I

Report of the Panel on Electromagnetic Observations from Space 1

SUMMARY

The panel on Electromagnetic Observations from Space 1 (EOS-1) was constituted to examine the state of ultraviolet (UV), optical (O), and infrared (IR) observations from space. This wavelength coverage is from approximately 0.09 to 5 μm. Observations in this wavelength range have been dominated by the Hubble Space Telescope (HST), and the IR wavelengths will be dominated in the near future by the James Webb Space Telescope (JWST). The Nancy Grace Roman Space Telescope (formerly the Wide-Field Infrared Survey Telescope [WFIRST]) will also make substantial contributions at near-IR wavelengths. The panel is charged with surveying the ability of these current near-term activities, as well as assessing the capabilities of proposed activities, to address the compelling science challenges identified by the science panels convened as part of the decadal survey. The panel is also charged with reviewing the white papers pertinent to the UV/O/IR activities in space. A total of 67 white papers were received by EOS-1. Additional written input included National Aeronautics and Space Administration (NASA)-funded mission study reports from the two flagships and five of the probes considered by this panel. The panel also benefited from the National Academies of Sciences, Engineering, and Medicine Exoplanet Science Strategy report¹ from 2018. The statement of task is given in Appendix A.²

The Astro2020 science panels outlined 24 questions and 6 discovery areas that will define progress for the next decade in astrophysics. Many of these areas need UV/O/near-IR data. The area of exoplanet research has seen enormous progress toward finding life elsewhere in the galaxy with great strides since the 2010 New Worlds, New Horizons in Astronomy and Astrophysics³ (Astro2010) survey report.⁴ First was

___________________

¹ National Academies of Sciences, Engineering, and Medicine, 2018, Exoplanet Science Strategy, The National Academies Press, Washington, DC.

² See Appendix A for the overall Astro2020 statement of task, for the set of panel descriptions that define the panels’ tasks, and for additional instructions given to the panels by the steering committee.

³ National Research Council, 2010, New Worlds, New Horizons in Astronomy and Astrophysics, The National Academies Press, Washington, DC.

⁴ National Academies of Sciences, Engineering, and Medicine, 2016, New Worlds, New Horizons: A Midterm Assessment, The National Academies Press, Washington, DC.

the unambiguous detection of an exoplanet in a direct image followed by NASA’s Kepler mission census, which reveals that there are large numbers of exoplanets with many more on the way from the Transiting Exoplanet Survey Satellite (TESS). The number of direct images of exoplanets and disks at visible and near-IR wavelengths has gone from a few in 2010 to a few dozen in 2020, thanks to new ground-based instruments that were deployed over the past decade, such as the Gemini Planet Imager (GPI), Spectro-Polarimetric High-contrast Exoplanet REsearch (SPHERE), the Subaru Coronagraphic Extreme Adaptive Optics (SCExAO) project, and the Large Binocular Telescope Interferometer (LBTI). All of these have advanced science as well as technology. Technological progress has made approaching the Panel on Exoplanets, Astrobiology, and the Solar System (EASS) discovery area of “The Search for Life on Exoplanets” potentially feasible. The panel recognizes the challenges inherent in searching for biosignatures, but progress has made this goal achievable and serves to organize and measure the best of NASA’s energies and skills. This goal will capture the imagination of all humankind, imply technical capabilities that will serve a great majority of the astronomical community beyond the field of exoplanets, and tie in to solar system and Earth science.

Other science areas will benefit from proposed missions that are aimed primarily at searching for life, but which provide a compelling suite of capabilities that can address other questions. Astronomy has made rapid progress when panchromatic data are available, and for this panel, of particular importance are the UV spectroscopic capabilities, which will surpass those available on HST, and which will address many issues identified by the science panels such as observing the circumgalactic medium (CGM) in emission lines and providing UV imaging and spectra for transient events such as mergers of compact objects.

EOS-1 considered several implementations of a mission aimed first at detecting biosignatures and with capabilities for expanding the understanding of exosolar systems in general and with capabilities for enabling broad ranging observations, especially significant at UV wavelengths. The panel is suggesting further study and technology development that could lead to a mission with the light gathering power of at least a 6-m primary mirror, and equipped with a light suppression system capable of achieving the goal of the 10^–10 contrast needed for detection of Earth-like planets in the habitable zone of solar-type stars. The mission will also need focal plane instrumentation to acquire images and spectra over the range of 100 nm to 2 μm with parameters similar to cameras and spectrometers proposed for the Large UV/Optical/IR Surveyor (LUVOIR) and the Habitable Exoplanet Observatory (HabEx). These instruments would include moderately wide field imaging at UV, optical, and near-IR wavelengths as well as multi-object spectroscopy over a similar wavelength range (see Table I.5, later in this appendix). The panel is not suggesting a named mission now because it is premature to do so this far ahead of when actual development could start, a situation caused by the fact that two astrophysics flagships are still being completed, as well as budgetary considerations. By not naming a particular mission configuration now, the panel is stating that more work is needed to ensure that a future mission’s budget envelope is well-constrained and that the mission will achieve its primary science goals given the uncertainties in parameters such as η_Earth. These points are discussed in greater detail in Section I.3, Flagships, below.

The panel reached its conclusion for a future flagship mission based on the following observations, which are described more fully later in this report. The panel viewed it as essential that a flagship-class mission be capable of achieving a compelling result for the first time while also being capable of supporting a broad range of other science investigations. First, both the LUVOIR and HabEx teams presented a convincing case that astronomers are positioned to make a serious attempt at searching for biosignatures on exoearth candidates, which is the compelling result projected for these missions. This goal aligns with one of the principal recommendations of the Exoplanet Science Strategy report that “The National Aeronautics and Space Administration (NASA) should lead a large strategic direct imaging mission capable of measuring the reflected-light spectra of temperate terrestrial planets orbiting Sun-like stars.”

The EASS panel states that searching for evidence of life “is only possible with a large, high-contrast, direct-imaging space telescope.”⁵ Neither JWST, the Roman Space Telescope, nor the Extremely Large Telescopes (ELTs) are predicted to have coronagraphs with adequate contrast, whereas small missions will have inadequate light-gathering capability for the needed spectroscopy of an exoearth. Second, any of the mission configurations that the panel evaluated that would be capable of finding more than one or two candidate exoearths require too much funding in the peak development years. Even the smallest mission considered in detail, HabEx 3.2S, requires a substantial increase in the budget allocation for new missions or else the mission would use almost all of the Astrophysics budget in its peak years according to the Technical, Risk, and Cost Evaluation (TRACE) analysis (see Appendix O). These budget concerns are exacerbated by the panel’s finding that a probe-class mission line with two launches per decade might also be added to the Astrophysics portfolio. Third, substantial technology development is still needed despite recent progress in starlight suppression techniques and ultra-stable telescopes, with at least 5 years and several hundreds of millions of dollars needed. The panel recognizes that for this level of technology development, it would be usual to choose a mission architecture to focus the technology development, but given the budgets and time scales needed, choosing a specific mission now is premature. But equally, the panel would like to ensure that substantive progress is made toward an eventual choice of a flagship capable of detecting exoearths and biosignatures, so the panel has considered how to achieve getting technologies to the point where a mission could start as early as the late 2020s and preferably launch no more than a decade later.

Observation: Considerable progress has been made in improving starlight suppression performance, which suggests that the desired 10-10 contrast ratio needed for direct detection of Earth-like exoplanets around Sun-like stars is achievable with adequate resources.

Observation: Progress in defining biomarkers as outlined in Exoplanet Science Strategy (2018)⁶ suggests that 0.1 to 2 μm is a rich wavelength regime, including the UV ozone feature as a robust indicator of oxygen in an exoplanet atmosphere (see Figure I.1, later in this appendix).

The panel also reviewed 11 probe-class missions. Two versions of the Cosmic Evolution Through UV Surveys (CETUS) probe concept were presented to the panel. The science goals and observational capabilities for the two versions are almost identical, with implementation being the major difference. In the remainder of this report, the two versions are treated as the same suite of observational capabilities. Collectively, these probe missions present a suite of capabilities that would address a number of the science questions identified by the science panels. While probes cannot fulfill all of the observational needs from space for astrophysics, they could provide very valuable data, such as multi-wavelength observations for time-critical observations that will be difficult to provide from flagships only. Probes have the potential to broaden the suite of capabilities available to astronomers and to respond to changing science priorities more quickly than flagship missions can. The panel prefers open competition for probe launches.

Observation: A variety of probe-class missions would be capable of delivering some but not all of the high-priority science identified in the Science Panel reports (see Table I.3, later in this appendix).

___________________

⁵Appendix E, section on Discovery Area: The Search for Life on Exoplanets.

⁶ National Academies of Sciences, Engineering, and Medicine, 2018, Exoplanet Science Strategy, The National Academies Press, Washington, DC.

I.1 PANEL INPUTS AND CONSIDERATIONS

The panel met face-to-face two times, remotely once, and conducted eight teleconferences to examine the state of UVOIR activities from space. The face-to-face and remote meetings concentrated on presentations from representatives of the two flagship missions that fall under the panel’s purview, LUVOIR and HabEx, and on presentations from representatives of all of the probe-class (~$1 billion missions) that fall in the UVOIR wavelength range. For the flagship missions, the panel sent questions to the teams both before and after presentations, although “export control” markings hampered evaluation of some responses. Resources to subject missions to the TRACE process were limited, so the panel chose to focus primarily on LUVOIR-B and on two versions of HabEx, 4H and 3.2S, as a balance between cost and exploring a range of implementation options. Some of the probe missions received some limited study funding from NASA, while others were submitted to the decadal survey solely as white papers. The panel considered probe-class missions that received NASA funding, including CETUS, Cosmic Dawn Intensity Mapper (CDMI), Transient Astrophysics Probe (TAP), Starshade Rendezvous, and Earthfinder (proof of concept study only). Exo-C received NASA study funding earlier than these missions. Other probe-class missions considered that did not receive NASA study funding include Occulting Ozone Observatory (OOO), an alternative formulation of CETUS, AstroNomical Uv proBe Imager & Spectrograph (ANUBIS), Astrophysics Telescope for Large Area Spectroscopy (ATLAS), and Nautilus.

The panel also heard from invited speakers on some of the techniques germane to missions under consideration. Talks on how starshades and coronagraphs work were included in the first meeting. The first meeting also included presentations on the status of technologies for these two starlight-suppression techniques. Because of its significance for predicting the exoearth yield from the flagship missions, the panel heard a presentation on the current state of knowledge of η_Earth at the second meeting. An additional talk on the Roman Space Telescope coronagraph was included in the panel’s third meeting. The panel also read 67 white papers submitted by the community.

The Aerospace Corporation briefed the panel on its TRACE results for LUVOIR B and two versions of HabEx at the panel’s third meeting. The panel also had access to the Large Mission Concept Independent Assessment Team report on LUVOIR and HabEx, and the Probe Cost Assessment Team report on the NASA-funded probe studies.

The panel also assessed the current near-term space capabilities in the UV-optical-IR wavelength range to see how the flagship and probe missions presented to the panel fit into the current observational opportunities. The range covered by JWST (0.6 to 28 μm) will be well-supported by the capable imagers and spectrometers provided on that mission. Study of exoplanet atmospheres via transmission spectroscopy observed in a transit will be the prime exoplanet observational mode on JWST. The Roman Space Telescope will include a demonstration coronagraph and will provide near-IR surveys over broad areas of the sky, and a range of small missions are enhancing many aspects of exoplanet investigations such as TESS, Atmospheric Remote-sensing Infrared Exoplanet Large-survey (ARIEL), PLAnetary Transits and Oscillations of stars (PLATO), CHaracterising ExOPlanets Satellite (CHEOPS).

I.1.1 Major NASA Operating Missions

The HST remains one of the world’s premier astronomical observatories 30 years after it was placed in low Earth orbit by the space shuttle. Its suite of imaging and spectroscopic instrumentation, restored to full functionality in the final HST Servicing Mission in 2009, covers the UV to near IR (0.1–1.7 μm). HST’s scientific accomplishments are legion, and the observatory remains in great demand by astronomers across the world. It has the only UV spectroscopic capability for the foreseeable future. While HST is no

longer serviceable with current capabilities, its orbit is stable against reentry until the 2030s, although a gyro failure could limit its operational lifetime.

The Stratospheric Observatory for Infrared Astronomy (SOFIA) is a telescope with an effective diameter of 2.5 m, carried aboard a Boeing 747-SP aircraft. The observing altitudes for SOFIA are between 37,000 and 45,000 feet, above 99 percent of the water vapor in Earth’s atmosphere. The present instrument suite provides coverage from 0.3 to 612 μm. New instrumentation is currently in development. The current instrumentation is complementary to the future capabilities of JWST, providing imaging and spectroscopy (R ~100 and R >10,000) in the mid-IR. Other instruments cover 30–600 μm with imaging, polarimetry, and high-resolution spectroscopy. SOFIA is made possible through a partnership between NASA and the German Aerospace Center (DLR). The observatory’s mobility allows researchers to observe from almost anywhere in the world. SOFIA has conducted regular observing campaigns in the Southern Hemisphere.

The Neil Gehrels Swift Observatory was launched in 2004 to study gamma-ray bursts (GRBs) with a range of gamma-ray and X-ray instrumentation. The mission has been a huge scientific success. Of particular interest to the EOS-1 panel is the UV-optical telescope (UVOT) that is part of Swift’s instrumentation suite. The UVOT (0.17–0.6 μm) takes images and grism spectra of GRB afterglows during pointed followup observations. The images are used for 0.5 arcsecond position localizations and following the temporal evolution of the UV-optical afterglow. Spectra can be taken for the brightest UV-optical afterglows, which can then be used to determine the redshift via the observed wavelength of the Lyman-alpha cut-off. The UVOT has made significant contributions to the study of a range of transient phenomena, not just GRBs.

NASA’s TESS is designed to search for exoplanets using the transit method of 200,000 nearby stars over 85 percent of the sky, an area 400 times larger than that covered by the Kepler mission. It was launched in 2018 and is expected to find more than 14,000 transiting exoplanets, compared to about 3,800 exoplanets known when it launched. As of March 2021, TESS has identified more than 2,645 candidate exoplanets, of which 122 have been confirmed so far. In January 2020, NASA announced the discovery by TESS of the first Earth-size planet in its star’s habitable zone. TESS will provide prime targets for further characterization by the JWST, as well as other large ground-based and space-based telescopes.

I.1.2 International Operating Missions

The European Space Agency’s (ESA’s) Characterizing Exoplanet Satellite (CHEOPS) is the first mission dedicated to studying bright, nearby stars that are already known to host exoplanets, to make high-precision observations of the exoplanet’s size as it passes in front of its host star. CHEOPS was launched in December 2019. It is focused on planets in the super-Earth to Neptune size range, with its data enabling the bulk density of the planets to be derived.

ESA’s Gaia is an astrometric mission to measure the positions and velocities of ~1 billion stars in the Milky Way Galaxy. These measurements are being used to construct a three-dimensional map of the galaxy. The observational phase of the mission is complete, but data analysis is still in process, with the full Data Release 3 (DR3) release expected in 2022. The mission is having both broad science impacts and an impact on operations of future missions with exquisite star positions that enable much better pointing than previously achievable.

I.1.3 Approved Missions in Development

The JWST is a NASA flagship mission, in partnership with ESA and the Canadian Space Agency (CSA), in development for launch in 2021. JWST will provide significantly improved infrared angular resolution (0.031 arcsec at 2 μm) and sensitivity over HST and the now retired Spitzer. It is designed to enable a

broad range of investigations across astrophysics, including finding and studying the first galaxies in the early universe. JWST will provide major capabilities for studying exoplanets and planet formation across its 0.6–28 μm wavelength range. All four scientific instruments have observing modes designed for exoplanet transits. Three instruments have coronagraphic or aperture mask imaging to obtain direct images of exoplanets. JWST will orbit the Sun at the Sun-Earth L2 point.

JWST is a new type of space telescope design. To achieve the cryogenic temperatures (<50,000 mK [millikelvin]) necessary for low-background IR observations, the telescope and scientific instruments are open to deep space and cool passively, and remain cold, because they are shielded from the Sun’s and Earth’s heat by a large sunshield. The sunshield and the telescope are folded for launch and are deployed in space. The deployable sunshield, telescope optics, and large composite structures are among the key technologies being demonstrated by JWST that are relevant for future mission designs.

The Nancy Grace Roman Space Telescope (formerly the Wide-Field Infrared Survey Telescope [WFIRST]) is a NASA mission to survey wide swaths of the sky at near-IR wavelengths to address fundamental questions about the nature of dark energy, to provide a statistical basis for understanding exoplanetary system architectures via microlensing that is free of the biases of transit studies, to demonstrate a more capable coronagraph than ever used previously in space (CGI), and to provide wide-area near-IR imaging for guest observer programs. Wide-field imaging will be performed from 0.5 to 2 μm, with a spatial resolution of 0.11 arcsec. All of these programs are as vital as they were when the mission was ranked highly in Astro2010 and when the coronagraph was added. The panel finds the mission compelling, and the CGI is a useful technology demonstration that will test deformable mirrors in the space environment and also closed-loop wavefront control. The anticipated launch date is in late 2025.

Euclid, named after the ancient Greek mathematician, is a visible to near-IR mission currently under development by the ESA for launch in 2022. The objective of the Euclid mission is to better understand dark energy and dark matter by accurately measuring the acceleration of the universe using gravitational lensing, baryon acoustic oscillations, and measurement of galactic distances by spectroscopy. Euclid will measure the shapes of galaxies at varying distances and investigate the relationship between distance and redshift out to z ~2. The link between galactic shapes and their corresponding redshift may reveal how dark energy is related to the acceleration of the universe. Euclid employs a 1.2 m telescope as compared to Roman’s 2.4 m telescope, and includes a slitless grism capability. Euclid does not have a coronagraph.

Spectro-Photometer for the History of the Universe, Epoch of Reionization, and Ices Explorer (SPHEREx) will perform an optical to near-IR all-sky survey to measure the spectra of approximately 450 million galaxies. SPHEREx is a NASA medium-class Explorer (MIDEX) mission planned for launch in 2023. SPHEREx will use a spectrophotometer for its all-sky survey that will record images in 96 wavelength bands from 0.75 to 5.0 μm. It has a single instrument with a single observing mode and no moving parts to map the entire sky four times during its nominal 25-month mission. The key technology is a linear variable filter that shifts the wavelength of the imaging bandpass. SPHEREx will measure galaxy redshifts, categorize galaxies, and fit measured spectra to a library of galaxy templates. SPHEREx will probe signals from the intra-halo light and from the epoch of reionization. SPHEREx will also contribute disk observations and data on the molecular content of stellar nurseries.

Planetary Transits and Oscillations of Stars (PLATO) is an ESA mission designed to detect Earth-like planets in the habitable zones of solar-type stars. By combining space-based visible light photometric transit measurements with ground-based radial velocity measurements, an assessment of the bulk properties of the detected planets will be possible that in turn will provide an indication of which planets might be habitable. PLATO, although superficially similar to TESS, differs in that it is not an all-sky survey like TESS but rather will point at a suite of FGK stars for long periods with the goal of detecting small planets with long periods. The mission is scheduled for launch in 2026.

Atmospheric Remote-Sensing Infrared Exoplanet Large-Survey (ARIEL) is an ESA mission in development for a 2029 launch. It has a 1.2 m × 0.7 m off-axis primary mirror equipped with infrared spectrometers covering 1.2 to 7.8 μm and visible light photometry. The telescope is cooled to 55 K and the focal plane detectors are cooled to ~42 K. The goal is to survey 1,000 transiting planets orbiting F to M stars. The survey will provide a statistical sample of exoplanetary atmospheres that can be used to address questions such as how the stellar environment affects exoplanet atmospheres, whether atmospheric compositions shed light on possible planetary migration, how atmospheres may evolve over time, and many related issues.

Observation: The current suite of missions, including those soon to be launched, do not include any replacement for the aging UV spectroscopic capabilities available with HST, nor do they include any far UV capabilities.

I.1.4 Technology Progress

The 2010 decade saw rapid and significant advances in many technology areas relevant for UVOIR astronomy, especially in exoplanet detection and characterization. A segmented telescope design has been realized with JWST. Other work has examined how to achieve even greater wavefront stability. Test-bed coronagraphs are within a factor of 10 of the desired contrast, and adaptive wavefront control systems capable of removing even secondary mirror strut artifacts have been developed. Post-processing techniques have been demonstrated on HST data to reveal sources previously missed. UV coatings for dramatically improved throughput have been tested. A number of white papers highlighted these technology developments that are important for the missions that the panel considered:

• Telescopes: Development of technologies/concepts for ultra-stable telescopes (e.g., Ultra-Stable Large Telescope Research and Analysis [ULTRA] study; and white papers [WP] submitted by Coyle,⁷ East,⁸ Feinberg,⁹ Nordt,¹⁰ and Wells¹¹); active mirror technologies (WP Lawrence).
• Starlight suppression: Starshades (WP Short). Subscale demonstration achieved 1.2 × 10^–10 contrast; stowage and deployment design developed and demonstrated on sub-scale prototypes; coronagraphs (WP Shaklan, Mazoyer). Broadband lab demonstrations include 3.8 × 10^–10 contrast (monolithic aperture, Decadal Survey Testbed/High Contrast Imaging Testbed [DST/HCIT]); ~1 × 10^–9 static and ~1 × 10^–8 dynamic contrasts (Roman Space Telescope aperture). In addition, coronagraph designs have significantly improved in performance, especially for obstructed apertures, and are continuing to improve every year. See also Table I.7, later in this report.
• Adaptive wavefront control: (e.g., WP Pueyo; WP Kasdin) Micro-Electro-Mechanical Systems (MEMS) deformable mirrors (DMs) successfully operated in vacuum in the lab and on a stratospheric balloon (Planetary Imaging Concept Testbed Using a Recoverable Experiment [PICTURE]); number of actuators has increased from typically ~1K to ~2K; algorithms advanced to where DMs can remove struts and

___________________

⁷ L. Coyle, S. Knight, A. Barto, C. Allard, S. Lipscy, M. East, C. Wells, K. Havey, C. Sullivan, L. Allen, J. Arenberg, T. Lawton, K. Patton, B. Hellekson, A. Van Otten, M. Bluth, M. Nielsen, L. Pueyo, and R. Soummer, 2019, “Ultra-Stable Telescope Research and Analysis (ULTRA),” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

⁸ M. East and C. Wells, 2010, “ULTRA Segment Stability for Space Telescope Coronagraphy,” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

⁹ L. Feinberg, B. Hayden, B. Saif, M. Bluth, S. Park, P. Greenfield, and R. Keski-Kuha, 2019, “Ultra-stable Technology for High Contrast Observatories,” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

¹⁰ A. Nordt and L. Dewell, 2019, “Non-Contact Vibration Isolation and Precision Pointing for Large Optical Telescopes,” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

¹¹ C. Wells and R. Egerman, 2019, “HabEx Primary Mirror: ULTRA Segment Stability for Space Telescope Coronagraphy,” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

• segmentation of apertures, suppress binary stars, as well as mitigate telescope instability. The latter is especially important and is a lever that has not yet been fully appreciated or utilized (WP Pueyo; WP Crooke et al.), and has the potential to bring stability requirements of LUVOIR and HabEx within range of current JWST segment drift requirements.
• Post-processing methods: Ground-based post-processing of directly imaged planets has matured significantly, with powerful methods such as Karhunen-Loeve image processing (KLIP) now being standard.¹²
• UV coatings: (WP Sheikh) Can improve telescope throughput and enable a smaller telescope to achieve results that would have required much larger telescope in the past.
• Photonic and related devices: (WP Jovanovic; WP Van Buren) May enable breakthroughs in the size and robustness of focal plane instrumentation.

I.1.5 Science Panel Inputs

A crucial component in guiding the formulation of a program for the future is the identification of the most compelling science questions to be addressed in the next decade. The high-priority questions and discovery areas identified by the science panels and corresponding observational capabilities have been mapped to the capabilities proposed for the flagship missions under this panel’s purview. Table I.1 presents the mapping with the questions identified by panel as shown after the table. The entries in the table show which proposed instrument (see Table I.5, later in this appendix) on the flagship would be used to address these questions. The wide-field imagers and multi-object spectrometers on these missions are very similar, which is why the missions address similar science questions. To a large extent, missions make progress on the question in proportion to their light-gathering capability. Note that shaded entries indicate that the flagship does not satisfy all the requirements needed for the science panel questions, but does provide some capability to address the questions posed, typically by virtue of covering some of the relevant wavelengths at similar spectral resolutions. The table indicates that the flagships considered have a wide scientific reach that goes beyond the exoearth search. The science panels each provided four question areas, and missing entries such as the Panel on Cosmology Question 1 (COS-1) indicate that the EOS-1 panel missions provide no relevant capability in that area. Note that although LUVOIR-B and HabEx missions can address similar science areas, the larger size of LUVOIR-B enables deeper studies with higher quality data which come closer to answering the questions posed by the science panels. The HabEx missions are too small to address some of the science areas effectively, such as those addressing z ~7 UV luminosity functions, which is part of the Panel on Galaxies Question 1 (GAL-1).

TABLE I.1 Mapping of Flagships to Science Panel Questions and Discovery Areas

___________________

¹² R. Soummer, L. Pueyo, and J. Larkin, 2012, “Detection and Characterization of Exoplanets and Disks Using Projections on Karhunen-Loève Eigenimages,” The Astrophysical Journal Letters 755:2, https://doi.org/10.1088/2041-8205/755/2/L28.

NOTE: LUMOS: LUVOIR Ultraviolet Multi-Object Spectrograph; HDI: High-Definition Imager; UVS: Ultraviolet Spectrograph; HWC: HabEx Workhorse Camera; SSI: Starshade Instrument.

Another way to judge the breadth of the science achievable with a flagship mission comes from Table I.2, taken from the LUVOIR final report, which lists “Signature Science Cases” indicative of the sweep of science that can be addressed with such a flagship. Science cases 1–3 are associated with the grand goal of Earth-like planet detection, while cases 4 and 5 address other exoplanet science, attainable even if the contrast goal were not met. These cases map directly onto the questions posed by the EASS panel. Case 6

TABLE I.2 Proposed LUVOIR Signature Science Cases

contributes to other questions posed by the EASS panel about the relationship of the solar system’s architecture to other systems’ architectures. Cases 7, 10, 11, and 12 are related to Galaxies panel questions. Case 8 contributes to work on the Cosmology panel’s question 2. Case 9 is related to GAL-1, but the GAL question is pointed at higher redshifts (7 < z < 9), while case 9 is aiming at sources up to z ~7. The EOS-1 wavelength range is crucial for studying galaxy evolution as reflected in the capabilities of JWST’s instruments. Data from JWST will no doubt completely revolutionize this field.^13,14,15 The Panel on Galaxies has indicated that wide-field and very wide field spectroscopy at 0.32 to 5 μm will be essential for addressing questions such as measuring the characteristics of the first stars, galaxies, and black holes. This wavelength regime is also important for taking a census of supermassive black hole (SMBH) growth and determining the threshold for galaxy formation, with LUVOIR and HabEx covering wavelengths up to 2 μm. The Panel on Galaxies also points out the need for measurements at 0.09 to 0.32 μm to connect low-redshift galaxies to high-redshift galaxies. The discovery area for the Panel on Galaxies is mapping the circumgalactic and intergalactic media in emission, which also requires UV spectroscopy at these wavelengths.

Table I.3 presents a science panel mapping for the probe-class mission considered by the panel. The entries in the table refer to the focal plane instrument providing the relevant capability. A white background indicates a major contribution to the question, whereas gray shading indicates a lesser but still significant contribution. Shading has a similar meaning as for the flagships: some capability is provided with similar wavelengths and spectral resolutions as suggested by the science panels.

The panel on Compact Objects and Energetic Phenomena (COEP) highlighted the need for multiwavelength observations on a variety of time scales. That panel cites the need for rapid follow-up of events at UV wavelengths, which will be impossible after HST ceases operation. Panchromatic observing capabilities needed in the coming decade were highlighted in WP Megeath.¹⁶ This need touches on a number of science themes identified by the science panels. The success of the great observatories (Spitzer, Hubble, Chandra, and Compton) is a strong motivator for providing panchromatic capabilities to the astronomical community in the coming decade. Some probe-class missions such as TAP seek to provide panchromatic capabilities in a single facility to address specific science themes. Overall, the panel determined that there is strong scientific motivation for facilities, especially space-based, to provide the broadest wavelength coverage possible.

TABLE I.3 Mapping of Probe-Class Missions to Science Panel Questions and Discovery Areas

___________________

¹³ N. Förster Schreiber and S. Wuyts, 2020, “Star Forming Galaxies at Cosmic Noon,” Annual Review of Astronomy and Astrophysics 58:661-725, https://doi.org/10.1146/annurev-astro-032620-021910.

¹⁴ L. Tacconi, R. Genzel, A. Sternberg, 2020, “The Evolution of the Star-Forming Interstellar Medium Across Cosmic Time,” Annual Review of Astronomy and Astrophysics 58:157-203, https://doi.org/10.1146/annurev-astro-082812-141034..

¹⁵ M. Volonteri, A. Reines, H. Atek, D. Stark, and M. Trebitsch, 2017, “High-Redshift Galaxies and Black Holes Detectable with the JWST: A Population Synthesis Model from Infrared to X-Rays,” The Astrophysical Journal 849:2, doi: 10.3847/1538-4357/aa93f1.

¹⁶ S.T. Megeath, L. Armus, M. Bentz, B. Binder, F. Civano, L. Corrales, D. Dragomir, M. Elvis, C. Espaillat, S. Finkelstein, D. Fox, M. Greenhouse, K. Hoadley, J. Kauffmann, A. Kirkpatrick, R. Kraft, G. Khullar, P. Hartigan, C. Lillie, J. Lazio, M. Marengo, S. McCandliss, M. Meyer, R. Mushotzky, A. Pope, P. Roaming, J.D. Smith, K. Stevenson, A. Tielens, G. Tremblay, D. Wang, and S. Wolk, 2019, “The Legacy of the Great Observatories: Panchromatic Coverage as a Strategic Goal for NASA Astrophysics,” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

^a The first letter of the science question identifier indicates the appendix in which the question is discussed.

^b Used in conjunction with the Roman Space Telescope; N/A = no contribution.

^c Used with CETUS or CASTOR (Canadian mission). Instruments: TAP: IRT (Optical Infrared Telescope); Starshade Rendevous: CGI (Coronagraph Instrument); EXO-C: IFS (Integral Field Spectrograph); OOO: 3-Band Phot (3-Band Photometer); CETUS: MOS (Ultraviolet Multiobject Spectrometer), PSS (Point Source Spectrometer); ANUBIS: FUV Spec (Far Ultraviolet Spectrometer); EarthFinder: Spec (Spectrometer); CDIM: Spec (Spectrometer); Nautilus: NAVIIS (Nautilus Visual-Near-Infrared Imager and Spectrograph); ATLAS: Spec (Spectrometer).

I.2 BUDGETARY CONSIDERATIONS

The large mission (flagship) category was the most difficult challenge presented to the panel, because all of the options under consideration proved to be significantly higher cost than even the most optimistic NASA budget can accommodate, particularly if a probe line is to be included in future budgets. The TRACE analyses revealed that the mission costs were likely underestimated by their project teams by nearly 50 percent, largely owing to a combination of TRACE including monetized risks and longer development times that would be driven by budgetary constraints. The largest of these missions (mirror ≥6 m) do an incredible job of searching for Earth-like habitable planets (see detection rates in Table I.4), but require a considerable investment to advance the enabling technologies to a readiness level that will permit them to achieve these results. Given that these technologies are not yet mature (the LUVOIR¹⁷ and HabEx¹⁸ final study reports tabulate Technology Readiness Levels (TRLs) that range from 3 to 5), the current cost

___________________

¹⁷ National Aeronautics and Space Administration, 2019, The LUVOIR Final Report, NASA LUVOIR Mission Concept Study Team, https://asd.gsfc.nasa.gov/luvoir/reports/LUVOIR_FinalReport_2019-08-26.pdf.

¹⁸ National Aeronautics and Space Administration, 2019, Habitable Exoplanet Observatory Final Report, NASA Habitable Exoplanet Observatory Study Team, https://www.jpl.nasa.gov/habex/pdf/HabEx-Final-Report-Public-Release.pdf.

TABLE I.4 The Numbers of Exoplanets of Various Types That Would Be Detectable via Direct Imaging

NOTES: These values assume η_Earth = 0.24 and were taken from the LUVOIR and HabEx study reports, which presented these same values. There is debate about whether hEarth may be lower; the latest analysis performed by the Kepler team (see S. Bryson, M. Kunimoto, R.K. Kopparapu, J.L. Coughlin, W.J. Borucki, D. Koch, V. Silva Aguirre et al., 2020, “The Occurrence of Rocky Habitable-Zone Planets Around Solar-Like Stars from Kepler Data,” Astronomical Journal 161(1):36, https://doi.org/10.3847/1538-3881/abc418) is consistent with this estimate, but see also Gaudi et al. (2021), arXiv:2011.04703v.

and schedule estimates assume that all technologies come to fruition as planned, and thus, the cost and schedule estimates are still very immature. Unfortunately, based on the panel’s and Aerospace’s analyses, the panel believes that large-aperture telescopes (even as small as 4 m) as presented to the panel are not affordable in a reasonable time frame with the currently forecast NASA Astrophysics budgets. They either take too long to develop or consume too much of the budget, making the Astrophysics Program unbalanced. JWST is a case in point—it will have taken more than 20 years to develop and launch. For current flagship design approaches to be affordable and include a probe line would require that the Astrophysics budget at least double to $3 billion per year. However, the panel believes that a flagship mission, designed to observe habitable exoplanets while providing UV, optical, and near-IR imaging and spectroscopy could be pursued once the following criteria are met:

• All enabling technologies have been advanced to a TRL of 6 prior to authorization to proceed with phase A, which will force earlier investment in technology and a change in NASA policy.
• Sufficient budget is planned to be available to permit completion of the mission development and launch within 10 years of Key Decision Point (KDP) B, which might require difficult decisions to be made because funding is not formally in place until KDP C.
• The mission has been designed to have a mission lifetime no less than the time to develop the mission as measured from KDP B. (This criterion is not meant to drive mission reliability but rather to drive getting missions developed more expeditiously.)
• In addition to the three suggestions above, other improvements in the management of Astrophysics flagships can be found in Bitten et al. (2019);¹⁹ WP Tumlinson;²⁰ WP Hylan;²¹ WP Crooke.²²

The panel also observes that NASA could consider pursuing different design approaches in the future for very large aperture telescopes, because the Astrophysics budget may not increase as rapidly as telescope costs. If very large aperture telescopes are required in the future, different design approaches could

___________________

¹⁹ R.E. Bitten, S.A. Shinn, and D. L. Emmons, 2019, “Challenges and Potential Solutions to Develop and Fund NASA Flagship Missions,” 2019 IEEE Aerospace Conference, Big Sky, MT, pp. 1–13, https://doi.org/10.1109/AERO.2019.8741920.

²⁰ J. Tumlinson, J. Arenberg, M. Mountain, L. Feinberg, J. Grunsfeld, K. Sembach, N. Levenson, J. O’Meara, and M. Postman, 2019, “The Next Great Observatories: How Can We Get There?,” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

²¹ J. Hylan, J.A. Crooke, and M. Bolcar, 2019, “Managing Flagship Missions to Reduce Cost and Schedule,” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

²² J. Crooke, M. Bolcar, and J. Hylan, 2019, “Funding Strategy Impacts and Alternative Funding Approaches for NASA’s Future Flagship Mission Developments,” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

be considered, including assembly in space, and servicing and modularity that would allow telescopes to evolve, including adding aperture, upgrading capabilities, and extending the life of the telescope. Existing capabilities for assembly of high-performance optical systems in space are nonexistent and will require substantial development if this type of construction is to be realized.

I.3 FLAGSHIPS

Flagships provide capabilities that cannot be achieved at smaller scales, with sensitivity and angular resolution typically being the drivers for large telescopes. Flagships include a range of instrumentation, whereas the requirements are set by the most scientifically compelling goals. The instrumentation enables a broad range of other science goals to be addressed, so large missions have broad support within the astronomical community. Flagships provide large numbers of astronomers with their first exposure to space data and space projects.

I.3.1 Flagship Science Capabilities

The EOS-1 panel was presented with two projects that would lead to missions capable of detecting biosignatures on Earth-like planets, a compelling goal on many levels. Figure I.1, which appears in both the LUVOIR and HabEx final reports, shows the richness of an exoearth spectrum in the 0.2 to 2 μm region, including the very strong ozone potential biosignature at 0.25 μm. This spectrum assumes that one is observing reflected light, which is possible with direct imaging using a high-performance starlight suppression system. Small-scale height features such as these absorptions would not be observable in a transmission spectrum from a transit observation for an exoearth around a Sun-like star. Earth-like planets around Sun-like stars transit very rarely (from geometry), infrequently (once per year), and at such a shallow transit depth that their atmospheric features are essentially not characterizable by plausible missions.²³

___________________

²³ A. Misra, V. Mesdows, and D. Crisp, 2014, “The Effects of Refraction on Transit Transmission Spectroscopy: Application to Earth-Like Exoplanets,” Astrophysical Journal 792:61, https://iopscience.iop.org/article/10.1088/0004-637X/792/1/61.

The LUVOIR team developed two concepts for consideration: LUVOIR-A with a primary mirror diameter of 15 m, and LUVOIR-B with an off-axis primary mirror with a diameter of 8 m. Both concepts rely on coronagraphs for starlight suppression. The LUVOIR team’s preferred configuration is the 15 m version. The HabEx team developed nine separate concepts ranging in size from 2.4 m to 4 m and with starlight suppression using either a coronagraph or a starshade. The HabEx team’s preferred configuration (4H) uses a 4 m mirror and both coronagraphy and a starshade. The panel examined LUVOIR-B, HabEx 4H, and HabEx 3.2S in detail. The choice of LUVOIR-B over LUVOIR-A was based largely on cost considerations because LUVOIR-A would take too long to build and test without an implausibly large increase in the NASA Astrophysics budget as judged by the team’s values and confirmed by TRACE. The panel choose two HabEx configurations, as they provided a comparison between a mission with two starlight suppression techniques (4H) and one with only a starshade 3.2-m telescope with six pie-slice segments(3.2S), and also provided a comparison between a monolithic primary (4H) and a segmented primary (3.2S).

Both LUVOIR and HabEx teams have selected a suite of focal plane instruments that are aimed at both detailed characterization of exoplanets and more general observations. Table I.5 lists the capabilities of both LUVOIR and HabEx instruments. Central to discussing which of these missions is needed to achieve the science goals outlined by the science panels is the size of the telescope’s primary mirror. The size of the primary mirror sets the angular resolution and the sensitivity achievable. The angular resolution scales with the diameter D, and the sensitivity scales as D⁴ for background- and diffraction-limited imaging. For

TABLE I.5 Focal Plane Instrumentation for LUVOIR-A (Upper Table) and HabEx 4H (Below)

NOTES: The baseline design for LUVOIR-B does not include POLLUX. HabEx 3.2S has no coronagraph and its HWC camera has a short wavelength cut-off of 0.32 µm. These tables are from the mission final reports.

the goal of measuring exoearth biosignatures, a mirror sufficiently large to see the exoearth outside the coronagraph’s inner working angle is required and needs to be large enough to measure the reflected light spectrum of the exoearth. An exoearth in the habitable zone of a G star at 10 parsecs has a magnitude of ~30 in the absolute (AB) system, which alone suggests that a mirror with a collecting area of at least 6 m is needed, and which is confirmed by the small number of candidates for small telescopes shown in Figure I.2. Table I.4 lists the estimated number of exoplanets detectable using any of the LUVOIR and HabEx configurations considered here (assuming η_Earth = 0.24).

Figure I.2 presents the expected number of exoearth candidates assuming a value of η_Earth = 0.24. During the span of the panel’s work, it became clear that the value of η_Earth could be as much as a factor of 2.5× smaller, with a corresponding drop in the number of exoearth candidates (but see Bryson et al., 2020,²⁴ which supports a value of 0.24, while Gaudi et al., 2021,²⁵ cite a range of 0.05 to 0.5). The panel assumed that an extensive flagship mission would be capable of detecting an exoearth with essentially complete certainty. The number of detectable planets depends on D^1.97 and on η_Earth^0.96 using the formulation in Stark et al. (2019).²⁶ If η_Earth proves to be as low as 0.05, then a mirror of inscribed size ≥ 6 m is required to ensure detecting at least one exoearth. Figure I.3 shows how such a change in η_Earth will change the exoearth yield as a function of telescope size and coronagraph type. Figure I.3 also illustrates that coronagraph performance has not reached physical limits, and that recent progress in coronagraph performance is very encouraging.

___________________

²⁴ S. Bryson, M. Kunimoto, R.K. Kopparapu, J.L. Coughlin, W.J. Borucki, D. Koch, V. Silva Aguirre et al., 2020, “The Occurrence of Rocky Habitable-Zone Planets Around Solar-Like Stars from Kepler Data,” Astronomical Journal 161:36, https://doi.org/10.3847/1538-3881/abc418.

²⁵ B.S. Gaudi, J.L. Christiansen, and M.R. Meyer, 2021, “The Demographic of Exoplanets,” version 3, https://arxiv.org/abs/2011.04703.

²⁶ C.C. Stark, R. Belikov, M.R. Bolcar, E. Cady, B.P. Crill, S. Ertel, T. Groff, S. Hildebrandt, J. Krist, P.D. Lisman, J. Mazoyer, B. Mennesson, B. Nemati, L. Pueyo, B.J. Rauscher, A.J. Riggs, G. Ruane, S.B. Shaklan, D. Sirbu, R. Soummer, K. St. Laurent, and N. Zimmerman, 2019, “ExoEarth Yield Landscape for Future Direct Imaging Space Telescopes,” Journal of Astronomical Telescopes, Instruments, and Systems 5(2), https://doi.org/0.1117/1.JATIS.5.2.024009.

Other science cases also point to needing a telescope of at least 6 m. Figure I.4, taken from the LUVOIR team presentation to the panel, shows the number of quasi-stellar objects (QSOs) detectable in the UV that could be used as background sources for tracing circumgalactic gas. Too few QSOs are accessible to HST for this technique to go beyond just showing the existence of such gas. Larger telescopes can detect more QSOs with adequate signal to noise to enable mapping of the circumgalactic medium, with Figure I.4 implying that D >6 m will make a significant impact in this area. This impact comes not just in the form of more accessible QSOs but more importantly in the ability to study absorption lines in the spectra of QSOs at z ~0.5–1, which are essentially inaccessible to smaller telescopes. Using higher-redshift QSOs enables study of the CGM around galaxies over nearly half the age of the universe. Many other exoplanet questions such as that posed by EASS-2, “What is the nature of individual planets, and which processes lead to their diversity?” also need D >6 m (LUVOIR report, p. 1-24).²⁷ Of the 12 “signature science cases” listed in Table 1.2, only 2 are cited as requiring a 15 m telescope. Four are cited as needing D ≥ 6.7 m, with the rest requiring 8 m. Some of the 8 m projects can be done using a slightly smaller telescope at the expense of needing more observing time. Based on these considerations and the expected yield of exoearths, the panel has set the minimum mirror size at 6 m. The panel notes that the time observing time difference between a telescope with a 6-m collecting area and 6.7-m collecting area is ~55 percent, which was judged an acceptable difference for these projects.

I.3.2 Flagship Costs

The two versions of LUVOIR, LUVOIR-A and LUVOIR-B, are estimated by their project teams to cost $16.0 billion and $12.2 billion in fiscal year (FY) 2020 dollars at a 70 percent confidence level. Because TRACE funding was limited, the panel asked the Aerospace Corporation to perform a TRACE analysis on only LUVOIR-B because LUVOIR-A seemed infeasible to complete during the next two decades without an unprecedented increase in the Astrophysics budget as judged from the team’s report. Table I.6 summarizes the team and TRACE cost estimates for the three missions considered in detail by the panel. These cost estimates assume an optimal funding profile, and if development is funding-limited and stretched over a longer period than the 10–12 years assumed for these missions, the total costs will be significantly higher.

TABLE I.6 Comparison of Team and TRACE Cost Estimates ($FY 2020, 70% Confidence)

___________________

²⁷ National Aeronautics and Space Administration, 2019, The LUVOIR Final Report, NASA LUVOIR Mission Concept Study Team, https://asd.gsfc.nasa.gov/luvoir/reports/LUVOIR_FinalReport_2019-08-26.pdf.

I.3.3 Technology Development Needs

All of these missions and the 6 m version as suggested by the panel will need significant technology development to reach the 10^–10 contrast needed for direction detection of exoearths. These needs are highlighted by the Aerospace risk ratings of medium-high for LUVOIR-B, medium for HabEx 4H, and medium-low for HabEx 3.2S. For LUVOIR-B, Aerospace lists these areas for significant development:

• Large, segmented mirror provides high-precision figure and stability;
• Coronagraph incorporates new technology for contrast improvements over the Roman Space Telescope CGI;
• UV instrumentation requires fabrication improvements to optical coatings and detectors; and
• Large × 48 m sunshade requires consideration of stowage and deployment.

The equivalent list of HabEx 4H includes:

• Starshade to be scaled-up to 52 m diameter with integration of optical cover and solar array in deployable center disc;
• Coronagraph incorporates new technology for contrast improvements over the Roman Space Telescope CGI;
• Telescope mirror required with high-uniformity CTE and coating over 4 m monolithic, lightweighted primary mirror; and
• Life testing required for colloidal thruster to meet 10-year objective.

The list for HabEx 3.2S includes only the starshade and thruster risks from the HabEx 4H list. In addition to the items listed above that were noted by Aerospace, the panel found a number of other technology development items that are discussed in Sections I.3.4 and I.3.5.

The technology progress described earlier provides an excellent starting point for these missions. The comparison of HabEx 4H with HabEx 3.2S in the Aerospace risk assessment reveals that a starshade with its highly desirable properties of very small inner working angle and overall larger area of high contrast is not a high-risk item at the size scale needed for the HabEx telescopes but is rated as a medium risk with further development needed. A starshade for a larger telescope would need to be larger and placed farther away from the telescope, which would increase the risk. Although the HabEx 4H observational plan includes a clever scheme to work around the limitation of needing refueling for a starshade to be positioned for many targets, the recent successes of refueling other satellites (Intelsat 901 was refueled by Northrop Grumman’s Mission Extension Vehicle-1 [MEV-1]) could be applied to starshades, making them even more attractive. A space demonstration of a starshade, even at smaller scale than needed for an exoearth mission, would be valuable in retiring operational risks and as further proof beyond the small-scale ground tests of the efficacy of starshades.

Telescopes employing coronagraphic starlight suppression to achieve 10^–10 contrast require a significant technology investment, probably as large as $600 million ($FY 2020) by the start of phase A. Approximately $130 million of this funding would be needed for the high-contrast coronagraph instrument as estimated in the LUVOIR final report. Given the outstanding science discoveries that these exoplanet missions can accomplish but only with critical technologies that need to reach TRL 6 or above in the next 5 years, the panel suggests that NASA fund the technology tasks outlined in the LUVOIR and HabEx reports and do so over the next 5-year period, which would then flow into detailed mission architecture studies that could be completed before the next decadal survey. A key decision is the size of the primary mirror, and whether

a starshade will be included in the mission. For a coronagraph-based mission, Figure 11-3 in the LUVOIR final report lays out a phased technology program that addresses the highest risk technology developments needed. This early funding of significant technology is not normal for a NASA flagship mission. However, given the significant cost uncertainties related to technology, a significantly more accurate overall flagship cost would result from this early technology development roadmap, and would prevent getting well into mission design and development only to discover a significant issue that would incur a high cost to work around.

The technology roadmaps outlined by both LUVOIR and HabEx are critical to the ultimate determination of whether these exoplanet missions can accomplish their scientific goals. While the two mission concepts have some significant technology needs that are quite different (e.g., large sunshade for LUVOIR, starshade for HabEx), they do have a number of technology areas in common. The panel suggests that all 2020 decadal study high-risk technologies for these missions be reviewed for overlap such that a single technology roadmap can be described, costed, and scheduled with Figure 11-3 from the LUVOIR report illustrating at a top level what such a plan might look like. Such a plan might include key decision points with oversight by a single program officer. A preliminary list of cross-program technologies that could be considered for this grand technology roadmap include ultra-stable structural composites, low-creep adhesives, charge transfer efficiency (CTE) measuring techniques, milli-K thermal sensing, finite element model/test surface figure error/wavefront error model correlation. In addition, the panel suggests that development of starlight suppression technologies be continued, both starshades-based and coronagraph-based ones. There is significant overlap in the advancement of coronagraph-based technologies between LUVOIR and HabEx, and high-priority technologies to mature include better coronagraph architectures to increase science yield, adaptive wavefront control algorithms that are more efficient and improve tolerance to instabilities, vacuum-compatible deformable mirrors with more actuators, post-processing algorithms, and vacuum testing of the entire starlight suppression system. High-priority starshade technologies include demonstrating starshade petal accuracy and stability, performance modeling and validation, as well as demonstrating acceptably low scattered sunlight from petal edges. Ideally, this grand technology roadmap would be funded such that it could be fully accomplished within 5 years. Without studying each of the building blocks prior to the system-level testing, additional program risk remains.

To keep this technology roadmap focused, the panel suggests setting a strong, concise goal for a flagship: detect a suite of biosignatures on an exoearth, or show that they are rare in a statistically meaningful sense. The mission could start development before the end of the 2020s if the technologies and funding have met the panel’s proposed criteria for technologies and budget (see Section I.2, “Budgetary Considerations”). Because of the combination of the preceding challenges and opportunities, a commitment to the above goal rather than choosing a specific mission implementation is most appropriate now. A mid-decadal committee would review technology and scientific progress with NASA leading an effort to optimize the best mission point design in advance of that review.

I.3.4 Detailed Technology Development Comments—Telescope Considerations

The panel is neither suggesting a preferred primary mirror configuration nor suggesting that monolithic primaries larger than 4 m are not feasible, but rather is indicating that further study is needed to make an informed choice. For LUVOIR-style telescopes, three technology areas enable the science objectives: the high-contrast coronagraph, the ultra-stable segmented telescope that enables picometer-level wavefront and contrast stability, and the UV instrument technology. The development of these technologies to a level of TRL 6 prior to start of Phase A will enable the overall cost and schedule risk to be minimized. The LUVOIR

mission study report has outlined the technology, engineering, and manufacturing needs and in general shows how the TRL in these areas could be improved to levels needed to start mission development. Table 11-5 in the LUVOIR report summarizes the ultra-stable enabling technology development activities needed. Risks that do not seem to have been fully considered in the LUVOIR documentation related to these technology activities include the following (section numbers taken from the LUVOIR report):

• 11.2.2.1 System-Level Model Development and Validation—The LUVOIR-B telescope has three times the number of mirror segments as JWST. The mechanical and thermal finite element models created for JWST required significant amounts of computer power and very long run times. With the model ~3 times larger, a concern arises about the ability to solve mechanical and thermal problems (with <milli-K resolution) and perhaps hundreds of millions of degrees of freedom.
• 11.2.2.2 Thermal Sensing and Control Development—The mission study report has shown that a very small solid mirror can be controlled in a small chamber to <1 milli-K over long periods of time. The engineering study would benefit from including the most complicated areas where thermal sensing and control may not be straightforward, such as the internal and external perimeters of the primary mirror. This may be best addressed during the development of the segmented telescope system (see below) but also can be addressed with very detailed thermal and mechanical modeling.
• 11.2.2.3 Composite Material Process Development and Optimization—Ultra-stable telescopes need support structures that show very limited time-dependent delta L owing to temperature, moisture, and material creep effects. This latter term does not seem to be included in the study plan and may be as important as the other contributing factors. Other materials, such as invar, also have long-term material instability and are currently being assessed on the Roman Space Telescope for their impact on short-term stability. The LUVOIR design includes materials and components that are not as stable as the optical materials themselves; physically testing these materials and components may be needed.
• 11.2.2.4 Mirror Substrate—The current plan does not include the creation of a 3D coefficient of thermal expansion (CTE) profile of each and every mirror (or in fact any single mirror). This may be necessary to get the predicted wavefront error (WFE) when very small temperature deltas are assumed. The substrate manufacturer currently does not thoroughly perform these tests.
• 11.2.2.12 Segmented Telescope System—This will be the most critical TRL element because it will include many of the completed activities in Table 11-5 in the LUVOIR final report. Even though this activity is a subscale of the telescope system, the panel suggests that it be designed to include any/all areas where the greatest uncertainties may exist. All of the TRL items that will feed into this test need to be completed early enough and with as much schedule margin as possible to allow the Telescope System test to be thoroughly debugged and tested.

The HabEx mission study report identified two primary technology study items that will allow the telescope risk to be reduced significantly; demonstrating large mirror fabrication and obtaining the necessary mirror coating uniformity. The panel has identified other significant risk-related shortcomings that include lack of a full 3D map of CTE data, which would allow test and model comparisons to be made and understood; the effectiveness of their 1 g offloader system to fabricate their 0 g mirror figure; and the effects of long-term material stability not correctable by their laser metrology system (such as mirror figure time-dependent errors). There is no test plan to reduce the risk related to the effects of the lack of a 3D CTE map. The team contends that if the wavefront error requirement in a nominal thermal environment is met, then there is no significant risk. The problem is that the on-orbit environment(s) cannot easily or feasibly be duplicated on the ground (as determined with JWST) and therefore no model correlation or WFE validation will have been proven. This is a correctable flaw in the HabEx technical path via independent testing of

CTE and performing model/test correlation. Another risk is related to the 1 g manufacturing and the need for actuators to achieve the correct mirror figure at 0 g. Figure 6.8-3 from the HabEx final report shows that errors would not be reduced sufficiently without actuators. Typical finite element model analyses are not likely to bring the errors below a factor of 2× above the stated HabEx requirement, which suggests that this is another area needing more attention. Mirror coating uniformity is adequately addressed in the HabEx report.

I.3.5 Detailed Technology Development Comments—Starlight Suppression Considerations

In high-contrast imaging, it is useful to distinguish between the planet:star flux ratio and the instrumental contrast. The flux ratio is a property of the astrophysical system—for example, Earth has approximately 10^–10 brightness of the Sun. The instrumental contrast is (roughly) the ratio of the surface brightness of the residual halo of scattered light surrounding a star to the peak surface brightness of an unocculted star. Using post-processing to fit and remove residual starlight, it is often possible to detect and characterize a planet whose flux ratio is lower than the instrumental contrast, just as ground-based infrared instruments can detect objects much fainter than the bright sky. The exact post-processing benefit will depend on the stability of the instrument and telescope, driving toward ultra-stable observatory architectures. Both the LUVOIR and HabEx study reports included results from very good and detailed integrated modeling of the telescope and starlight suppression systems: dynamical Structural, Thermal, and Optical Performance (STOP) models, end-to-end diffraction and wavefront control loops, and so on. These models have a level of fidelity that goes well beyond prior studies of coronagraphic or starshade instruments (except for the Roman Space Telescope). The mission studies leverage the experience and some of the machinery of Roman models, as well as Roman hardware demonstrations. Both of these increase confidence in the coronagraphic and starshade starlight suppression technologies.

However, the TRL of some of the subsystems is still as low as 3. The reports describe coordinated plans to advance the technologies to TRL 6 prior to phase A, with a schedule and cost commensurate with past coronagraph technology development projects, including the Roman Space Telescope. Some of the key remaining risks are as follows:

• The gap between laboratory demonstrations of coronagraphs and LUVOIR/HabEx mission requirements is rapidly shrinking, but nonnegligible (see Table I.7). Contrast levels approaching 1 × 10^–10 in broadband have been demonstrated, but with a simpler coronagraph than baselined for LUVOIR and HabEx, and for monolithic apertures. Currently, the deepest demonstrations for nonmonolithic apertures are order (1 × 10^–9) for Roman coronagraphs (Hybrid Lyot and Shaped Pupil). For more aggressive coronagraphs, such as the Apodized Vortex that was chosen for LUVOIR-B, the demonstrations are of order (1 × 10^–8). Validated models predict that required coronagraph performance is achievable, which increases confidence that this gap will be closed soon. However, just as with any technology development project, there is always a small possibility of some unexpected limiting factor that escaped attention, because working at 1 × 10^–10 contrast levels is still somewhat unexplored territory. (Possible examples are vector effects from segmentation edges, amplitude errors from reflectivity nonuniformity, sub-wavelength physics [if any], effects of dust or other contaminants, deformable mirror mechanical and electrical stability and reliability, etc.) Both LUVOIR and HabEx mission study reports present a good technology development plan to close this performance gap, and the panel emphasizes that a well-planned early technology development effort is absolutely critical for these missions.

TABLE I.7 Selected Laboratory Demonstrations of Coronagraphs

^a J. Llop-Sayson, G. Ruane, N. Jovanovic, D. Mawet, D. Echeverri, A.J. Edlorado Riggs, C.T. Coker, G. Morrissey, and H. Sun, 2019, “The High-Contrast Spectroscopy Testbed for Segmented Telescopes (HCST): New Wavefront Control Demonstrations,” Proceedings of SPIE: Techniques and Instrumentation for Detection of Exoplanets IX (S.B. Shaklan, ed.), vol. 11117, International Society of Optical Engineering (SPIE), Bellingham, WA.

^b E. Cady, K. Balasubramanian, J. Gersh-Range, J. Kasdin, B. Kern, R. Lam, C. Mejia Prada et al., 2017, “Shaped Pupil Coronagraphy for WFIRST: High-Contrast Broadband Testbed Demonstration,” Proceedings of SPIE: Techniques and Instrumentation for Detection of Exoplanets VIII (S. Shaklan, ed.), vol. 10400, International Society of Optical Engineering (SPIE), Bellingham, WA.

^c J. Llop-Sayson, G. Ruane, D. Mawet, N. Jovanovic, C.T. Coker, J. Delorme, D. Echeverri, J. Fucik, A.J. Edorado Riggs, and J.K. Wallace, 2020, “High-Constrast Demonstration of an Apodized Vortex Coronagraph,” Astronomical Journal 159(3):79–87.

• The amount of time and/or number of iterations required to achieve 10^–10 contrast is based primarily on experience with Roman Space Telescope CGI modeling and demonstrations, which is at ~10^–8 and 10^–9 levels. It is possible that convergence to 10^–10 will be slower than expected.

Fortunately, coronagraphic and wavefront control technologies are still improving, including several promising directions that mitigate the above risks:

• Advances in wavefront control algorithms (e.g., WP Pueyo,²⁸ WP Kasdin²⁹) may relax the stability requirements of the telescope, potentially to JWST levels.
• Advances in post-processing algorithms may relax raw contrast requirements.
• Advances in coronagraph designs may relax telescope stability requirements, as well as enable reducing the size of the telescope and enable cost savings.
• The panel emphasizes that the technology is sufficiently mature to begin more detailed studies that would eventually lead to a LUVOIR- or HabEx-like mission now, and the above opportunities for improvement offer some combination of risk mitigation, increased science, and/or cost savings (e.g., by reducing telescope aperture without sacrificing exoplanet science). “Sufficiently mature” means that the interplay between mission architecture and starlight suppression techniques can be examined in greater depth effectively.

Observation: For either LUVOIR-B, a 6 m HabEx 4H, or HabEx 3.2S to proceed to development, substantial investments in technology development will be required.

___________________

²⁸ L. Pueyo, C. Stark, V. Bailey, M. Bolcar, L. Coyle, L. Feinberg, T. Groff, O. Guyon, J. Jewell, J. Kasdin, S. Knight, D. Mawet, J. Mazoyer, B. Mennesson, M. Perrin, D. Redding, A.J. Riggs, G. Ruane, R. Soummer, W. Scott, and N. Zimmerman, 2019, “Wavefront Sensing and Control Technologies for Exo-Earth Imaging,” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

²⁹ J. Kasdin, L. Pogorelyuk, N. Zimmerman, A.J.E. Riggs, and L. Pueyo, 2019, “Relaxing Stability Requirements on Future Exoplanet Coronagraphic Imaging Missions,” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

I.4 PROBES

At approximately $1 billion for development, a probe mission line would fill the gap in cost between Explorers that are cost-capped at $150 million to $250 million and flagships (typified by the >$3 billion Roman Space Telescope and ~$10 billion JWST). This distribution of mission sizes leaves a gap of more than an order of magnitude in cost and scale between the large missions and the next category. A probe line would provide an opportunity similar in scale to the smaller Great Observatories such as Spitzer and Compton. A probe line could help ensure the availability of the pan-chromatic coverage needed because many problems in astrophysics benefit from multi-wavelength studies (e.g., WP Megeath³⁰) and a diversity of techniques such as ultra-high-resolution spectroscopy, polarimetry, and high time-resolution observations. WP Elvis also presents arguments supporting the need for probes.³¹ The panel heard presentations from all probes that included any UVOIR wavelengths as listed in Table I.3. The mapping of probes to science panel questions in Table I.3 shows that all of the probes that were considered by the panel can make a major or significant contribution to one or more of the science questions. Past missions that would have fallen into this cost category such as Spitzer and Kepler have made impressive contributions in a range of astrophysical areas. Probes also provide a mechanism for responding to new science opportunities on a more rapid time scale than is possible with larger missions.

The panel looked at the Planetary Science Division’s New Frontiers program, similar in cost scope to what is envisioned for a probe line, as a possible model. The panel envisions developing a probe line as soon as is practical. Astrophysics will need to modify the program somewhat, but the EOS-1 panel suggests that a probe line would have the following characteristics. The line could support two announcements of opportunity (AOs) and two launches per decade. This cadence would be high enough to support principal investigators (PIs) proposing but not being selected and coming back for a second try within a 10-year span. Examination of the probes presented to the panel suggests that a cost cap in the $1 billion to $1.5 billion range not including launch vehicle and science operations would be appropriate with the understanding that the total mission life cycle cost would be larger because of launch costs. Community participation in the science operations phase such as through a guest observer program is strongly suggested. NASA could consider incentivizing the use of new technologies needing demonstration in space by not counting the cost of such technologies against the total mission cost. The technologies to be allowed in this manner would be specified in the AO.

Given the quality of the probe missions presented to the panel, the panel found it tempting to specify the first Probe candidates. However, the panel considered the value of open competition for fostering new ideas and creativity, and suggests that the probe line AOs not specify topical areas beyond the possible technology use mentioned above.

___________________

³⁰ S.T. Megeath, L. Armus, M. Bentz, B. Binder, F. Civano, L. Corrales, D. Dragomir, M. Elvis, C. Espaillat, S. Finkelstein, D. Fox, M. Greenhouse, K. Hoadley, J. Kauffmann, A. Kirkpatrick, R. Kraft, G. Khullar, P. Hartigan, C. Lillie, J. Lazio, M. Marengo, S. McCandliss, M. Meyer, R. Mushotzky, A. Pope, P. Roaming, J.D. Smith, K. Stevenson, A. Tielens, G. Tremblay, D. Wang, and S. Wolk, 2019, “The Legacy of the Great Observatories: Panchromatic Coverage as a Strategic Goal for NASA Astrophysics,” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

³¹ M. Elvis, J. Arenberg, D. Ballantyne, M. Bautz, C. Beichman, J. Booth, J. Buckley, J.O. Burns, J. Camp, A. Conti, A. Cooray, W. Danchi, J. Delabrouille, G. De Zotti, R. Flauger, J. Glenn, J. Grindlay, S. Hanany, D. Hartmann, G. Helou, D. Herranz, J. Hubmayr, B.R. Johnson, W. Jones, N.J. Kasdin, C. Kouvelioutou, K.E. Kunze, C. Lawrence, J. Lazio, S. Lipscy, C. F. Lillie, T. Maccarone, K.C. Madsen, J.E. McEnery, R. Mcentaffer, R. Mushotzky, A. Olinto, P. Plavchan, L. Pogosian, A. Ptak, P. Ray, G.M. Rocha, P. Scowen, S. Seager, J. Tomsick, G. Tucker, M. Ulmer, Y. Wang, and E. Wollack, 2019, “The Case for Probe-Class NASA Astrophysics Missions,” white paper submitted to Astro2020: Decadal Survey on Astronomy and Astrophysics.

I.5 OVERALL PROGRAM BALANCE

The sections above lay out a strong case not only for exoplanet science but also for other astrophysics science goals and objectives. These were derived from the science panel inputs, and the panel has used these to help assess the various probe and flagship missions. In doing so, the panel considered program balance, which focuses on the need to ensure that opportunities are available for the whole astrophysics community to conduct science investigations, both on the ground and in space, and across all disciplines. With respect to the EOS-1 charge for electromagnetic observations from space, the current astrophysics program of record includes a plan for four Explorer missions and four missions of opportunity per decade and extremely large flagship missions that have been taking more than 20 years to complete. The flagships consume nearly half of the astrophysics budget for decades until they launch and start producing science results.

I.5.1 Explorers Program

Excellent science is best enabled by ensuring programmatic balance across small, medium, and flagship-class missions, with robust support for research, analysis, and theoretical underpinning of the science data returned by all of these missions. Flagship missions have the potential to enable unique scientific opportunities and can tackle “big questions” in science that are otherwise unachievable. However, the extremely slow cadence between flagships (>20 years in some cases) means that the time between new starts is a significant fraction of a career. This slow cadence means that a flagship’s scientific requirements are developed long before it can be deployed, and its technology is frozen into place well before launch, resulting in a potentially very long cadence between infusions of new technology. Thus, flagships can be slow to respond to new and emerging science questions and new technology. By ensuring programmatic balance across small, medium, and large missions, along with adequate support for scientific analysis of data from NASA missions, NASA can answer big, ambitious scientific challenges while still remaining nimble enough to take advantage of emerging scientific breakthroughs and new technology. The panel observes that for the reasons outlined below, the astronomical community is well served with the programmatic balance as currently practiced by NASA.

Smaller competed missions offer opportunities for a wide array of scientists to lead missions and have the potential to broaden and diversify the pool of scientific leaders (provided meaningful efforts are made to ensure full participation of the entire community). They provide more rapid results compared to flagships, helping to ensure that science is accomplished in a timely fashion and injecting frequent new experimental data to provide necessary feedback to theoretical work (which needs to be well-supported). This class of missions offers an opportunity for the science community to take a direct hand in setting priorities and leading the effort to answer key and emerging scientific questions.

Smaller missions such as Explorers and a new line of probes provide an opportunity to demonstrate new technologies that can be used to break open new fields, provide multi-wavelength access, respond to new and emerging scientific discoveries, and reduce risk for more costly flagship missions. The instrumentation, observing strategies, and analysis techniques employed on smaller missions can serve as valuable test beds and risk reduction measures for larger, costly flagships. For example, the Kepler mission, which originated as a Discovery mission, greatly expanded understanding of the number and diversity of planetary systems outside of our own solar system using the transit technique outside the confounding effects of Earth’s atmosphere. The Spitzer Space Telescope, equivalent in today’s dollars to a probe-class mission, was creatively repurposed beyond its original scientific objectives to probe the thermal structure, chemistry, and atmospheric dynamics of extrasolar planets. Additionally, both the Spitzer Warm Mission and the Near-Earth Object Wide-field Survey Explorer (NEOWISE) Reactivation served to validate passive cooling techniques that will be employed by JWST.

Small missions also play a vital role in ensuring technological vitality. Many areas of astrophysics depend critically on the availability of specialized technologies that have limited commercial or military applications, meaning that they are kept viable only through the continued investment of NASA’s scientific programs. The more rapid cadence of smaller, competed missions compared to flagship missions allows these specialty technology areas to be maintained. Examples include far-IR detectors, X-ray detectors and optics, polarimetry, and UV optics and detectors. Small missions are essential for keeping specialty astrophysical technologies alive between flagships that can be separated by decades.

Despite the important role they play in providing risk reduction and ensuring availability of key technologies, small missions currently face a different standard of risk than flagships. The current competed line of Explorer missions, along with Discovery and New Frontiers missions in the Planetary Science Division, are strictly cost-capped and limited to employing only technologies that are at or above TRL 6 or above by the time of their Preliminary Design Reviews (PDRs). The Discovery and New Frontiers lines in the Planetary Science Division have recently somewhat ameliorated the concern that these lower-cost missions are unreasonably restricted from using less mature technology by providing opportunities for ride-along technology demonstrations. For example, the Deep Space Optical Communications (DSOC) package was offered with an incentive to proposing teams in the 2015 Discovery round. Offering similar incentivized technology demonstration opportunities for the Explorer and probe lines in Astrophysics may offer some chance to mature technology and expand wavelength coverage and diversity of available techniques.

The emerging areas of SmallSats and CubeSats are gaining the attention of astronomers. SmallSats are being proposed to fill several key gaps in astrophysical research—namely, the monitoring of sources for weeks or months at time, and at wavelengths not accessible from the ground such as X-ray, UV, far-IR, and low-frequency radio. Time-domain astronomy has been an enormous challenge for any flagship mission, because those telescopes are highly sought after and shared among hundreds of programs annually, making long-duration observing nearly impossible. Other science cases for SmallSats being developed now include a wide variety of astrophysical experiments, including exoplanets, stars, black holes and radio transients, galaxies, and multi-messenger astronomy. Achieving high-impact research with SmallSats is becoming increasingly feasible with advances in technologies such as precision pointing, compact sensitive detectors, and the miniaturization of propulsion systems.³²

I.5.2 Foundational Programs

The panel observes that several areas outside the usual mission development flow could have profound and far-reaching consequences if pursued. Servicing such as refueling could be beneficial, because one of the limitations in the use of starshades is the need for fuel to reposition the starshade. HabEx 4H devised a hybrid coronagraphy scheme to use the starshade only for the most important targets. Refueling the starshade might be preferable.

Assembly of structures in space might alleviate some of the issues with launching large telescopes, but current capabilities are far from what is needed for precision structures. Investment in developing this area could enable ambitious future missions that would be considered by future decadal surveys. More innovation may be needed to enable very large future telescopes in space. Funding some “blue sky” thought teams and including some engineering support is needed to reduce the cost of future large space telescopes. Management and funding are other areas that could benefit from innovative approaches.

___________________

³² E. Shkolnik, 2018, “On the Verge of an Astronomy CubeSat Revolution,” Nature Astronomy 2:374–378, https://doi.org/10.1038/s41550-018-0438-8.