2

A New Cosmic Perspective

The past decade has been one of extraordinary discoveries in astronomy and astrophysics, and a realization of the scientific vision of the 2010 decadal survey, New Worlds, New Horizons in Astronomy and Astrophysics (Astro2010).¹ Scientific advances range from the detection of gravitational waves from merging black holes, to a direct image of a black hole in a nearby galaxy, to the production of heavy elements in the merger of two neutron stars, long-hypothesized but observed in exquisite detail for the first time. Increasingly sensitive instrumentation and powerful simulations are uncovering connections between the complex gaseous surroundings of galaxies and the forces that shape them. An explosion in the number of known exoplanetary systems has been accompanied by the detailed characterization of a subset of these other worlds, with insights into their formation arising from imaging of the disks where young planets are assembling. Newly discovered fossil structures from the formation of the Milky Way Galaxy open a window on the Milky Way’s distant past, and observations take several steps closer to diagnosing the conditions present shortly after the Big Bang.

The investments of previous decades bore fruit in this decade in the awarding of Nobel Prizes in Physics for six discoveries derived from astronomical measurements: dark energy, neutrino oscillations, gravitational waves, exoplanets, physical cosmology, and black holes (Figure 2.1). At the same time, international collaborations greatly expanded. A salient example of this is the seminal paper detailing the discovery of two merging neutron stars and their gravitational and electromagnetic signatures: it encompassed nearly 4,000 authors from 900 institutions and 70 observatories, spanning all seven continents and space-based facilities, or roughly one-third of the professional astronomical community as well as most of the gravitational wave community worldwide.

The Astro2010 decadal survey identified three main science objectives for its decade, while also enabling a wider discovery potential. The resulting scientific program is still bearing fruit in this decade and will in the next. On the topic of Cosmic Dawn: Searching for the First Stars, Galaxies, Black Holes, the James Webb Space Telescope (JWST) (launched in December 2021) will directly examine the youngest observable galax-

___________________

¹ National Research Council, 2010, New Worlds, New Horizons in Astronomy and Astrophysics, The National Academies Press, Washington, DC.

ies, the Vera Rubin Observatory (science first light expected in 2023) and the Nancy Grace Roman Observatory (formerly Wide-Field Infrared Survey Telescope, with launch expected in 2025) will transform views of dwarf galaxies at the extremes of galaxy formation and the record of ancient stars they left behind, and the European Space Agency’s (ESA’s) Laser Interferometer Space Antenna (LISA) (launch expected in mid-2030s) should identify mergers of black holes all the way back to their earliest formation. In the arena of New Worlds: Seeking Nearby Habitable Planets, the Transiting Exoplanet Survey Satellite (TESS) (launched 2018), JWST, and soon Roman will explore a widening array of exoplanet types. For the Physics of the Universe: Understanding Scientific Principles objective, myriad ground- and space-based telescopes will use the universe as a laboratory to probe the nature of dark matter and dark energy. Roman, Rubin Observatory, ESA’s Euclid (launch expected in the latter half of 2022), the most recent Laser Interferometer Gravitational-Wave Observatory (LIGO) upgrade to Advanced LIGO Plus (operations expected to begin 2024), the Event Horizon Telescope, and ground cosmic microwave background (CMB) experiments will advance understanding of the conditions present in the infant universe, the properties of dark energy, and the fundamental physics associated with black holes.

The development of scientific priorities for this survey began with the receipt of 573 science white papers (written by 4,516 unique authors and/or endorsers) in early 2019. These papers formed the starting point for deliberations by six expert science panels, organized by subfield: Compact Objects and Energetic Phenomena; Cosmology; Galaxies; Exoplanets, Astrobiology, and the Solar System; the Interstellar Medium and Star and Planet Formation; and Stars, the Sun, and Stellar Populations. Each panel reviewed recent progress in its field and identified key challenges and priorities for the coming decade and beyond. Their reports are included as appendixes to this report. From this wide array of opportunities, each panel identified four key science questions regarded as being especially ripe for investigation in the coming decade, along with a “discovery area” where emerging capabilities or techniques offer great promise for major advances. Tables 2.1 and 2.2 at the end of this chapter provide a summary list of these questions and discovery areas. These are not intended to encompass all of the important science needed, but rather to highlight particularly important questions and opportunities.

The panel reports were then integrated by the steering committee into the summary science case that forms the remainder of this chapter. It soon became clear that most of the science questions and discovery areas could be organized into three broad thematic areas: Worlds and Suns in Context highlights the extraordinary advances over the past decade in the study of exoplanets, stars, and their associated planetary systems, and the opportunities for transformational advances in these areas, including the ultimate search for and characterization of habitable planets, in the decades ahead. Cosmic Ecosystems represents an integration and culmination of understanding the origins of galaxies, stars, planets, and massive black holes, and the realization that the life cycles of the universe over this billionfold range of scales are intimately connected, through feedback processes propagating through the gas within, surrounding, and between galaxies. The New Messengers and New Physics theme embodies the dual revolutions brought about by the marriage of observations of light with those from gravitational waves and elementary particles (multimessenger astrophysics) along with the expansion of measurements of the sky over time (time-domain), as well as the opportunities for major advances answering some of the most fundamental questions in astrophysics and physics, such as the nature of dark matter and dark energy. Each of these three themes is summarized below, and Tables 2.1 and 2.2 list the thematic areas associated with each of the science panel questions and discovery areas. All are represented in the themes and many cross multiple themes.

Although the three overarching themes effectively distill the multitude of science goals and priorities contained within the six panel reports, their encompassing natures do not provide clear scientific goals that can be easily grasped by the non-professional readers of this report. With that in mind, the committee has identified a key priority science question for each of the themes, which will help to motivate the strategic program later in this report: Pathways to Habitable Worlds; New Windows on the Dynamic Universe; and

Unveiling the Hidden Drivers of Galaxy Growth. The priority science areas for each theme are described at the end of each respective section.

2.1 WORLDS AND SUNS IN CONTEXT

The planets in our solar system, and the Sun at the center of it, provide the most direct connection to the myriad other stars and planets in our galaxy and the universe. With the flowering of capabilities expected in the next decade, progress in both stellar astrophysics and planetary science will expand and provide a broader context with which to understand and appreciate our cosmic perspective.

Within the past decade, progress in planetary demographics has achieved a reversal of sorts; new knowledge that planets are common, along with multi-planet systems, translates to a new cosmic understanding that there are likely more planets than stars in the universe. The Copernican revolution continues, in the realization that among the incredible variety of exoplanetary systems, our own solar system may prove to be an outlier rather than an exemplar. Along with pondering the cosmic order to understand stars, their formation, evolution, and ultimate fate, there is a parallel track for planets—how do they form and evolve?—and a merging of these tracks when considering questions related to habitability. It is an exciting time in which to practice the astronomical craft, as humanity edges ever closer to being able to answer the age-old question “Are we alone?” It is humbling and exciting to contemplate that the question of whether life exists elsewhere could be answered with the technology humanity now possesses.

The past decade has also witnessed a renaissance in stellar astronomy. Gaia, an ESA mission to deliver fundamental stellar parameters such as distances and three-dimensional motions on the sky, has proved revolutionary even with its initial data releases, to articulate the connections between and among groups of stars. Other time-domain observatories primarily designed to detect exoplanets have returned a wealth of asteroseismological observations allowing us to probe the interior structures of stars. Whereas in the past only a few fundamental parameters of a star could be determined accurately, and most of these in a relative sense, now mass, size, distance, age, and chemical makeup for a wide swath of stars are measurable. The coming decade will continue this trend: where Gaia enabled precision stellar parameters of roughly a billion stars in the Milky Way, the Rubin Observatory promises to expand by a factor of 10 the number of main sequence stars for which distances can be determined.

2.1.1 Stellar Demographics

We are in the midst of a stellar renaissance, as astronomers come to know the individual journey of each star, separating these stars from anonymous points of light in the heavens to having the equivalent of detailed dossiers of physical characteristics and histories.

This increasing complexity of stellar astrophysics extends to exposing the internal states of stars and the insights that come along with that revelation. Asteroseismology was a once-bespoke method of sounding the internal state of a star via detections of oscillation modes. The technique is now an established tool to determine precise stellar ages in large data sets, as well as returning masses and radii that can be used to calibrate models. The emergence of ultra-precise, long-duration, and continuous light curves from space (first with the European mission Convection Rotation and Planetary Transits [CoRoT], then with NASA’s Kepler and TESS as well as ESA’s Characterizing Exoplanets Satellite [CHEOPS] and soon, ESA’s Planetary Transits and Oscillations of stars [PLATO] mission) served both the exoplanet community and the stellar astrophysics community by enabling planet detections via the transit method and vastly expanding the number of stars for which asteroseismic oscillations (and subsequent stellar astrophysics studies; see below)

could be detected. Probes of the internal states of stars via this method now return constraints on stellar structure previously only theorized. The cores of red giant stars appear to rotate faster than the surface, and oscillation frequencies differentiate red giant stars in which core helium burning is occurring, versus those only burning hydrogen in a shell. Latitudinal differential rotation in the convection zones of Sun-like stars, revealed through asteroseismic observations, indicate a shear much larger than predicted from numerical simulations. The mass distribution of red giant stars probed by asteroseismology does not match predictions from stellar population models. Asteroseismology detects strong magnetic fields in the cores of red giant stars. Indeed, high-precision, high-cadence light curve observations now join traditional methods of photometry and spectroscopy as essential tools of observational stellar astrophysics.

The next decade will continue to provide precision tests of stellar evolutionary models, and ultimately advance our understanding of the structure, dynamics, and evolution of galaxies as a whole (Figure 2.2). The elements present in a star’s atmosphere reveals its origins and refines the understanding not just of that star, but when applied to large samples of stars, reveals how the assemblage of stars in our galaxy came to have its form, structure, and content. Identifying and studying particular stellar subsamples such as cool subdwarfs provide the fossil record of the early history of star formation in the Milky Way Galaxy, as their elemental compositions are unchanged from that at their birth. Necessary links in this chain of chemical tagging are improvement in the laboratory and numerical calculations of atomic and molecular transitions and opacities, together with inclusion of more realistic geometries beyond assuming that stars are spherical, and non-equilibrium effects in stellar model atmospheres. In the coming decades, high-resolution spectroscopy with extremely large telescopes will expand these abundance measurements throughout the Local Group. Connecting this chemical record of the galaxy with the dynamical record from Gaia and spectroscopy should produce a definitive fossil record of the assembly and life history of our galaxy. Progress in the next decade will require advances in industrial-scale spectroscopy (e.g., Sloan Digital Sky Survey [SDSS] V and future even larger surveys) along with computational methods to harness the large data sets of photometric, spectroscopic, and astrometric surveys.

Whether a star has one or more partners, and the nature of those partners, influences its life history because of mass exchange, particularly for massive stars. This is especially true for the end state of stellar evolution, as the formation of compact objects provides connection points to probes of extreme gravity and extreme particle acceleration. The detection of gravitational waves from astrophysical objects this decade leads to invigorated research into the endpoints of stellar evolution for the next decade. Stars in multiple systems can have very different evolutionary pathways compared to their single counterparts; detailed photometric and spectroscopic electromagnetic observations spanning infrared through X-ray wavelengths, coupled with gravitational wave measurements and attention from theory, will elucidate these multiple routes and their consequences. Theoretical modeling of binary co-evolution is necessary to understand the fate of close binaries that do interact. White dwarfs caught in the act of merging will be studied by the Rubin Observatory and can be linked to resolved gravitational wave signals detected by LISA once it launches, as these systems will also likely contribute a major part of the stochastic background expected from the Milky Way Galaxy. Further observational and theoretical research will sharpen constraints on the mass threshold that separates an end state of a white dwarf from core-collapse supernova formation. The effects of rotation, binary, and magnetic fields can be implemented in population synthesis models, with wide applicability ranging from understanding the ionizing output of massive stars to interpreting the spectra of distant galaxies. More generally, mapping out the myriad evolutionary pathways by which stars come to populate every part of the Hertzsprung-Russell diagram will require a robust observational and theoretical understanding of the formation, evolution, and especially interaction of stars in multi-star systems.

The angular momentum of a star, typically measured by its surface rotation period, is a key parameter to unlocking its current state and is a fundamental parameter in its own right. Determining rotation periods for a few tens of stars would in years past have been the subject of a Ph.D. dissertation. Now, with automated light curve analysis of stellar targets to search for transiting planets, a click of a button is practically all that is needed to compute rotation periods of tens of thousands of stars within a single research paper. This has proceeded in recent years largely as the stellar by-product of exoplanet studies (Figure 2.3), a fortuitous result but one biased by the particular target selection criteria used for exoplanet searches. The next decade will see rotation rates of stars (and other time-domain stellar astrophysics measures such as flaring) determined over nearly the entire galaxy, likely with the aid of modern machine learning practices. Application of these rotation periods as clocks for stellar age dating converts stellar time-domain astronomy to the industrial-scale extraction of stellar birth dates. This furthers the personalization of stellar histories and adds precision to testing stellar models, as well as providing ground-truth age dating to constrain models of the Milky Way’s formation and evolution history.

Among all the detailed stellar knowledge, astronomy still discovers entirely new types of stars and stellar phenomena. Progress in the next decade is expected to reveal more and different types of stellar exotica, as testaments to the incompleteness of the knowledge about the many forms and varieties of stellar behavior. The unprecedented dimming of Betelgeuse, one of the brightest stars in the sky and the shoulder of the Orion constellation, from December 2019 through May 2020, proved that even well-studied stars can throw an unexpected astronomical curveball. Multi-wavelength observational capabilities that include ultraviolet and even X-ray wavelengths provide key diagnostics of temperature, density, abundances, and kinematics, essential for the post-mortem understanding of these events. The rare phenomena, such as the unusual dimmings of Boyajian’s star, sparsely erupting pre-main sequence stars, or the elusive luminous blue variables, will become commonplace, and understanding will hopefully follow discovery of stars and their environments in a virtuous cycle.

The Sun is a singular star among all the others, primarily for its proximity but also because of our dependence on its behavior for our existence. Observations of the Sun are necessarily a touchstone for virtually all of stellar theory, with ripples throughout the understanding of all of stellar astrophysics. The

anticipated first science to be done with the revolutionary Daniel K. Inouye Solar Telescope (DKIST) facility promises more answers to fundamental questions of how magnetism controls our star. This 4 m, solar-dedicated telescope on the ground (a “Hubble for the Sun”; see Figure 2.4) represents a large leap in solar observing capabilities. The interplay between magnetic flux and mass flows is of universal astrophysical importance. This decade’s detailed observations will have the requisite spatial, temporal, and spectral resolution as well as the dynamic range to provide a ground-truth for models of basic magnetic structures. Advances originating from this zoomed-in view of solar magnetic structures require complementary global measurements—particularly of the solar corona—to understand magnetic energy storage and release. Complementary radio observations could generate three-dimensional mapping of the magnetic field in sunspots, and magnetic field measurements in the global corona provide context for the more detailed, restricted field of view measurements.

The singular nature of the Sun is both its blessing and its curse. Its ability to provide constraints on astrophysical questions with unrivaled accuracy is in tension with its fitness as a prototype for all things stellar. While it is a spectacularly well-studied star, it is a unique case study observed at one point in its 4.6-billion-year evolutionary history. Observations in the past decade provided tantalizing hints that the Sun’s cycle of magnetic activity does not behave the same as other solar-like stars. Work in the next decade will broaden the range to include a larger sample and provide the perspective of stars of different ages and spectral types. While hints exist now of a disconnect in the relationship between fundamental stellar parameters and magnetic properties—magnetically active stars can have larger radii than predicted based on their temperatures; stellar twins identical in all other properties can have very differing levels and amounts of magnetic field distribution and magnetic activity signatures—observations in the next decade will further this discovery space and fill important gaps in the foundation of stellar structure theory.

On a broader scale, it is clear that stars are not uniform discs of light. Time-domain astronomy enables the characterization of surface structure using the changes in light seen either in broad wavelength bands (Figure 2.3) or in narrow spectral features that reveal inhomogeneities. Optical and infrared interferometric observations have begun to localize these regions. Observations spanning multiple regions of the electromagnetic spectrum provide independent constraints on the properties of these features, characterizing the magnetic, chemical, or cloud-like nature of the features. Indeed, the ability to measure rotation periods arises from the periodic brightening and dimming caused by blemishes passing across the face of each star. On stars like the Sun, those spots reveal the telltale existence of subsurface magnetism; the multitude of long-duration high-precision photometric light curves available from space missions, including CoRoT,

Kepler, and TESS, reveal the ubiquity and variety of surface inhomogeneities of cool stars. Future progress in interferometers, from optical to infrared to radio wavelengths, will advance knowledge of the underlying physics causing spatial inhomogeneities on stellar surfaces.

The importance of magnetic fields in a stellar context was expected but is now unequivocal. Magnetism in massive stars has been confirmed (but not entirely explained) in a minority of cases, and stronger magnetic fields than initially expected have been found at the end of the main sequence, in the ultracool dwarfs spanning stellar and substellar origins. A variety of observational tools enable these discoveries: high-energy observations detail the magnetic shaping of stellar winds and unveil the existence of heated plasma in an above-the-surface corona; optical and infrared spectra reveal the telltale signature of magnetic splitting of atomic and molecular lines; and radio observations expose the action of accelerated electrons, thereby revealing the presence of magnetic fields in their atmospheres. Stellar coronae are common in cool main-sequence stars, but the origin of this heated plasma is still unknown, even on the Sun. Future sensitive high-energy and radio measurements will probe the coronae of stars in more detail and expand the number of objects amenable to study. Ultraviolet spectra probe the chromosphere and above, regions in a cool star’s atmosphere where magnetic forces begin to dominate. These observational advances also require associated advances in modeling the intricacies of stellar structure for full understanding of the impact of convection, rotation, and magnetic field generation on the complex and dynamic nature of stars.

2.1.2 Exploring Alien Worlds

Over the past two decades, an incredible diversity of worlds and systems has been discovered, from giant planets over 10 times the mass of Jupiter to the Trappist-1 system with seven Earth-sized planets packed in an area smaller than the orbit of Mercury. This revolution is ongoing. New capabilities led to an explosion of discoveries in the past decade (Figure 2.5), filling in demographics of planets, expanding the ability to characterize the composition and nature of individual members, and advances in new facilities adding to the depth of characterization possibilities.

These discoveries are helping us understand fundamental questions about ourselves. How did the solar system form? Are systems like our own common or rare? Are planets like Earth common or rare? And, ultimately, do any of those Earth-like planets harbor life? With current technology, the planets in solar systems like our own are nearly undetectable; with new facilities over the next two decades, this will change, and the picture of other worlds will start to become complete.

Finding exoplanets is challenging. Many exoplanet detection methods have been developed, and each gives some important, but limited, information. The radial velocity (RV) technique measures orbits and constrains the mass of a planet using the motion of its parent star. This technique has been used for a number of exoplanet “firsts” over 25 years (including the Nobel Prize–winning discovery of 51 Peg b), and astronomers are pushing the technology to be able to detect Earth-mass planets on periods of months, to perhaps a year, in search of objects similar to Earth. In the past decade, another technique, exoplanet transits, has transformed the view of exoplanetary systems, in particular when employed via high-precision space missions like NASA’s Kepler and TESS missions, and before them the European CoRoT mission. A transit, when a planet passes directly in front of its parent star, allows for a measurement of a planet’s radius and orbital period. Kepler showed that planets on close-in orbits (within 100 days) are extremely common (Figure 2.5), which has revolutionized the understanding of planet formation, and is one of many examples that show that architecture of planetary systems is exceedingly diverse.

The radial velocity and transit methods are best suited for finding planets on relatively close-in orbits. A method prioritized by Astro2010, gravitational microlensing, will be used by the Roman Space Telescope in this decade to complete the planetary census by finding planets from 1 to 100 AU (astronomical units),

and even free-floating planets. Microlensing exploits the bending of light by the gravity of the star and planet to detect distant systems. This technique will be powerful for measuring how common such planets are on these wider orbits. However, these planets cannot be followed up for future in-depth observations, because this chance star-planet alignment does not repeat.

Another major technique of long-term importance to the field is direct imaging, where the light of the parent star is blocked to make a faint planet directly visible. Currently, this is only practical for young giant planets in the outer parts of solar systems. However, with suitable technology development, there is now a clear path forward to use direct imaging techniques on ground-based extremely large telescopes (ELTs) and optical interferometry with 8 m–class telescopes to study the atmospheres of temperate rocky planets around low-mass stars, and to use future space missions to study potentially Earth-like planets around Sun-like stars, as well as the enormous diversity of non-Earthlike planets.

With these diverse techniques, astronomers have produced a partial census of exoplanets. The results have been extraordinary, showing that on the average there are at least two planets per star in the Milky Way and that many planetary systems are very different from our own, with crowded systems of planets intermediate between Earth and Neptune in size. So far, it appears that lower-mass M-dwarf stars have more planets than Sun-like stars. Careful characterization of the stars themselves has been crucial to interpreting Kepler results, and broader studies will reveal how these distinctions persist across different regimes. The census remains incomplete; all current planet detection surveys would be essentially blind to every planet in the solar system at their known orbital separations, except Jupiter.

The Kepler Mission has also shown that the changing nature of planets—their time evolution—is also fundamental to understanding the planetary population. Figure 2.6 shows a recent discovery using Kepler data, of a fundamental divide between rocky planets below about 1.7 times the radius of Earth and larger “sub-Neptunes” that have thick hydrogen-dominated envelopes that increase the planetary radius. The “gap” between these two populations is thought to be owing to the gradual loss of hydrogen-envelopes, possibly owing to a stellar wind that drives these atmospheres off the planet and into space. The detection of this gap was made possible because of exquisite characterization of the host stars and shows the important connection between stellar and planetary physics.

Approximately 20 Earth-sized (<1.7 R_Earth) planets have been discovered within the habitable zone of low-mass stars—neither too close nor too far, but receiving enough energy to allow liquid water on their surface. A key question that Kepler and other projects have tried to address is the frequency of potentially habitable planets—the average number of Earth-sized planets within the habitable zone of their star, particularly around Sun-like stars. A decade ago, this quantity—known as η_Earth—was almost completely uncertain. These planets are very difficult to detect directly, so their exact occurrence rate must be extrapolated from planets higher in mass, closer to their star, and/or from the frequency of habitable zone planets orbiting lower-mass stars. Current constraints indicate that 18–28 percent of Sun-like stars have an Earth-sized (80 to 140 percent of Earth’s radius) planet in their habitable zone, enough to make their study with a future large mission practical. (The number is likely a factor of 2 higher for cooler lower-mass stars.) The nature of these planets is of course almost completely unknown, whether they have followed a path of planetary evolution amenable to life or something completely different from Earth; but that very uncertainty is a compelling scientific reason to try and answer this most profound question.

2.1.2.1 Planet Formation

The detailed process of planet formation is one of the great unknown frontiers in astrophysics. The core concept of gas and tiny dust particles in a circumstellar disk assembling into planets is well known. But the details of how particles assemble, how the process overcomes barriers to operate quickly enough

before the disk dissipates, how this leads to the incredible diversity of known planets, and whether this process can frequently produce planets with Earth’s key characteristics, remain unknown. Understanding this is crucial to placing potentially habitable planets in context, including questions as to how water and atmospheric volatiles are delivered to Earth, whether giant planets affect the evolution of terrestrial worlds, and whether “mini-Neptune” planets can evolve toward habitability. More sensitive observations of planet-forming disks to understand the astrochemistry, dynamics, and role of water in the formation of habitable planets through radio, mm, and far-infrared (IR) spectroscopy will help advance this field.

Planetary demographics and composition reveal only the endpoint of this process. Stunning observational advances from Atacama Large Millimeter/submillimeter Array (ALMA) show complex structures in planet-forming disks (see Figure 2.20), and recent advances with large ground-based telescopes have even caught a protoplanet with its own accretion disk (Figure 2.7). These structures, however, still correspond to large planets in the outer parts of solar systems; in the habitable zone or the regime of the Kepler super-Earths the process itself remains almost inaccessible. Current observations of continuum dust emission cannot probe the innermost regions of the planet-forming disks where Earth-like planets may be located; sensitive radio interferometry at longer wavelengths than those probed by ALMA is needed to see the optically thin emission originating in these locations. Spectra at far-IR wavelengths would provide a

unique and revolutionary census of water within these disks, which is a key to understanding giant planet formation and the distribution of water among terrestrial planets. New radio and far-IR surveys of planet-forming disks would enable leaps in understanding that would surpass this era’s ALMA-driven revolution.

There have been tremendous advances in the theory that underlies planet formation, as new ideas such as the “streaming instability” and “pebble accretion” use gas and particle interplay and physical changes in the disk to trap planet-forming material and rapidly form giant planets. The models incorporate uncertainties such as turbulence and feedback between the forming planets and disks. Better observations of gas and dust in disks (especially on smaller scales and across a range of ages) and of forming planets (especially of lower masses) will advance this field, combined with larger-scale computational models and laboratory experiments.

2.1.2.2 Atmospheric Spectroscopy to Characterize Worlds

Only a small fraction of the thousands of known planets have been characterized beyond their basic mass or radius, along with orbital properties, but those available observations further illustrate the diversity of worlds. In addition to completing the planetary census, the other major frontier of the 2020s is the spectroscopic observation of planetary atmospheres. A spectrum can yield the abundances of atoms and molecules as well as the temperature of a planetary atmosphere (Figure 2.8). Already several dozen planetary atmospheres have been characterized by the Spitzer and Hubble space telescopes and by large ground-based telescopes. Observations of the transiting and directly imaged planets target the extreme inner and outer reaches of planetary systems, respectively. So far, this work has mostly focused on the easier to

observe giant planets, but already the diversity of planets beyond what is seen in the solar system is on full display. Detections have been made of clouds of rock dust in the hottest planets, metal vapor such as sodium, potassium, iron, and calcium, escaping hydrogen, and molecules including water vapor, carbon monoxide, and methane. Abundance determinations show the current state of these atmospheres, and—as in the solar system—the history of planetary formation and evolution is embedded in them.

NASA’s TESS mission, along with ground-based surveys, are finding planets around the nearby bright stars, to further fuel this revolution in atmospheric spectroscopy. The premier tools for obtaining exoplanet spectra in the 2020s will be JWST and extremely large ground-based telescopes. These telescopes will revolutionize the understanding of the composition of exoplanets. Atmospheres are the window into many physi-

cal, chemical, and formation processes, and these platforms will deliver spectra for a continuum of worlds, from Jupiter-size gas giants, to the Neptune-size planets, to the mysterious mini-Neptunes that dominate the exoplanet census, and down to Earth-size rocky worlds. The lessons of atmospheric physics and chemistry learned from one planetary class can readily inform the understanding of other classes. As astronomers will be far outside the realm of solar system expertise, and while tentative predictions of what to expect have been made, the joy of discovery will be seeing the diversity of these new worlds. In particular for terrestrial planets, significant progress in the 2020s will occur for systems around low-mass M-dwarf stars. Nearly all of this science will focus on planets within a few tenths of an AU of their parent stars, orbits tucked in far closer than Mercury is to the Sun. Moving beyond this to characterization of systems that look more like the solar system will require new capabilities. To understand potentially Earth-like planets around stars like the Sun, which is the only example, so far, for life, will drive the field toward even loftier goals beyond the 2020s (Box 2.1). Are such planets habitable? Do they show signs of life?

2.1.2.3 Connections to the Solar System

Studies of exoplanets and of the solar system are tightly intertwined and have enjoyed a profitable give and take in contributing to the understanding of planet formation and atmospheres. Many aspects of exoplanet atmosphere modeling use numerical techniques originally developed for the solar system’s own giant and terrestrial planets. The solar system provides “ground truth” to validate spectroscopic observations against in situ measurements such as elemental abundances on Jupiter or Earth or hazes in satellite atmospheres (Figure 2.9). On the other hand, the vast diversity of exoplanet types—many with no analog in the

solar system—and the opportunity to study systems over a range of ages provides insights into formation mechanisms and evolutionary processes that improve the understanding of the solar system’s own history.

The solar system contains asteroids and comets as remnants of its original formation. Analogs around other stars take the form of debris disks, massive dust belts produced by collisions among these small bodies. Measurements of the composition, orbital dynamics, and size distributions of small bodies in the solar

system provide crucial benchmarks for understanding solar system formation in one spectacularly detailed instance. Time-domain surveys such as the Vera Rubin Observatory’s Legacy Survey of Space and Time will greatly expand the number of known small bodies in the outer solar system, and provide information about its early evolution. Results from recent studies analyzing dynamics of small bodies in the Kuiper Belt provide tantalizing hints, to be confirmed, about the possibility of additional planets.

While the detailed information available from solar system observations informs theories of planetary system formation in general, such studies are also synergistic. Observations of young planet-forming disks provide a window into the early conditions that led to the formation of the giant planets in this solar system. New generations of radio interferometers will probe inner solar system scales of planet formation, approaching scales of Earth’s orbit and super-Earth planet masses. High angular resolution near- and mid-infrared observations of the innermost regions of planet formation around young stars directly image thermal continuum emission of forming planets and provide kinematic constraints. Increased capabilities to track circumplanetary disks and substructure add the possibility to measure orbital motions of Earth-like planets as they are forming and provide context for factors potentially affecting our early Earth. Detection and study of complex organic molecules provide the chemical initial conditions of forming solar systems. With these high angular resolution and high sensitivity techniques, young planetary systems can be observed—analogs to the solar system—at their moment of formation, validating models that explain the origin of the life-bearing conditions on Earth.

As the only place for which in situ and other detailed studies are possible, advances in knowledge of the solar system are a crucial part of the astrobiological endeavor, and astronomical observations in turn inform the understanding of the planets in direct reach. Future space coronagraphic observations of young Venus or Mars analogs, for example, could help confirm whether those planets had more Earth-like atmospheres in the past. Continued solar system space missions, complemented by long-term monitoring from telescopes at Earth, provide essential details to broadly understand these planets. Comprehensive theories of planet formation must be able to explain both the solar system and the diverse array of known exoplanetary systems.

One of the most exciting astronomical discoveries of the past few years is two interstellar interlopers—the asteroid ‘Oumuamua and the comet 2I/Borisov—that originated around another star and passed through our solar system. Large-scale surveys such as the Rubin Observatory Legacy Survey of Space and Time will discover many more such objects and will provide an increased understanding of context and impact for these interstellar visitors.

2.1.2.4 The Star-Planet Connection

Stars and the planets that orbit them are inexorably connected to each other by their evolutionary history. The star’s properties and evolution influence the evolution and habitability of the planets, particularly of terrestrial planets. The end stages of star formation provide the initial conditions for planet formation. The coming decade will see the implementation of a systems-level approach to understanding the many factors that influence a planet’s habitability.

Most workhorse planet detection methods are a relative measurement, made with reference to the properties of the host star. Stellar surface inhomogeneities pollute the planet measurements due to the combined light of the system and the breakdown of simplistic assumptions about the stellar surface characteristics. Starspots and other variations produce spurious Doppler shifts that can mask Earth-sized planets. Precision planet characterization in the coming decade will motivate the need for better knowledge about the star, particularly its magnetic properties. Conversely, close-in planets can be used as test particles to reveal small-scale stellar inhomogeneities (Figure 2.10) and increase understanding of stars.

A star can influence its near environment by a combination of its radiation, gravity, and particles. The difference between “super-Earth” and “mini-Neptune” planets can potentially be explained by processes driven by the star. A star’s magnetic field is responsible for the heating producing stellar emissions above the photosphere, which manifests as enhanced ultraviolet through X-ray radiation. High-energy stellar radiation is a risk to, and potentially a catalyst for, life. The UV emission of a star influences the planetary

atmosphere photochemistry, and can create false positives for biosignatures. The extent and amount of high-energy radiation from the star determines how much of the atmosphere a close-in planet keeps over evolutionary time. For planet-hosting stars that are cooler than the Sun in temperature, the magnetic field is also complicit in the star losing mass, through a steady stellar wind and potentially transient coronal mass ejections. Eruptive events—characterized as some combination of stellar flaring, coronal mass ejections, and energetic particle events—produce space weather and in extreme events, influence habitability. Magnetic fields of low-mass stars may prevent some eruptive events from ejecting mass, lending complexity to a blind application of the solar analog, and there are currently few observational constraints on stellar

winds or stellar coronal mass ejections. The next decade will see progress in characterizing the habitability prospects, particularly of M dwarf planets. Due to their proximity to the star, they are the most vulnerable to atmospheric loss, coronal mass ejections, tidal heating, and orbital evolution.

Priority Science Area: Pathways to Habitable Worlds

Among the many exciting discoveries and opportunities ahead within the Worlds and Suns in Context theme, one question stands out: “Are there habitable planets harboring life elsewhere in the universe?” The answer to this question, and the questions that immediately follow—“Is the Earth unique?” “Are humans alone?”—would have impacts reaching far beyond astronomy, within science and humankind overall. Advances in the understanding of exoplanets and in astronomical instrumentation now allow the planning of major steps toward identifying and studying candidate habitable planets, and making the first tests for the signatures of habitability and life. Laying out the Pathways to Habitable Worlds is the priority science area for this theme.

Meeting this ambitious goal will require progress on a variety of fronts, observationally, theoretically, and in the laboratory. Potentially habitable planets are the exception rather than the rule (roughly 20 out of the ~2,400 planets discovered by Kepler; see Section 2.3.2), and those close enough to be studied in detail by future facilities is smaller yet, so current and planned exoplanet surveys will play a critical role in enlarging the candidate list. Ongoing efforts to characterize the surfaces and atmospheric compositions of known exoplanets, from space and the ground, will be important for quantifying the demographics of the overall planet population and refining the diagnostic tools that will eventually be brought to bear on the candidate habitable worlds.

Observing and characterizing candidate habitable planets themselves, however, whether by direct imaging or indirect observations in transiting systems, requires new telescopes on the ground and in space. The first target systems are likely to be planets orbiting close to the lowest mass and faintest stars, M dwarfs. Their low luminosities (thousands of times less than the Sun) make it easier to detect an orbiting planet against the stellar glare, but the habitable planet zones will lie close to the stars themselves. JWST may be able to measure a few such systems in transit, and spectroscopic observations of larger samples form a keystone part of the science cases for the 24–40 m ELTs (E-ELT, Giant Magellan Telescope [GMT], Thirty Meter Telescope [TMT]).

A long-standing goal ever since the discovery of the first exoplanets is to image and characterize planets orbiting Sun-like stars, including those in the habitable zone. Such stars have long been suspected as offering the most likely sites not only for liquid water but also for life, and space-based telescopes with sufficient stability and high-contrast capabilities to image and spectroscopically measure planets around the nearest such stars are now within our reach. A space telescope similar in wavelength coverage to the Hubble Space Telescope (HST), and with an aperture of at least 6 m and coronagraphic imaging capability, should be capable of observing approximately 100 nearby stars and of successfully detecting potentially habitable planets around at least a quarter of the systems. Such an observatory would also provide valuable information on other extrasolar planets, and be versatile enough to carry out groundbreaking observations of stars, galaxies, black holes, and the gases and baryons within and between galaxies, with a scientific impact rivaling that of previous “great observatories” such as HST.

The potentially habitable worlds ripe for discovery in this decade and the next represent only the tip of a deep iceberg in exoplanetary exploration. Studies of super-Earths and mini-Neptunes as well as larger Jupiter-sized planets fill in critical areas; these measurements can be done with a range of facility sizes and round out a comprehensive approach to the subject. Considering solar systems as a whole and understanding how the individual components interact with each other is a crucial component of understanding the

processes at work to arrive at the observed state. While the ultimate quest is to find potentially habitable planets, the perspective of a wide variety of planetary demographics, characteristics, and evolutionary paths (including those that lead away from habitability) is needed to understand where the multiple highway exits lead along the route to answering the question “Are humans alone?”

The path to habitability starts at the beginning of a planet’s journey and requires investigation of the chemical and dynamical processes at work to determine conditions early on in a planet’s life that lead both toward and away from habitability. Improvements in imaging and spectroscopy with future large facilities at mm and sub-mm wavelengths will probe the rings and gaps caused by forming planets in the disk of gas and dust around a young star and create a census of the properties of these forming planets. Spectroscopic probes of planet-forming gas reveal the chemical initial conditions that solar systems and individual planets in formation experience; the study of complex prebiotic species paints the picture of chemical evolution pathways needed at the beginning of this path. Measuring the water reservoirs in star- and planet-forming disks is crucial for understanding the mechanisms by which terrestrial planets gain their water.

Although this mapping of the Pathways to Habitable Worlds has emphasized the central roles to be played by major existing and planned observatories, success will also require a wide array of enabling and foundational projects and studies. These include observational and theoretical studies of the linking between star and planet formation, the chemical processes—both inorganic and organic—critical for planet formation, the evolution of planetary atmospheres, and the origins of life, laboratory measurements of spectroscopic tracers and astrochemical processes, and detailed study of stellar activities, variability, and surface structures, all of which can mimic observational signatures of transiting

2.2 NEW MESSENGERS AND NEW PHYSICS

Our understanding of the universe has been repeatedly transformed by observing across the entirety of the electromagnetic spectrum, from the radio to the gamma rays. These observations can test or reveal physics in ways that are not possible on Earth. Now, nearly daily movies of the sky, as well as the new messengers of neutrinos, particles, and gravitational waves, provide new ways of doing astronomy and new methods for uncovering and testing new physics.

Observing the universe in new ways has historically involved expanding our observations to cover the full electromagnetic spectrum, not just the part accessible with our eyes. In doing so, X-ray observations revealed the dynamic coronae of the Sun and other stars, accreting neutron stars and black holes, and the hot plasma that pervades interstellar and circumgalactic environments. Radio observations revealed the existence of neutron stars, whose remarkably stable rotation rates have since been used to discover planets and confirm the theory of general relativity’s prediction of orbital decay via the emission of gravitational waves. And telescopes observing in the infrared can peer into the enshrouded stellar nurseries where stars and planets form.

New views of the universe also come from observing in new ways—for example, by making observations with much higher sensitivity, angular resolution, or time resolution, or by obtaining higher dynamic range views of previously hidden phenomena. The movies of stars orbiting the 4 million solar mass black hole in the center of the Milky Way Galaxy would not have been possible without adaptive optics and near-IR interferometry to overcome the blurring effects of Earth’s atmosphere. More recently, the Event Horizon Telescope’s unprecedented angular resolution at millimeter wavelengths (via interferometry) provided the first direct image of the near-horizon environment of a black hole (see Figure 2.11), captivating scientists and the public alike.

By observing ever fainter sources across the electromagnetic spectrum, it is also possible to peer back into the distant past when the universe and its constituents were young, and unravel the history of the

universe. Perhaps most spectacularly, observations of the CMB radiation measured the geometry and mass-energy content of the universe and helped transform cosmology into a precision science.

The most radically different views of the universe are provided by messengers other than electromagnetic radiation in the form of photons or waves. Neutrino observations confirmed the theoretical prediction that hydrogen fusion powers the luminosity of the Sun and demonstrated that neutrinos have mass, a key insight into physics beyond the standard model of particle physics. Cosmic-ray measurements have found particles whose energies are enormous compared to those that can be produced in the Large Hadron Collider (LHC) at European Council for Nuclear Research, but whose astrophysical origin remains a mystery. In 2013, the south pole IceCube observatory detected a diffuse high-energy neutrino flux of astrophysical, but unknown, origin. Starting in 2015, LIGO opened up the gravitational wave view of the universe by detecting merging binary black holes. The simultaneous detection of gravitational waves and electromagnetic radiation from a binary neutron star merger in 2017 showed the power and complementarity of multi-messenger observations (see Box 2.2).

As views of the universe have expanded, so has astronomy’s impact on basic physics. Many frontier science questions identified by the survey’s science panels center on the intertwined themes of using new techniques to see the universe in new ways (new messengers) and uncovering new physics with advanced

astronomical observations. In what follows, the presentation of this science theme is organized by first discussing cosmology and the dark sector and then turning to the new astronomy enabled by observations with particles, neutrinos, gravitational waves, and light.

2.2.1 Cosmology and the Dark Sector

The fundamental paradigm of modern cosmology is the Hot Big Bang, in which an initially hot, dense, and nearly smooth universe rapidly expands and cools. All evidence suggests an initially simple universe, made of a nearly uniform collection of light nuclei and electrons, a sea of radiation, a similar sea of cosmic neutrinos, and dark matter of an unknown nature. As time passed, small initial differences in density grew under the action of gravity to form the rich structure described in the Cosmic Ecosystems science theme of this report. The transition from a smooth universe to one with stars and galaxies occurred less than 500 million years after the Big Bang. Finding innovative ways to probe cosmology in the “dark ages” prior to any significant star formation is one of the discovery areas identified here. The development of “LambdaCDM,” our current standard cosmological model, is one of the major intellectual triumphs of the past few years; the nature and origin of its key ingredients—dark matter, dark energy, and a nearly scale-invariant spectrum of primeval mass fluctuations—remain, however, some of the biggest mysteries in science.

Observations of the motions within galaxies and clusters to the large-scale distribution of intergalactic gas and the CMB require more matter than can be explained by the observed baryons. In the common interpretation, this is cold dark matter, an unseen gravitating material—nearly 10 times more abundant than baryonic matter—that moves non-relativistically in the recent universe. Over the many decades since dark matter was posited by Fritz Zwicky to explain galaxy motions in the Coma cluster and by Vera Rubin Observatory to understand galaxy rotation, astronomers have learned primarily what it is not: not like anything that has been seen on Earth. It could be a particle with the mass of a proton that does not significantly interact with normal matter or, at another extreme, it could be a “particle” with a quantum-mechanical wavelength the size of a small galaxy. While physicists attempt to identify dark matter through ambitious and excruciatingly careful laboratory experiments, astronomers in the coming decade will wield the threefold tools of theory, simulation, and observations to search in parallel. Dark matter could leave detectable traces of its potentially more complex interactions through deviations from the simplest version of the cold dark matter paradigm or through emission of unexpected particles (gamma rays, positrons, narrow radio frequency lines) produced by dark matter interactions. The signatures of complexity in the properties of dark matter could come from the most distant sources (e.g., the CMB) or some of the nearest (e.g., nearby wide stellar binaries or the internal kinematics of dwarf galaxies). Nearly all astronomical facilities—current, imminent, and aspirational—contribute to the study of dark matter. New large ground-based optical-infrared telescopes would be particularly impactful—for example, by studying the internal motions of stars in dwarf galaxies and testing the nature of the dark matter that holds those galaxies together.

Even with the current poor understanding of what dark matter is, there is broad consensus around how it behaves gravitationally on large scales. This allows theoretical models to robustly connect the final distribution of dark matter to the conditions in the very early universe. Astronomers are on the brink of using this connection to make 3D maps of the dark matter distribution and use those maps as probes of fundamental physics. The properties of the dark matter distribution can be traced by the ways it subtly shifts light rays as they propagate through the universe, through the effect of gravitational lensing (Figure 2.12). Analyzing the fine details of maps of the CMB or of the shapes of distant galaxies, reveals the 2D matter distribution, which can then be deprojected into the full 3D view. This is one of the primary goals of Roman, Rubin Observatory, Euclid, and new, more powerful CMB instruments. The statistical proper-

ties of these maps will eventually be able to measure the sum of the masses of the neutrinos, which will be yet another major contribution of astronomical observations to basic physics. In parallel with lensing studies on large scales, more detailed, focused studies of individual lensed galaxies, supernovae, and quasars (accreting black holes), and even lensed stars at cosmological distances, will test the predictions of the cold dark matter paradigm on the smallest scales, where we expect deviations to be most evident.

Along with dark matter, the second great cosmological unknown is the nature of the “dark energy.” Observations of the recent universe have shown that its expansion is accelerating. This remarkable discovery is not explained by a model containing only matter (even dark matter), but instead indicates a new feature, dubbed dark energy. Arguably the simplest explanation for dark energy is Einstein’s cosmological constant—an energy and pressure characterizing empty space whose gravitational effect drives the acceleration—but the incredibly small value, compared to what is expected from quantum theories, leads physicists to think that dark energy may be more complicated. The cosmological constant predicts a specific form for the acceleration in the expansion of the universe as a function of time. Observations in the coming decade using Rubin Observatory, Roman, Euclid, and higher-resolution, higher-sensitivity CMB facilities will test whether observations are, or are not, consistent with a cosmological constant. These results will be a major legacy of the Astro2010 decadal survey. Any deviations from a cosmological constant model could touch off a second revolution as powerful as the initial discovery of the accelerating universe itself.

While the next-generation tests of dark energy are under way, astronomers are already finding hints that the current cosmological model may be incomplete. An unexpected tension has developed between two different ways of determining the present expansion rate of the universe. In one, based on measuring distances and velocities in the local universe, the current expansion rate is roughly 5 percent higher than inferred from the second method, which uses the standard cosmological model to extrapolate from early universe CMB measurements to today. Continued measurements from satellites and the ground, including those using gravitational waves, will be able to distinguish between a systematic effect in one of the methods or the need for a previously unknown component of the universe.

The fluctuations seen in the CMB are believed to have been imprinted in the earliest phases of the Big Bang during a period of cosmological inflation in which extraordinarily rapid expansion established the large-scale homogeneity and flatness of the universe while also causing quantum fluctuations that subsequently grew into the fluctuations we observe. One of the most exciting opportunities in the coming decade is that CMB measurements may reveal remnant gravitational waves from this early epoch, as depicted in Figure 2.16, later in this chapter. The presence of gravitational waves would manifest as a distinctive polarization pattern in the CMB, called “primordial B-modes,” at angular scales of 3 degrees and larger. These scales are accessible from the ground and space. If primordial B-modes are found, they will provide critical constraints on the physics of inflation and give new insights into physical processes at energy scales orders of magnitude larger than can be attained at the LHC. Efforts are already under way to detect them, but higher angular resolution and sensitivity CMB observations from the ground and in space will be needed given the difficulty of detecting the small B-mode signal amid the polarized galactic foreground.

In summary, the standard cosmological model is both a remarkable triumph and an astonishing puzzle. With a relatively modest number of parameters, it continues to match observational results despite orders of magnitude improvement in cosmological measurements over the past 20 years. However, there is a mystery novel’s worth of hints that the same model is incomplete, as the most important components are not yet understood. This represents one of the great contributions of astronomical observations to basic physics, and a continued opportunity for breakthroughs and surprises in the coming decades. Unraveling the many cosmological mysteries will require a particularly close interplay between theory, simulation, observations, and laboratory experiments.

2.2.2 New Messenger

Astronomy has long been a science rooted in the observation of light (photons). The past decade, however, has overturned this understanding of what astronomy is and could be, thanks to observations with new messengers that carry new information about the workings of the universe. Gravitational waves, neutrinos, and cosmic rays (Figure 2.13)—long viewed as largely the province of physics—have all now passed into the realm of astronomy. This is owing to multiple breakthrough discoveries in the past decade, using facilities such as Auger, IceCube, and LIGO. At the same time, astronomy’s traditional pursuit of photons is being transformed by new observational facilities that probe time variable and transient phenomena. Characterizing the time-variable electromagnetic universe has become increasingly sophisticated, thanks to pathfinding optical telescopes such as the All-Sky Automated Survey for Supernovae (ASAS-SN) and the Zwicky Transient Facility (ZTF) that are setting the stage for the Rubin Observatory in the coming decade. Outside the optical, dedicated radio surveys with the Karl Jansky Very Large Array (VLA) and the Canadian Hydrogen Intensity Mapping Experiment (CHIME), and other international facilities are uncovering new and unexpected phenomena, such as fast radio bursts, while high-energy space telescopes sensitive to explosive events like gamma-ray bursts become ever more central to interpreting signals from new cosmic messengers. It is not an exaggeration to say that nearly daily movies of the sky made across the

electromagnetic spectrum are their own form of “new messenger,” with information that is fundamentally distinct from a static view of the universe. This led to the identification of time-domain astronomy as a key discovery area in Astro2010.

These seemingly separate advances in observational techniques are in fact intimately related: most of the known and anticipated sources of gravitational waves, neutrinos, and cosmic rays are also time variable or transient electromagnetic sources (e.g., neutron star mergers, gamma-ray bursts, black hole jets, and stellar explosions). Combining information from all messengers can unravel the physics at the heart of these objects, as was demonstrated so spectacularly in the case of the binary neutron star merger GW170817 (Box 2.2).

2.2.3 Neutrinos and Cosmic Rays

Trillions of neutrinos stream through us from the Sun every second, with a similar number from the cosmic neutrino background. This is an indication of just how difficult neutrinos are to detect relative to how common they are, but in this difficulty lies their promise. Precisely because neutrinos interact so little with matter, neutrinos carry information about the inner workings of some of nature’s most important energy sources: the nuclear furnaces inside stars, the formation of neutron stars in stellar explosions, and the conditions in the jets of relativistic particles that originate near the event horizons of black holes in galaxy nuclei. Neutrinos that are detected also point directly to the celestial position of their source. In the next few decades, astronomical observations will begin to routinely measure the cosmos with these most elusive of particles. Here, it is important to distinguish between cosmological neutrinos (which, like the CMB, have redshifted to millielectronvolts [meV]), neutrinos emitted by fusion processes in stars, which are in the megaelectronvolt (MeV) range, and neutrinos in the gigaelectronvolt (GeV) range and above, which are produced by hadronic collisions between cosmic rays and ambient matter. The lattermost are the focus here, as they are an unambiguous signature of ion acceleration, follow straight-line orbits from their source to our detectors (unlike cosmic rays themselves, which are scrambled by magnetic fields), and unlike their photon counterparts, are immune from optical depth effects.

Theoretical models of the early phases of the Big Bang predict that there should be a nearly uniform cosmic neutrino background similar to the CMB. There are laboratory experiments aimed at detecting the cosmic neutrino background, but it is an extremely challenging measurement. In contrast to the sea of low-energy neutrinos produced in the Big Bang, higher-energy neutrinos are messengers from some of nature’s most dramatic events. When the relatively nearby supernova 1987A exploded, some 25 neutrinos were measured in the Kamiokande, Irvine-Michigan-Brookhaven (IMB), and Baksan neutrino detectors, which were state of the art for their times. Thirty-five years later, modern neutrino detectors would measure tens of thousands of neutrinos from a supernova in the Milky Way. As similar supernovae explode throughout the universe, they collectively produce a diffuse background of MeV neutrinos, one that should be within reach of forthcoming experiments if current theories are correct. Detection of either a galactic supernova or the diffuse background would test models of the formation of neutron stars and black holes and the core-collapse explosion mechanism. Only neutrinos and gravitational waves can peer into the inner regions of these events where densities reach nuclear scales.

The skies are full of even more energetic sources that likely produce neutrinos, some continuous, some episodic, and some transient. For example, relativistic jets—collimated beams of ejected material moving at nearly light speed—are known to emanate from supermassive spinning black holes in active galactic

nuclei (see Figure 2.11). These jets span up to millions of light years in extent. Similar relativistic jets are also associated with gamma-ray bursts (GRBs), extremely energetic events that emit in just seconds the same amount of energy the Sun will emit over its lifetime. This rich population is likely to be a significant source of neutrinos detectable with future facilities.

The prospects for future neutrino astrophysics are promising. Over this past decade, the IceCube experiment at the South Pole detected an unresolved extragalactic background of 60 neutrinos with energies in the teraelectronvolt to petaelectronvolt (TeV–PeV)² range. Their distribution on the sky indicates that they are produced by distant sources well outside our galaxy, and thus are likely to be by products of energetic events throughout the universe (much like diffuse X-ray and gamma-ray backgrounds). To date, only the Sun (Figure 2.14), SN1987A, and potentially one blazar have been imaged in neutrinos of any energy, but with new facilities, the excitement of identifying specific individual sources is likely just beginning.

Neutrinos are not the only messengers of relativistic and energetic astrophysical phenomena. Quite independently, other experiments have detected ultra-high-energy cosmic rays (UHECRs, in the ~10¹⁸ eV [EeV] range). What produces these remarkably energetic particles? Are the TeV–PeV neutrinos and UHECRs produced in the same sources? They are widely surmised to be accelerated in the relativistic jets of accreting supermassive black holes or GRBs, but this has yet to be tested observationally. Unfortunately, direct

___________________

² A teraelectronvolt (TeV) is 10¹² eV, a petaelectronvolt (PeV) is 10¹⁵ eV, and an exaelectronvolt (EeV) is 10¹⁸ eV. The protons in the LHC at CERN have an energy of about 10 TeV, orders of magnitude below those of the highest-energy cosmic rays.

identification of a cosmic-ray source is difficult, since UHECRs are charged particles and are thus deflected as they travel through the magnetic fields that permeate the universe. However, a clear directional signature would be provided by the high-energy neutrinos that the UHECRs produce in the regions where they are accelerated. Higher-sensitivity neutrino observations with better sky localization are critical for unraveling how nature’s most extreme particle accelerators work.

2.2.4 Gravitational Waves

Gravitational waves are the newest detectable messenger traveling through the astronomical landscape. They provide a unique probe of regions with large amounts of mass moving at near the speed of light: black hole and neutron star collisions, neutron star and black hole formation in stellar explosions, and the first fractions of a second of the Big Bang (Figure 2.15). The importance of gravitational waves lies in part with the central role that black holes and neutron stars play in many areas of astronomy, from stellar evolution to galaxy formation. In just the 5 years since LIGO’s first detections were announced, gravitational wave measurements have already left astronomers in awe, with insights into the origin of neutron-rich elements (see Box 2.2), the detection of stellar-mass black holes much more massive than previously known, and new tests of gravity in the strong field regime.

Similar to the electromagnetic spectrum, a variety of astronomical phenomena produce a broad frequency range of gravitational waves (see Figure 2.15). The kilometers-long ground-based detectors LIGO and Virgo are sensitive to signals from relatively small systems: neutron stars and stellar mass black holes with radii from 10 to a few hundred kilometers, and with masses up to several hundreds of solar masses. Colliding massive black holes with millions of solar masses at the centers of galaxies and inspiraling white

dwarfs are thousands to millions of kilometers in extent and need space-based detectors, such as the LISA mission led by ESA with NASA contributions, scheduled to launch in the mid 2030s. To detect the gravitational waves from yet larger objects, like black holes weighing billions of solar masses, requires galaxy-sized baselines. This is done with high-precision timing of the arrival of radio pulses from pulsars distributed throughout our galaxy. The loss of the Arecibo Observatory³ is a setback for pulsar timing experiments and underscores the need for radio facilities that can continue this critical science.

The larger and more sensitive ground-based gravitational wave interferometer network anticipated in the coming decade will detect a large number of stellar-mass black hole mergers and determine their masses and spins. Careful comparison of the properties of these mergers with theoretical models will inform which arise from binary or triple stellar evolution, which arise in dense stellar systems like globular clusters, and which might have entirely different origins. The population of neutron star and black hole mergers with masses of ~2–5 M_sun⁴ in the “mass gap” between neutron stars and black holes will provide key constraints on our understanding of massive stellar evolution, the maximum mass of neutron stars, and core-collapse explosion physics.

Combined gravitational wave and electromagnetic observations have the potential to finally crack the long-standing puzzle of the origin and growth of massive black holes, one that lies at the intersection of the understanding of stars, galaxies, accretion disks, and cosmology. LISA and future ground-based gravitational wave detectors will detect black hole mergers in the earliest phases of galaxy formation less than a billion years after the Big Bang, while LISA, ground-based detectors, and pulsar timing arrays will constrain the merger rate over cosmic time for a range of black hole masses. Combining this gravitational wave data with deeper electromagnetic observations of accretion onto black holes, particularly in the infrared (e.g., JWST) and X-ray, will inform whether massive black holes originate from the remnants of massive stars. Are they built up by stellar and black hole collisions in dense star clusters, or perhaps they formed from the direct collapse of gas clouds in some of the first galaxies? Gravitational wave measurements will also be critical for disentangling the role of mergers and gas accretion in growing black holes over cosmic time.

The simultaneous detection of gravitational waves and light from the binary neutron star merger GW170817 was a transformative event (see Box 2.2). More sensitive ground-based gravitational wave interferometers will provide a much larger sample of binary neutron star and neutron star–black hole mergers. The light from GW170817 was powered by material ejected during the merger and from the accretion disk left behind after the merger. Theoretical models predict that there will be considerable diversity in the electromagnetic counterparts to such events depending on the total mass of the binary neutron star system, the mass ratio, and the equation of state of dense nuclear matter, which sets the maximum mass of a neutron star beyond which it collapses to a black hole. Characterizing this diversity using observations of light and gravitational waves will be critical for unraveling the physics of accretion and jet production in binary mergers and their role in the origin of the heavy elements. Given the larger distance to gravitational wave sources as the facilities become more sensitive, this will require new observational capabilities for electromagnetic follow-up. Large ground-based optical-infrared telescopes for spectroscopy to characterize heavy element production and sensitive ground-based cm-wavelength interferometers and space-based high-energy telescopes to characterize relativistic jets will be particularly important. The science produced by joint gravitational wave and electromagnetic observations is likely to extend to at least a subset of LISA sources—namely, binary black hole mergers in gas-rich galaxies. This will enable detailed studies of host galaxies, reveal the role of gas in facilitating massive black hole mergers in galactic nuclei, and provide a sample of black hole mergers at high redshift suitable for cosmology.

___________________

³ See Section 5.15 for a detailed discussion of the Arecibo Observatory.

⁴ One M_sun is the mass of the Sun.

Part of the power and promise of gravitational wave measurements is their ability to simultaneously enable “new astronomy” and “new physics.” Tests of general relativity (GR) using gravitational wave measurements are in their infancy. In the coming decades, these tests will become far more stringent, with increasingly more precise measurements either cementing the quantitative applicability of GR in the strong field regime, or perhaps revealing new physics. Sensitive ground-based gravitational wave detectors at higher frequencies and louder signals will constrain the radii of neutron stars through their tidal deformation. This, in turn, will constrain the equation of state of nuclear matter better than with current measurements, and in a way that is not possible in laboratories on Earth. Gravitational wave constraints on the neutron star equation of state will complement ongoing electromagnetic efforts using radio and X-ray timing and X-ray spectroscopy, likely leading to high-precision measurements of neutron star radii in the coming decade. Larger samples of binary black hole and neutron star mergers will also significantly increase their utility for cosmology. Gravitational wave detections determine the distances to sources. Simultaneous electromagnetic observations of host galaxies to determine redshift can thus provide a measurement of the Hubble constant that is completely independent of current techniques.

Gravitational wave astronomy started with a bang in 2015, opening a new window to the universe with the detection of merging black holes by LIGO/Virgo. Over the few years since then, gravitational wave observations have become an indispensable astronomical tool. The coming decade, with the potential of detections in other parts of the gravitational wave spectrum, signals from new sources, and large numbers of black hole and neutron star detections, will be the start of a new era of precision and multi-wavelength gravitational wave astronomy.

2.2.5 Astronomical Transient Events

Although the night sky looks placid, millennia of observations have shown that it in fact varies systematically on many time scales. There are secular changes in the positions and speeds of objects due to their motion—be it interstellar interlopers in the solar system (see Section 2.1) or stars orbiting the 4 million solar mass black hole in the center of the Milky Way Galaxy. On top of this, there is a dizzying variety of dynamic astrophysical events that emit large amounts of energy in anywhere from the blink of an eye to time scales longer than human lifetimes. Some of these are cataclysmic events that herald the formation of neutron stars and black holes in stellar core-collapse, the thermonuclear explosions of white dwarfs, or the mergers of stars or compact objects. Others are repeating phenomena, such as stellar flares or the explosions of the surface layers of white dwarfs. All of these phenomena involve astronomical sources that produce light at many wavelengths, as well as in some cases gravitational waves and high-energy neutrinos and cosmic rays. Astronomical transients impact nearly every area of astronomy. Supernovae and neutron star mergers disperse heavy elements into the interstellar medium, seeding gas with the elements necessary for life. Thermonuclear explosions of white dwarfs are valuable “standard candles” used to trace the acceleration of the universe. Fast radio bursts, millisecond bursts of radio emission of uncertain origin, have the potential to become a powerful new probe of the distribution of baryons throughout cosmic ecosystems (see Section 2.3). Dedicated archives of brightness measurements now spanning more than 100 years enable the study of transient phenomena that recur on decade time scales or longer, such as stellar occultations by circumstellar material, the slow but dramatic brightening of newly unveiled young stars and of old stars like Betelgeuse in their death throes (see Section 2.1).

Rapid advances in detector technology and computing power have led to a revolution in astronomical time-domain surveys, which in turn have led to the discovery of new classes of transients (e.g., fast radio bursts and stellar mergers; Figure 2.16 depicts various classes of optical transients). The Rubin Observatory will take this effort to the next stage. Just one night of observation is expected to detect 10 million transient

events with sub-arcsecond positional accuracy, sending triggers to telescopes around the globe and in space, to follow up with observations at other wavelengths and to correlate with observations using other messengers. This revolution is extending to a broader range of wavelengths outside the traditional optical and gamma-ray bands. The Roman satellite will carry out a near-infrared supernova survey that will also likely discover new classes of infrared transients. Roman’s microlensing survey will allow characterization for the first time of the mass function of the majority of neutron stars and black holes in the galaxy. The extended Roentgen Survey with an Imaging Telescope Array (eROSITA) is carrying out the first all-sky X-ray survey since the Roentgen Satellite (ROSAT) in the early 1990s. Additionally, there is a tremendous increase in the number of international radio facilities searching for radio transients—for example, CHIME, the Australian Square Kilometre Array Pathfinder (ASKAP), and MeerKat.

The next decade has the potential to realize the full capabilities of transient and multi-messenger science and, very likely, to discover more surprises along the way. This work will ultimately provide a mapping between transients, the energy sources and central objects (e.g., black holes, neutron stars, stars) that power them, and their broader astronomical consequences (e.g., feedback and nucleosynthesis). Fully realizing this vision will, however, require a wide range of observational, theoretical, and software capabilities. The data required are obtained with simultaneous and coordinated operation of different instruments on Earth and in space. Ensuring easy user access to this wealth of data and the ability to cross-correlate multiple sources of data will maximize the science enabled by existing and planned facilities (see Sections 4.4 and 7.4.1). In addition, NASA’s workhorse hard X-ray and gamma-ray transient facilities (Swift and Fermi, respectively) are aging, and their longevity is uncertain. Higher sensitivity all-sky monitoring of the high-energy sky, complemented by capabilities in the optical such as from Kepler and TESS, is a critical part of

our vision for the next decade in transient and multi-messenger astronomy. Likewise, there are tremendous scientific opportunities for dedicated transient facilities at other wavelengths (e.g., the ultraviolet from space and radio on the ground) and dedicated spectroscopic follow-up facilities, to complement the major U.S. investment in optical and near-infrared imaging surveys.

Priority Science Area: New Windows on the Dynamic Universe

The combination of new multi-messenger probes of astronomical phenomena with the maturation of time-domain observations opens up tremendous discovery spaces across nearly all areas of astrophysics. Within this discovery landscape, driven by improvements in gravitational wave and neutrino detection, and upcoming facilities such as the Rubin Observatory, one priority area stands out: the application of these new tools to the formation, evolution, and nature of compact stellar remnants such as white dwarfs, neutron stars, and black holes, as probed by the gravitational wave signatures of their mergers, together with rare explosive events that can be explored by the unique cadence and multi-color sensitivity of the Rubin Observatory. Sensitive observations of high-energy neutrinos and charged particles add new elements of discovery space, which will probe the universe’s most extreme particle accelerators—New Windows on the Dynamic Universe.

The formation and evolution of compact stellar remnants, signaled by their accompanying multimessenger transient phenomena are now serving to probe the range of neutron star and black hole masses in entirely new ways. These measurements provide information about the nature of matter under the extreme conditions that cannot be replicated in the laboratory and about how the most extreme compact stellar remnants are formed and evolve. The mergers of neutron stars, uniquely observable at very early times through their gravitational wave signatures, can inform how elements such as gold and platinum are produced, which has been a mystery for many decades. Physical conditions near the surfaces and event horizons of neutron stars and black holes also represent extremes of matter, density, and gravitation and serve as unique probes of fundamental physics. The coming revolution in temporal observations of the sky through non-electromagnetic messengers, and through new observational cadences and spectroscopic follow-up across the electromagnetic spectrum, is at the frontier of modern astrophysics.

As with the other priority science areas, progress will require coordinated advances in observations, experiments, and theory. The Vera Rubin Observatory and future large-area high-cadence radio facilities will increase the numbers of variable and transient objects by orders of magnitude, with regular sampling over time. The Rubin Observatory will be unique for an optical time-domain facility, as it will provide multiple optical colors at uniform cadence with unprecedented sensitivity. Advanced algorithms utilizing artificial intelligence algorithms to sift through Rubin Observatory’s massive amounts of data will find the interesting outliers that will be the heart of future progress and discovery. Current facilities such as the Chandra, Swift, and Fermi space observatories and ground-based radio and optical and infrared observatories will play vital roles in such follow-up work. The full exploitation of this potential for multi-wavelength time-domain observations will require maintaining and expanding these observatories. Such facilities need not be large or expensive but need to be optimized for the task, and this survey recommends the establishment of such facilities both for space (including needed replacement of capabilities currently handled by aging facilities) and on the ground. Since this science tends to be global by its very nature, international cooperation both in complementary facilities and data sharing would greatly enhance the scientific outputs from these investments.

The major new facilities recommended by this survey will all play important roles in extending the power of these capabilities in the future. Many of the visible counterparts of these sources (often originating at cosmological distances) are extremely faint, beyond the limits of present-day telescopes, but should be within reach of the next generation of ELTs. The relativistic outflows produced by these events often can be readily

detected in the radio, and will be prime targets for next-generation facilities such as the Next Generation Very Large Array (ngVLA) (recommended for design studies by this survey). Design studies for a next-generation ground-based gravitational wave observatory will set the pathway toward a revolutionary new facility in future decades. X-ray observations—critical to fully understanding the physics of these phenomena—motivate the design and construction of new facilities ranging from the scale of Explorers to a future large mission.

Foundational research is also essential for maximizing progress in this field. Chief among these will be support for theoretical modeling and simulations of these highly relativistic and energetic phenomena, including numerical relativity to determine the nature of the gravitational signatures, plasma physics to understand the particle acceleration, and dynamical modeling to determine the populations that could lead to compact object mergers. Efficiently assimilating the large flows of data from the surveys, extracting the measurements, and interpreting the observations pose major challenges, but with benefits that will expand our astrophysical knowledge in entirely new ways.

2.3 COSMIC ECOSYSTEMS

Processes on a wide range of time and length scales together drive the formation, evolution, and interaction of the remarkable diversity of objects we observe, from exoplanets and stars to black holes and galaxies. A confluence of advances in theory, computational modeling, and observational capabilities expected in the next decade will transform our understanding by identifying the key mechanisms shaping this web of interconnected systems.

2.3.1 Overview

Arguably the single most important lesson in the past ~30 years of understanding the origin of structure in the universe is that it is not a one-way street, dictated solely by gravity from large scales to small. The formation of some of the smallest and densest objects in the universe, stars and massive black holes, dramatically alters how most other astronomical objects form, from planets and galaxies to stars and black holes themselves. Stars and black holes impact their surroundings through a broad set of energetic processes collectively known as feedback. These span an enormous range of time and length scales, from gas as close as the planet-forming disk around a young star to as distant as in another galaxy millions of light-years away.

Many aspects of star and galaxy formation can be viewed as a cosmic tug-of-war between feedback and gravitational collapse, as illustrated schematically in Figure 2.17. It is now known that the luminous bodies of galaxies, far from being disconnected from their surroundings, are part of a vast system that includes their surrounding circumgalactic medium out to intergalactic scales. Theory predicts that giant rivers of gas flow into galaxies, but most of the gas in galaxies is subsequently ejected back out into the circumgalactic medium by powerful galaxy-scale outflows. The flow of matter and energy throughout the entire system is likely responsible for the commonalities and differences among galaxies, but the details of how have been elusive. Likewise, the flow of matter and energy within a galaxy—again due to the combined effects of gravity and feedback—determines the distribution of gas in the interstellar medium and where and how stellar and planetary systems form. The same flows depicted in Figure 2.17 also disburse the heavy elements produced by stellar processes, from the carbon in our bones to the rare-Earth metals in phones.

Understanding the interplay of gravitational and feedback-driven processes is challenging in part because it involves such a wide range of length and time scales. In addition, much like understanding human health requires understanding how cells function, myriad small-scale physical processes regulate the flow of mass and energy illustrated in Figure 2.17, because they determine how gas cools, sheds its angular momentum, and mixes with other gas.

Small-scale physical processes at work in the cosmic ecosystem can thus have a surprisingly large impact on the large-scale behavior of astrophysical systems. An example is ionizing radiation, which is produced by massive stars and black hole accretion disks and can regulate star formation and black hole accretion on sub-parsec scales. Some of this radiation can escape from the dense gaseous environments in which it was produced and propagate out of galaxies into the intergalactic medium. In this way, the first stars and black holes were able to cause a global phase transition over scales of hundreds of Megaparsecs, in which most of the hydrogen in the universe was converted from a neutral to an ionized state, during what is referred to as the “Epoch of Reionization.” Identifying the sources of cosmic reionization, and better understanding how photons escape from their gas-rich and dust-enshrouded sources, will be an important science area over the coming decade. Numerical simulations of the propagation of radiation through galaxies (radiative transport) are highly computationally intensive but critical for gaining a full picture of the role of this key process. Observational studies of gas kinematics, luminosity functions, and chemical compositions of galaxies spanning the Epoch of Reionization are needed, as well as studies of local galaxies that “leak” ionizing radiation, which may help shed light on the process of photon escape.

The symbiosis among cosmological phenomena on such different scales has been recognized for decades. Now, however, a confluence of advances in theory, computational modeling, and new observational capabilities will enable us to identify and understand the actual mechanisms at work in regulating this cosmic ecosystem. New observational probes of these interconnected flows at two widely separated spatial scales are highlighted in two of the discovery areas identified by the science panels: mapping the circumgalactic and intergalactic medium in emission and detecting and characterizing planets as they form.

2.3.2 Stellar and Black Hole Feedback

Stars have densities exceeding the mean density of the universe by more than 30 orders of magnitude; for black holes, the conditions are even more extreme. Despite their small sizes, stars and black holes are the most efficient sources of energy production yet discovered. This efficiency is a consequence of nuclear fusion in stellar interiors and the deep gravitational potential wells produced in stellar core-collapse and tapped by accretion onto black holes. Additional energy can be extracted at the end of a star’s life. There is approximately one core-collapse supernova for every 100 M_sun of stars formed, and this single 10⁵¹-erg supernova explosion can in principle accelerate 1,000 M_sun of gas to speeds sufficient to unbind the gas from most galaxies. For accreting black holes, the energy per unit mass is even higher. Thus, objects occupying a small fraction of the total mass and volume of the universe are critical to the evolution of much of the structure that is observed.

Many aspects of how stars live and die are currently uncertain enough to limit the ability to model and interpret the effects of stellar feedback, be it in the form of radiation, stellar winds, or supernovae. Even in the local universe, for example, there are significant gaps in the understanding of stellar winds. This leads to large uncertainties about which massive stars become neutron stars and which become black holes. Stellar winds are also believed to play an important role in dispersing star-forming molecular clouds and driving turbulence in the interstellar medium, but the efficiency of this feedback again depends on the uncertain strength of winds from massive stars.

Even less is known about how the properties of stellar feedback vary with quantities like stellar metallicity, which strongly affects the efficiency of driving stellar winds but varies from galaxy to galaxy and throughout cosmic time. Much of this uncertainty can be traced to the need for better theoretical models and observational diagnostics of stellar mass loss. More sensitive space-based ultraviolet (UV) spectroscopy is necessary to better characterize the spectra and winds of massive stars. Also, deep visible-wavelength and near-infrared spectroscopy of large numbers of stars in the Milky Way and other nearby galaxies from the ground—“industrial-scale” spectroscopy, one of the discovery areas identified by this survey—would significantly sharpen constraints on stellar models.

Similarly large unknowns about feedback stem from uncertainties in binary stellar evolution. High-mass binary stars evolve differently than single stars, which affects both their radiative output in life and the supernova explosions in which they die. Understanding binary stellar evolution is thus critical for understanding the global energetics of stellar feedback in galaxies. It is likewise important for understanding the spectra of galaxies—in particular, the UV radiation that photoionizes the interstellar medium and likely reionized the universe during the initial epoch of galaxy and star formation. Separately, the dramatic advances in directly seeing the outcomes of binary stellar evolution with transient detection and gravitational wave facilities—for example, stellar mergers and compact object mergers—will continue to provide critical new insights into the life cycle of binary stars (see Section 2.2).

In addition to understanding the energy, mass, and momentum that stars supply to their surroundings, determining how these winds, radiation, and supernovae interact with the surrounding gas on different scales in the interstellar medium is equally important to untangling their interplay. Newly forming low-mass stars produce winds and jets that modify the structure of the clouds in which they form. They also produce radiation that heats and evaporates their surrounding protostellar disks, influencing the conditions for planet formation. Evaporation of planetary atmospheres by the same stellar radiation can also explain a bimodal distribution of planetary radii seen in transit observations (see Section 2.1). The higher-energy radiation (UV and X-ray) from low-mass stars can also drastically alter the chemistry of planetary atmospheres, and thus the habitability of planets, but more precise determinations are needed of how the relevant radiation changes as a function of stellar mass or age.

In regions of high-mass star formation, the radiation and stellar winds produced by massive stars can dominate the dispersal of molecular clouds (Figure 2.18), but observationally diagnosing which processes are the most important in different environments has proved challenging. There are tantalizing observational and theoretical clues suggesting that star-forming clouds with sufficiently high densities are difficult to disrupt by stellar feedback and may form super-star clusters and perhaps globular clusters at high redshift. Studies of reionization era galaxies (e.g., with JWST) and local analogs in the coming decade may finally resolve this long-standing puzzle. Regions of high-mass star formation are often buried behind huge layers of dusty gas, so improved long-wavelength observations (far-infrared, sub-mm, radio) are required to peer through the dust.

Supernovae generally explode too late after stars form to dominate the dynamics within most molecular clouds, but they are a critical source of feedback on galactic scales. The Chandra X-ray observatory led to major progress in the past two decades on understanding supernova feedback, but higher resolution and higher sensitivity X-ray imaging and spectroscopy would enable much more quantitative probes of supernova feedback and its role in powering galactic winds. Supernovae are important for a second, less direct, reason, in that they produce the GeV cosmic rays that dominate the energy of relativistic particles in many galaxies. The impact of cosmic rays is one of the largest uncertainties in understanding feedback in galaxy formation. The primary uncertainty is how cosmic rays are scattered by small-scale fluctuations in the magnetic field, which sets whether cosmic rays can escape a region or whether their pressure builds up to the point where it can drive an outflow. On smaller scales, these cosmic rays can affect the thermal balance and chemistry of molecular clouds and their ability to form stars. It is remarkable that tiny solar system–scale fluctuations in the galactic magnetic field are a key ingredient in understanding how galaxies drive winds on scales of tens of kiloparsecs, or that the large scale magnetic field properties or distant supernovae can affect the formation of pre-stellar cores. This is an area where additional theoretical advances are particularly needed, including advances in plasma simulation techniques.

Feedback from supermassive black holes during galaxy formation and evolution is one of the most dramatic examples of the small scales in the cosmic ecosystem impacting the large (Figure 2.19). Accreting black holes can influence their surroundings through UV and X-ray radiation, collimated relativistic jets, and wider-angle winds. However, the understanding of how active galactic nuclei (AGN) spectra, jets, and winds vary with luminosity, black hole mass, black hole spin, and perhaps other properties is rudimentary. This currently is a major bottleneck in understanding the evolution of galaxies. Theoretical progress in these areas is likely to continue to be rapid, with advances in general relativistic simulations of black hole accretion predicting increasingly realistic jets and winds for comparison to observations. These predictions push the frontiers of radiation theory, plasma theory (to determine how and where the plasma is heated), and computational astrophysics. Observationally, a combination of high-resolution millimeter (mm) imaging and multiwavelength, multi-messenger observations can reveal the launching mechanism and particle content of relativistic jets (electron-proton versus electron-positron) and thus the energetics of jet feedback. More detailed studies of molecular gas emission will reveal how the densest star-forming components of the interstellar medium are impacted by AGN jets. And higher sensitivity and spectral resolution optical-UV and X-ray spectroscopy of broad emission and absorption line outflows from AGN are needed to better diagnose the physical properties of accretion disk winds and determine how much mass and energy they actually carry. In addition to measuring the jet and wind properties, better constraints on the masses and spins of the supermassive black holes themselves will play an important role in understanding how these objects formed and how they grow, as well as their role in the feedback processes described above. X-ray telescopes and next-generation gravitational wave experiments can constrain the black hole spins, and their masses will be better constrained by measurements with next-generation optical and radio telescopes.

Observations of galaxy clusters have revealed the critical role of black hole feedback by jets in the intracluster medium (see Figure 2.19), although exactly how the energy from the black hole couples to the surrounding gas is still uncertain. This is a prime example of the need for multiwavelength observations: the

combination of radio, X-ray, and optical data reveals the interplay between the centimeter-wave-emitting relativistic jets, mm-emitting molecular gas, X-ray emitting thermal intracluster plasma, and optical-emitting photoionized gas. Higher-spectral-resolution X-ray spectroscopy of clusters would sharpen the understanding of AGN feedback in this critical environment that serves as a laboratory for understanding AGN feedback more broadly. Another observational frontier lies in extending studies of the hot intracluster medium to galaxies, probing the transition from galaxies that have largely ceased star formation to actively star forming galaxies. This is likely to transform the understanding of the role of feedback across a wide range of environments by directly observing the impact of feedback on the gaseous halos that contain the fuel for galaxy growth (see Box 2.3). Multiwavelength studies will again be key. UV and optical spectroscopy and mm dust continuum observations probe the cooler multiphase gas. Combining UV absorption and UV emission studies of the CGM would be particularly valuable, as would deeper X-ray imaging and spectroscopy. The Sunyaev-Zel’dovich (SZ) effect is a direct probe of the thermal pressure of ionized gas in galactic halos. Of all the observational diagnostics at our disposal, the SZ effect most directly constrains the energy content of gas in galactic halos produced by the combined effects of gravitational collapse and stellar and black hole feedback. Higher-sensitivity and higher-resolution CMB observations, motivated to a significant extent by cosmology (see Section 2.2), will have a large impact on the understanding of galaxy formation as well.

2.3.3 Multi-Scale Cosmic Flows of Gas

Feedback and gravity are the key ingredients that determine how gas flows across cosmic scales. Directly observing these gas flows is challenging because of the diffuse nature of the gas in galaxy halos and the high spatial resolution required to peer into regions of ongoing star, planet, and massive black hole formation. New theoretical and observational capabilities are, however, allowing this critical aspect of the tug-of-war between gravity and feedback to be tackled.

Within galaxies, star-forming clouds form and disperse on time scales of millions of years. The structures are thus constantly being reshaped by cosmic flows of gas driven by the interplay between gravity and stellar feedback. Within those clouds, a major puzzle is how turbulence and magnetic fields determine the gas flows down to young stars and the planet-forming disks that surround them. Theoretically, this subtle problem requires understanding the degree to which the mostly neutral gas is coupled to the magnetic field—small-scale physics that dramatically impacts the large-scale problem of how disks around young stars form. Observationally, higher-resolution radio and infrared imaging of protostars and their surrounding gaseous environments with ALMA and other instruments are required for progress in this area (Figure 2.20). Images of dust emission from ESA’s Herschel Space Observatory revealed that gaseous filaments are responsible for fueling star formation on the scale of star clusters, but the role of filaments in determining cluster structure and stellar fragmentation is not yet clear. The accretion disks around young stars fed by these gas flows in star-forming cores set the conditions under which planets form.

Studies with Spitzer, Herschel, and the Atacama Large Millimeter/submillimeter Array (ALMA) of local star-forming regions have identified protostars down to low masses and have brought the complex interactions with their surroundings to light. An area of particular focus is the relation between protostars and the dynamics, chemical evolution abundances, and physical conditions in protostellar disks. High-resolution ALMA images of gas and dust have shown that many disks have small-scale substructures such as gaps and rings (see Figure 2.20), which are created either by already formed planets or by gas instabilities that could shape future planet formation. An unexpected result of these studies is that the central regions of the disks are often dust-obscured even at submillimeter wavelengths. Imaging of disks at high spatial resolution by ALMA or JWST can map the distribution within disks of a wide variety of molecules, with lines sampling

an extensive range of density and temperature. These molecules, as interpreted by chemical models, are beginning to sketch out the early chemical evolution of planetary systems as shaped by the young star, but many key molecules, such as water, remain to be observed. A question for the coming decade is to understand the coupling between these small-scale, solar system–forming regions, the larger cloud environment, and the diffuse interstellar medium (ISM). The challenge is that these different scales harbor gas with a wide range of temperatures, phases, and chemical species, which require panchromatic observations from the millimeter and far-infrared through UV. Ground-based spectroscopy in the radio, optical, and near-infrared will be complemented by UV and far-IR spectroscopy from space to map the full cascade of star formation from the diffuse ISM down to individual protostars.

Our understanding of the flow of gas on the much larger distance scales of galaxies is also evolving rapidly. Despite the vast distances between observed galaxies, theoretical models predict that they are connected via the cosmic web of dark matter that forms the backbone for gas flows on cosmological scales (Figure 2.21). Theory predicts that gas flows into galaxies from the circumgalactic medium that fills the dark matter halos surrounding galaxies (see Box 2.3). The circumgalactic medium is in turn fed by flows of gas from the more diffuse intergalactic medium and from gas ejected into the circumgalactic medium in galaxy-scale outflows. Characterizing the structure, metallicity, and dynamics of the circumgalactic and intergalactic medium with optical-UV and X-ray spectroscopy is a major frontier highlighted by the science panels.

During the formation of galaxies, a small fraction (~10^-3) of the gas somehow loses essentially all of its angular momentum to end up in a massive black hole at the center of the galaxy. Exactly how this happens remains a mystery, although theoretical models are beginning to connect the growth of black holes to the properties of gas in the host galaxy on much larger scales (Figure 2.21). Given the energetic importance of black hole feedback, we need to understand how and when black holes grow, to assess their role in galaxy formation. High spatial and spectral resolution molecular gas observations are the key to peering into galactic nuclei, and can reveal how gas accretes and the properties of the black hole itself. One key problem highlighted by two of the science panels is whether massive black holes first form and grow by accretion from seed black holes formed in stellar core-collapse or whether under rare circumstances it is possible for gas clouds to collapse directly to ~10⁴-⁵ M_sun black holes, perhaps via a supermassive star intermediary. Or does an entirely different set of processes seed galactic nuclei with massive black holes at high redshift? A combination of gravitational wave measurements of black hole mergers across cosmic time (see Section 2.2), deep near-infrared (JWST), far-infrared, and X-ray imaging and spectroscopy of high redshift accreting black holes would help piece together the origins story for massive black holes.

The collective result of feedback from star formation and black hole accretion powers galactic winds, galaxy-scale outflows of mass, energy, and heavy elements that have a major impact on the star formation histories and chemical evolution of galaxies and on the gaseous medium surrounding galaxies. The morphological and spectroscopic evidence for galactic winds is overwhelming (e.g., Figure 2.22), as is the heavy element enrichment they produce hundreds of kiloparsecs from star-forming galaxies and even out into the diffuse intergalactic medium.

Massive stars and their supernovae are sources of mass, energy, and momentum, which emerge in the form of fast-moving shock-heated gas, and relativistic cosmic ray particles as well as photons. Radiation, winds, and jets from accreting supermassive black holes are also likely important in powering galaxy-scale outflows in galaxies. It is, however, an open question whether AGN are only important for high-mass galaxies or have a significant impact in low-mass galaxies as well. There are major uncertainties about how the ingredients of stellar and AGN feedback combine to produce the outflows that we observe. Perhaps the biggest puzzle lies in understanding the subtle problem of the co-existence of gas at very different temperatures and densities in the outflows. Galactic winds are observed to be multiphase, with molecular,

atomic, and warm and hot ionized gas apparently coexisting in the same outflow, but at different velocities. This is revealed by multi-wavelength images and spectra ranging from the millimeter to the X-ray: the wide range of physical conditions requires a panchromatic observational approach. It is unclear how much of the colder gas is launched from the galaxy and how much forms in situ in the outflow. Recent far-infrared dust polarization maps showing smooth vertical magnetic fields over large portions of galaxy disks argue that the cold gas in which this field is embedded is influenced by the wind; future infrared data will test these ideas and provide direct information on feedback processes in and around the cold ISM. Theoretically, it is likely that small-scale physical processes such as instabilities and thermal conduction regulate the transfer of gas between different phases. Although they are small in scale, these processes are in fact critical because they determine how the wind material cools and thus how much mass and energy can actually be ejected from a galaxy. A better understanding of how mass moves between phases is also necessary to quantitatively interpret the wealth of multi-wavelength data on galactic winds and connect those observations to the physical quantities of most interest, such as the mass and energy that winds carry. Some of the most pressing questions, and greatest opportunities, are on the largest scales, as outlined in Box 2.3.

Priority Science Area: Unveiling the Hidden Drivers of Galaxy Growth

By its very nature, the Cosmic Ecosystems theme is very rich, exploring interconnected processes ranging over a billionfold range of linear scales, from individual supernovae to the virial radii of galaxies and beyond. Among this abundance of science objectives, the survey has chosen as its priority science area Unveiling the Hidden Drivers of Galaxy Growth. The pathway toward achieving this goal is bracketed by two sets of transformational observational opportunities: first, the revolutionary measurements of the cosmic history of galaxy growth to come very soon from JWST, and second, culminating with the revolutionary capabilities for ground-based and space-based imaging and spectroscopy from the 24–40 m ELTs and a large infrared/optical/ultraviolet (IR/O/UV) space observatory, respectively.

An essential milestone on the pathway to revealing these drivers is a complete understanding of the formation and buildup of galaxies and their structures, stellar populations, metals, and central black holes over cosmic time. JWST is poised to revolutionize this subject by imaging and identifying galaxies out to some of the first generations during the epoch of reionization, and by obtaining deep redshifted ultraviolet–visible spectra for thousands of a galaxies spanning the period from reionization to the present. When combined with observations from ALMA, the Northern Extended Millimeter Array (NOEMA), and the VLA, it should be possible to trace the transformation of baryons from interstellar molecules to stars and metals within a self-consistent framework. This information will be complemented by wide-field galaxy surveys from the Vera Rubin Observatory, the Spectro-Photometer for the History of the Universe, Epoch of Reionization, and Ices Explorer (SPHEREx) Explorer mission, the ESA Euclid mission, and the Roman Space Telescope, which will provide large statistical inventories of galaxies and rich target lists for future massively multiplexed spectroscopic surveys on the ground. Observations of nearby galaxies and our galaxy with JWST and the Roman Space Telescope will also provide new insights into the physical processes driving star formation and the feedback processes powering the ecosystem.

Beyond that, the key missing link in unveiling the physics driving galaxy growth is to measure the properties of the diffuse gas within, surrounding, and between galaxies. These are the sites where baryons accrete on to galaxies, star formation is triggered, central black holes accrete and grow, and the feedback processes regulating galaxy growth are manifested. The key observational probes of all of these processes are emission and absorption-line spectroscopy of the diffuse gas, which contains a wealth of diagnostic information on the physical conditions, compositions, and dynamics of the gas. Current telescopes, whether located in space (HST) or on the ground (6–10 m apertures) are capable of probing only the densest regions in emission and occasional single sightlines through rare galaxies that happen to be superimposed in front of a bright quasar. These sporadic observations have been sufficient to demonstrate the power of spectroscopy for probing the baryon cycle but far too little to build up a robust physical picture of the processes at work. A large-aperture UV/optical space telescope, however, also envisaged for addressing Pathways to Habitable Worlds, would transform this subject. The combination of 6 m-class aperture and a high-efficiency spectrograph with modern detectors would provide thousands of potential sightlines to nearby galaxies, enabling “tomographic” studies of their circumgalactic and interstellar media, as well as rich new observations of the intergalactic gas clouds along the same lines of sight. With a versatility comparable to that of HST but an aperture comparable to JWST, such an observatory would carry out groundbreaking observations of galaxy growth and address a majority of the science questions identified in the survey overall.

Major inroads to addressing this problem can also be met by the next generation of ELTs. This would include similar absorption-line spectroscopy (but targeted at the redshifted spectra of more distant galaxies observed at earlier cosmic epochs), as well as studies of the evolution of active galactic nuclei and the growth of the corresponding central black holes over cosmic time. The giant-aperture telescope alone will be able to obtain spectra for the faintest galaxies imaged by JWST, including some of the first genera-

tions of objects formed during reionization. The same spectroscopic capabilities, when trained on ancient stars in the halos of the Milky Way and in its nearest companion galaxies, can provide a detailed record of the assembly and chemical enrichment of these galaxies in their formative stages. More generally, the vast majority of faint galaxies revealed by deep JWST images will be too faint for follow-up spectroscopy either with the largest current ground-based telescopes or with JWST itself; the ELTs will play major roles in carrying out deep spectroscopic observations of these targets, probably including some of the most important “Rosetta stones” of the first generations of galaxies, to confirm their redshifts and measure their physical properties. Imaging observations with the ELTs (with adaptive optics [AO]) and with an IR/O/UV space telescope of the resolved stellar populations of nearby galaxies out to the Virgo cluster can extend these fossil records to the full demographic population of galaxies today.

As with our other two priority science areas, these large facilities, although transformational in their potential impact, cannot address these critical scientific questions by themselves. Over the coming decade, major contributions to these investigations will come from multi-wavelength observations with the VLA, ALMA, Rubin Observatory, Euclid, Roman, 3.5–10 m OIR telescopes, Chandra, and toward the end of this period by the ESA Athena mission. Athena’s emphasis on The Hot and Energetic Universe will provide especially unique and powerful constraints on the cosmic feedback cycle, especially from AGNs, and it underscores the critical importance of future X-ray and infrared missions if these scientific questions are to be fully addressed. Last, perhaps as much as in any branch of astrophysics, progress in this field critically rests on a vibrant program of theoretical modeling and numerical simulations of galaxies, over the full dynamic range of scales over which the ecosystems operate.

2.4 SYNTHESIS AND CONNECTIONS

The science themes envisioned for the next decade and beyond reflect an interconnected cosmos. The hierarchies of structure inherent in the universe are dependent on interactions across this wide range of scales. Planet formation and evolution is influenced by the local stellar environment; star formation is affected by galactic environments and strongly influences them through supernovae and other feedback processes; and galaxy formation and evolution depends on the circumgalactic and intergalactic environment and interactions therein as well as on the energy output produced by massive black hole growth in galactic nuclei. Measurements of the positions and properties of individual stars in nearby galaxies enables an extraction of their evolution and life history from the larger forces at play in the galaxy’s structure, formation, and evolution. Disentangling this web requires understanding the complex local interdependencies. The interconnectedness even extends to the power of applying multiple wavelength observations, multiple messengers, and novel observational techniques being applied to vastly different science areas.

The need for observations across the electromagnetic spectrum is common for all of these science themes, as the complexity of these questions requires a multiplexed effort. The interconnected nature of cosmic ecosystems necessitates an observational and theoretical program in which traditional boundaries between disciplines (e.g., radio, X-ray, star formation theory, galaxy formation theory) give way to a more comprehensive approach. Improved multi-wavelength observational capabilities are particularly important for probing the full range of physical conditions, from cold, dusty gas and synchrotron-emitting cosmic rays observed in the radio and infrared, to optical and UV observations of stellar and black hole radiation, to the hot interstellar and circumgalactic medium observed in the UV and X-ray and by the SZ effect on the CMB. The study of radio jets in galaxies informs the mechanical and radiative inputs to galaxy structure and details the large-scale consequence of supermassive black holes at a galaxy’s center, itself amenable to study at high energies, ultraviolet, optical, and infrared wavelengths. Research into aspects of star formation requires a panchromatic approach: deeply embedded sources appear only in the infrared,

while magnetically active young stars emit copious high-energy radiation, and accretion processes appear at far-infrared, optical, ultraviolet, and even high-energy wavelengths. Compact objects can emit at the longest and shortest wavelengths of the electromagnetic spectrum, as well as potentially be a source of other particles like neutrinos or cosmic rays.

The universe is not static, and the time-domain is important to many aspects of astrophysics. Constraints on the local value of the expansion of the universe rely on the identification and use of variable stars as one avenue for measurements, while a different route uses time delays in variability from multiple gravitationally lensed images of quasars to provide the constraint. Most of the detection methods for exoplanets use some means of detecting changes in stellar properties over time to infer the presence of a planet. Many objects in the universe change their intensity with time, in a way that illuminates the astrophysics of the object itself or makes it amenable to study some other phenomenon: witness the monitoring of quasar spectra over time to illuminate the innermost regions near the center of a galaxy around the central supermassive black hole, as well as studies of the lifetime and distribution of starspots through long-term precision light curves. The time-domain also encompasses eruptions and explosions, from nova outbursts of episodic accretion to the merging of two neutron stars, which produce both electromagnetic and gravitational wave signals during the final coalescence, to the high-energy flares that may prevent the development of life on otherwise Earth-like planets orbiting magnetically active stars.

In the same way that answering the science questions of the next decade requires a multi-wavelength and multi-messenger approach, these objectives also require the synergy of space, ground, and even underground facilities. In the gravitational wave arena, space-based detection of inspiraling intermediate mass black hole mergers can signal the need for ground-based gravitational wave observation of the final merger, with searches for possible electromagnetic counterparts. Transients in electromagnetic emission will come by the millions per night from the Rubin Observatory once it is operational, with follow-up in other ground- and space-based telescopes needed for further characterization of the most interesting transient events. Neutrino observatories are opening an entirely new window for probing the most energetic processes in the universe. Exoplanet atmosphere characterization needs the combination of high-precision spectro-photometric measurements obtainable from space with the high spectral resolving power available from large-aperture ground facilities, to probe planets in the habitable zone around low-mass stars.

Theory, simulations, and laboratory measurements are just as crucial as new observations in making headway on these important lines of inquiry. A core challenge is understanding how to model systems in which processes on a wide range of time and length scales interact to produce the universe that we observe. This includes both the need to model how the smallest objects (stars, supernovae, and black hole accretion disks) interact with their environment to understand the universe on large scales (e.g., galaxies and the circumgalactic medium), as well as the need to include small-scale physical processes (e.g., dust-gas instabilities in protostellar disks) to understand astronomical systems (planet formation, in this example). Theoretical mechanisms can explain the variety of transients observed and anticipated in the coming decade. General relativistic simulations of black hole accretion advance state-of-the-art predictions of the behavior of jets and winds from these compact objects, which can be compared with observations. From theory and simulations, a more complete knowledge of stars—as far as their rotation, binarity, and the impact of magnetic fields—improves the ability to model and interpret the expected ionizing output from massive stars. These parameters are also critical for understanding the evolution of massive stars and their feedback effects on galactic ecosystems. Likewise, simulations of convection, rotation, and magnetic field generation illuminate the dynamic nature of stars. Computational models and laboratory experiments, along with observations of gas and dust in planet-forming disks, are needed to understand planet formation in a more holistic manner. Knowledge of atomic and molecular properties gleaned from the laboratory can be essential for fully understanding the microscopic processes that have macroscopic consequences for stars, galaxies, and the cosmos.

TABLE 2.1 Science Panel Questions

TABLE 2.2 Science Panel Discovery Areas