all astronomical questions will require the power of a filled-aperture telescope of 8- to 16-m diameter designed to cover a wavelength range of 912 Å to $30\mu \mathrm{m}$, with ambient cooling to 100 K to maximize IR performance.” The 1986 Report of the National Commission on Space, “Pioneering the Space Frontier”³ (aka, the Paine report) recommended a broadly similar mission as it drew heavily on the nearly simultaneous SSB report and personnel, notably telescope operation at $\sim 100\mathrm{K}$.

Not long after the SSB and Paine reports, in his overview of the influential 1989 Space Telescope Science Institute (STScI) community workshop report, “The Next Generation Space Telescope,”⁴ Illingworth urged the participants to recognize the limited lifetime of major astrophysics missions, such as HST, and “look beyond, to the missions that will succeed the great observatories.” Illingworth summarized the basic parameters of the notional mission concept embraced at the workshop: again, a 10- to 16-m class telescope operating from $\sim 0.1\mu \mathrm{m}$ to “beyond $10\mu \mathrm{m}$” due to passive cooling to 100 K. (As Smith and McCray¹ point out on page 49, “This widely quoted apparent limit to a low temperature achievable via passive (aka, radiative) cooling alone became for some years a major hindrance. . . in achieving sensitive observations at long wavelengths.” Wide recognition that radiatively cooled optical system temperatures far colder than 100 K were possible emerged as a significant factor in the mid-1990s. This recognition contributed to an emphasis on IR observations, rather than UV/optical wavelengths, for the first major post-HST observatory as discussed later in this section.)

In the proceedings of the STScI workshop, Illingworth summarized a selection of some of the most exciting science that a large post-HST observatory would be capable of, including revealing important structures in cosmologically distant galaxies observable throughout the Universe with scale sizes of 100 to 1000 pc. (Notional details of the large UVOIR space observatory are listed on Table 1 of the workshop report.) Furthermore, Illingworth pointed out the exciting prospect of such a mission being able to detect Earth-like planets orbiting stars within 10 pc of the Sun. In the same proceedings, Angel⁵ observed that the search for Earth-like planets “was in large measure the original rationale for such a large telescope” in the SSB report published in 1988. His article, which emphasized the daunting engineering challenges to such a mission, included discussion of the required performance of an observatory able to search for biomarkers in hypothetical Earth-like planets and discussed those biomarkers considered at the time to be most revealing.

Table 1

ATLAST observatory requirements derived from science goals.

With respect to the search for life, interestingly NASA apparently first explicitly considered direct imaging of extrasolar planets with a large-aperture telescope as part of the Project Orion design study⁶ in the late 1970s. Both ground- and space-based imaging were evaluated in this study, with the final report optimistic that a 2.4-m space telescope would enable the detection of large Jovian planets around stars within 10 pc. Coronagraphic imaging of the beta Pictoris circumstellar disk by Smith and Terrile⁷ in the mid-1980s provided impetus to the first mission study work leading to a proposal for the modest-aperture, ultrasmooth Circumstellar Imaging Telescope by the Jet Propulsion Laboratory.^8,9 A major technical challenge at the time, however, was manufacturing large optics to the required smoothness, which motivated years of subsequent NASA technology investments.

Momentum in the scientific and technical communities was building. In 1991, two years after the NGST meeting at STScI, JPL hosted a workshop, “Technologies for Large Filled-Aperture Telescopes in Space,” which summarized developments necessary to enable the 8- to 16-m UVOIR telescope operating from 0.12 to $\sim 10\mu \mathrm{m}$ described in Illingworth’s¹⁰ workshop executive summary. To demonstrate the impressive capabilities of such an observatory, page 6 in the workshop report compares simulated visual-wavelength images of distant galaxies as observed by the 10 m Keck telescope, HST, and a hypothetical future 16-m space observatory. (See also the cover of Refs. 4 and 5.)

For advocates of future major space observatories, it is worth noting that less than a year had passed after launch of HST, multiple engineering assessments and community workshops of a follow-on UVOIR flagship converged on basic observatory parameters, technology investment priorities, and science objectives, including observations of nearby Earth-like worlds, broadly similar to the concepts discussed elsewhere in this special issue of JATIS. Of course, to a large degree this is simply due to the fact that major scientific objectives evolve slowly and the observational requirements to achieve those objectives are set by the basic laws of physics and optics. Note, however, these early 1990s workshops represented a high-water mark in proposing large UVOIR observatories on the lunar surface or in high-Earth orbit: Farquhar and Dunham¹¹ wrote briefly on the value of the Sun-Earth libration points for space observatories. In 1995, NASA’s very successful Solar and Heliophysics Observatory began operating at the Sun-Earth L1 point and HST had by then amply demonstrated sustained precision operation of an observatory in free space.

The mid-1990s also began an extended hiatus in the candidacy of a large UV/optical-optimized observatory for selection as the high-priority major post-HST mission for NASA. In late 1993, the Association of Universities for Research in Astronomy (AURA) empaneled an 18-person committee, chaired by Alan Dressler, to assess compelling science objectives for the coming decades. In 1996, the committee issued its influential report, “HST and Beyond: Exploration and the Search for Origins: A Vision for UV–Optical–IR Space Astronomy”^12,13 (aka, The Dressler Report). That report identified two high-priority science objectives intended to motivate selection of the mission concepts to achieve them, which were again very similar to the goals of the workshops noted previously: (1) visiting a time when galaxies were young (i.e., formation and early evolution of galaxies) and (2) the search for Earth-like worlds. To achieve these goals, the Dressler Committee made three recommendations, two of which were in a substantially different direction from previous community recommendations and would ultimately lead to selection by the 2000 NRC Decadal Survey¹⁴ of the Next Generation Space Telescope (NGST): (1) extend the lifetime of HST, (2) build a large filled-aperture IR-optimized space observatory, and (3) develop and demonstrate space interferometry. With respect to the third recommendation, although beyond the scope of this paper, space astrometry became a priority of NASA’s thinking about exoplanet observations with the establishment of a Space Interferometry Science Working group in 1991, which led to the development of the Space Interferometry Mission¹⁵. To underscore European interest, about the time of the release of the Dressler Committee report in mid-1996, a major workshop in Toledo, Spain was held on the subject of IR interferometry from space and the search for life-bearing planets.¹⁶

With HST’s flawed optics corrected in 1993, and two new instruments installed during a servicing mission in 1997, numerous individuals continued to remind the astronomical community that HST would eventually reach its end-of-life and no mission proposed to extend its science programs at UV/optical wavelengths was under consideration. Given the long gestation period for building a flagship-class mission (over $\sim 20$ years) and with the IR-optimized NGST recommended by the Dressler Report very different from the UVOIR version of NGST recommended by the STScI conference⁴ in 1989 as the next large mission to follow HST, advocacy resumed for a large UV–optical–near–IR flagship. Such a mission would now be a candidate to follow NGST.

In 1998, a workshop was held at the University of Colorado to discuss the future of UV–optical astronomy from space.¹⁷ The meeting consensus was that the next step after HST at UVOIR wavelengths should be a 4- to 6 m-class instrument to complement NGST, an aperture notably more conservative than that generally advocated a decade earlier. In response to this workshop, NASA’s UV–optical working group (UVOWG) was commissioned to study the scientific rationale for new missions in the UV–optical bandpass. This group produced a report¹⁸ that also recommended (1) a 4-m aperture telescope that emphasized wide-field imaging and UV spectroscopy and (2) investigation into the feasibility of an 8-m telescope with deployable optics similar to NGST. The 4-m concept, dubbed the Space Ultraviolet Observatory, was subsequently proposed to the 2000 Decadal Survey. Technology development was identified as a priority by a Survey panel, although not recommended by the full committee. The search for Earth-like worlds was a high priority for the Survey for which the IR-optimized Terrestrial Planet Finder (TPF; see Sec. 1.2) was the recommended mission to eventually achieve this NASA goal, although the 2000 Decadal Survey’s highest-priority major recommendation was NGST.

1.2.

Preparing for the 2010 Decadal Survey

1.3.

The NASA Astrophysics Roadmap: Vision for the Next Three Decades

In spring 2013, the Astrophysics Subcommittee of the NASA Advisory Council’s Science Subcommittee chartered a community-wide task group to develop a long-range vision for NASA’s Astrophysics Division that would span the subsequent three decades. This vision built upon the products of the 2010 Decadal Survey and included science objectives likely to remain priorities until the middle of the century; thus the name of its final report, released in late 2013: “Enduring Quests, Daring Visions: NASA Astrophysics in the Next Three Decades.”⁴² The roadmap also identified key technology investments necessary to enable the missions recommended to NASA Headquarters’ Astrophysics Division.

The scientific objectives, mission concepts to achieve them, and technologies required to enable them were divided into three time periods, each about a decade long, beginning with the near-term missions already under development (e.g., JWST and WFIRST). Following this period was the Formative Era, which identified for NASA science priorities now familiar from multiple conferences and workshops referenced previously, with a special emphasis on the search for and characterization of exoplanets – and perhaps even Earth-like worlds—in the solar neighborhood, as well as the birth and evolution of galaxies, stars, and planets. To achieve these goals, the report recommended, as had many others before it, a large UV/optical/near–IR mission now dubbed the LUVOIR Surveyor, which was described in the final report as having an aperture of 8 to 16 m with wavelength coverage from “near–IR to near–UV,” the specific wavelengths to depend upon technology development.

The roadmap technologies are consistent with the more detailed description developed by Bolcar et al.⁴³ for the ATLAST concept described in Sec. 2: precision deployment and wavefront control, mirror coatings, detector systems, and—very critically—high-performance starlight suppression.

1.4.

Association of Universities for Research in Astronomy’s High-Definition Space Telescope

In early 2013, the Association of Universities for Research in Astronomy (AURA) chartered a team of seventeen scientists and technologists to assess further how the challenging dual goals of cosmic origins science, especially of extremely distant objects and processes, and the search for life-bearing planets, could be combined into a single mission. Released in mid-2015, the report, “From Cosmic Birth to Living Earths,”⁴⁴ described in depth the science cases and technology drivers for a space telescope concept, dubbed the High-Definition Space Telescope (HDST). While the basic HDST concept was broadly similar to earlier UVOIR observatory concepts and science goals and the ATLAST designs of the last decade, its science drivers and design concepts had progressed substantially.

Significant scientific advances over the performance of HST at these wavelengths require developing the capability to deploy an aperture much larger than can currently be accommodated within the inner diameter of existing launch vehicles, as earlier concepts described (e.g., Refs. 4 and 10). Success with JWST continues to build confidence that precision deployment of very large segmented optical systems is a successful engineering solution. To give an indication of the advancements required in deployed optics, Fig. 1 compares the relative sizes of the primary mirror for HST, JWST, and HDST. In the particular concept shown in the figure, the HDST primary is made up of 36 1.7-m segments, although segments of different sizes could be adopted.

Fig. 1

Relative sizes of the primary mirrors for HST, JWST, and HDST (Fig. 5-1 from Ref. 44).

The AURA study developed a notional instrument suite that permitted realistic estimates of observatory performance consistent with being both a powerful general-purpose flagship, as well as capable of detecting biomarkers in the candidate Earth-like worlds. HDST was proposed to have 25 times the pixel density per area of HST at the same optical wavelengths, than four times better resolution at near–IR wavelengths JWST, and up to 100 times the point-source UV spectroscopic sensitivity. Furthermore, HDST was proposed to have multiobject UV spectroscopy for up to 100 sources in an $\sim 3\mathrm{arcmin}$ field of view, as well as extremely stable wavefronts to provide precise point-spread functions over long observational timelines.

For galaxies out to cosmological distances, a mission such as HDST would have the capability to reveal features on a scale size of 100 pc or smaller (Fig. 2), which will reveal structures and processes critically important to major leaps in our understanding of the cosmos.

Fig. 2

An observatory with an aperture the size proposed for HDST will permit observations on a scale smaller than 100 pc throughout the universe (Fig. 4-1 from Ref. 44).

Images from HST have thrilled both scientists and the general public for a quarter century and a mission such as HDST will do no less. Figure 3 compares synthetic images of a galaxy at high redshift as would be observed by HST, JWST, and a mission such as HDST. The breathtaking capability of such a mission would continue public and professional support for flagship observatories exploiting the limits of optical design and instrument sensitivity.

Fig. 3

Comparison of synthetic images taken of an extremely distant ($z=2$) galaxy as would be revealed by three flagship observatories. Diffraction-limited angular resolution of each observatory is indicated in each of the three smaller insets (Fig. 4-5 in Ref. 44).

Both the most challenging scientific goal proposed for HDST, as well as for what many believe is its most exciting, is the study of UVOIR “biomarkers” in the spectra of Earth-like worlds orbiting neighboring stars. This goal requires a very large aperture in space in order to be able to (1) observe a sufficiently large number of stars to provide meaningful constraints on the occurrence of potentially habitable exoplanets and (2) acquire the spectra of those exoplanets, which are extremely faint, and (3) to resolve the angular separation between the exoplanet from its host star, as discussed in the work summarized in Sec. 1. Technologies to permit extremely stable wavefront control of a large aperture, as well as starlight suppression to about one part in 10 billion (${10}^{10}$) via a coronagraph and/or starshade, were identified in the AURA report for HDST as priority investments to make the concept possible.

2.1.

Scenario Planning for the Advanced Technology Large-Aperture Space Telescope

Our scenario planning activity for the study summarized here recognized, e.g., uncertainties in major science goals for this wavelength range in the 2020s, characteristics of plausible future launch vehicles at that time, as well as capabilities for astronauts and robotic systems at notional operations’ site for an ATLAST-type mission. Moreover, at the time we began our assessment, NASA HQ had yet to prepare plans for the 2020 Decadal Survey. Our final reference design was, therefore, intended to be flexible enough to be a successful candidate for future development, regardless of the scientific, programmatic, and engineering environment of the 2020s. In addition, as mandated by Congress, the ATLAST designs that our team assessed were all intended to be serviceable, if not actually serviced, when eventually operational.

Basic science requirements were derived from the ATLAST design from last decade,³⁷ and updated by our team at the start of our current assessment in 2013. These requirements were consistent with those described in NASA’s 30-year roadmap⁴² and AURA’s HDST concept⁴⁴ and determined reference design requirements, as summarized in Table 1 and notional candidate instruments summarized in Table 2. In our scenario planning, our team was unambiguous about the design being capable of carrying out a broad range of astronomical investigations, including our team’s priority of the challenging search for and characterization of Earth-like worlds in the solar neighborhood. We concluded that an aperture of $\sim 10\mathrm{m}$ was a compellingly attractive advance in general scientific capabilities over HST. Moreover, it is larger than the minimum aperture ($\sim 8\mathrm{m}$) that we judged would produce a sample size of candidate exo-Earths of sufficient size to permit us to confidently estimate statistically the yield of life-bearing planets in the solar neighborhood.⁴⁶

Table 2

Notional ATLAST instrument candidates.

2.2.

A Deployable Concept for Advanced Technology Large-Aperture Space Telescope

By adopting a segmented design as one concept for ATLAST, we had a basic design that was adaptable to different launch vehicles by increasing or decreasing the size of the segments, their number, or the adopted deployment scheme.

Our engineering design reference mission (EDRM) consists of a deployable primary mirror. Consistent with our scenario planning, the aperture is scalable, meaning that its engineering design supports adding more rings of segmented mirrors to increase the aperture in response to the refinement of the science requirements or availability of larger launch vehicles. With study resources at first limited largely to internal funds, we were constrained to assess in depth only a single segmented design, a 9.2-m configuration. This aperture was validated to fit within a 5-meter launch vehicle fairing, an industry standard, which we chose as part of our strategy to build confidence that our design is feasible and costs are controllable. An image of the final ATLAST reference design for this concept from mid-2015 is shown in Fig. 4. The 9.2-m aperture is made up of 36 hexagonal segments and reflects the design heritage of JWST.

Fig. 4

Final visualization of the ATLAST segmented reference design from mid-2015.

2.3.

A Monolith Concept for Advanced Technology Large-Aperture Space Telescope

In addition to the chord-fold deployable concept, our EDRMs included concepts using an 8-m monolith primary mirror and a monolith surrounded by deployable mirror petals.

The latest iteration of the 8-m monolith concept has been led by the ATLAST team at MSFC and is described by Stahl et al.⁴⁷ A major appeal for this design is the advantage of not having gaps due to segments in the primary mirror, which is characteristic of the deployable options for ATLAST. A monolith thus provides advantages with regard to some current coronagraph designs. However, 8 m is the approximate upper limit for a monolithic aperture telescope. Consequently, as described also by Stahl and Hopkins,⁴⁸ the MSFC team also produced a 12-m concept consisting of a central monolith surrounded by a ring of identical petals. Both of these concepts, as well as segmented designs much larger than described in Sec. 2.2, would require the space launch systems (SLS) Block II launch vehicle with either its 8- or 10-m fairing. This vehicle is slated for eventual development and there is at present no alternative means of launching the largest ATLAST apertures. Depending on the SLS mass capacities, the 8-m monolith concept could use existing ground based mirror technology or the new “deep-core” mirror technology being developed via NASA’s advanced mirror technology development program.⁴⁹