D

Report of the Panel on Galaxies

FRAMEWORK

Galaxies are the basic unit of observable structure on cosmic scales, themselves residing within a hierarchy of groups, clusters, and superclusters and displaying an astonishing diversity of properties. The luminous regions of galaxies are vast, but the dark spaces between them are much vaster still. We now understand that this apparent emptiness is partly an illusion. The dark matter halos of galaxies extend to great distances and meld into a filamentary cosmic web. The dark matter structures are permeated by a tenuous circumgalactic, intracluster, and intergalactic medium, made of primordial hydrogen and helium that may eventually join galaxies and form into stars, as well as of enriched gas that carries the products of previous stellar generations back into intergalactic space.

The goal of the field of galaxy formation and evolution is to achieve a predictive formulation of the assembly histories of galaxies and their dark matter halos, together with the evolution of their stellar populations, black holes, physical structures, chemical content, and circumgalactic, intracluster, and intergalactic media. Observations characterize both the common trends and the diversity of galaxy properties and their evolution with time. Theoretical models aim to explain these observations with a priori physics while providing predictions that can be tested with new data.

Over the past two decades, enormous progress has been made in linking galaxies and larger baryonic structures to their dark matter halos and in understanding the processes responsible for that link. We know that the luminous bodies of galaxies are part of an interconnected ecosystem that includes their surrounding medium out to intergalactic scales. The flow of matter and energy throughout the entire ecosystem is likely responsible for both the diversity and the regularity of galaxies. Stars and black holes, the prime engines of the matter and energy flow, are believed to have powered the major phase transition—cosmic reionization—that the universe underwent in the relatively short period between a half and one billion years after the Big Bang.

These newly established paradigms point to critical, and addressable, paths forward. We must observe and understand the sources that caused cosmic reionization, and we must isolate the individual physical processes that drive the evolution of the ecosystem and govern the connection between gas, stars, black holes, galaxies, and their dark matter halos. These challenges can be addressed only with coupled advance-

ments in multiwavelength and multiscale observations, models, and simulations, while leaving room for the serendipity that has often led to major discoveries and leaps in understanding. The four science questions formulated in this report, while not exhaustive of this rich field, are the most compelling to address in the 2020s and beyond. These questions are expected to experience major advances over the next decade thanks to upcoming powerful space and ground facilities, and to the increasing sophistication of simulations and models; but they also require new, future capabilities. The discovery area highlights a new observational approach that is becoming technologically feasible for the first time and will produce a more complete picture of the baryons in the universe.

Question D-Q1. How did the intergalactic medium and the first sources of radiation evolve from cosmic dawn through the epoch of reionization?

Question D-Q2. How do gas, metals, and dust flow into, through, and out of galaxies?

Question D-Q3. How do supermassive black holes form, and how is their growth coupled to the evolution of their host galaxies?

Question D-Q4. How do the histories of galaxies and their dark matter halos shape their observable properties?

Discovery Area: Mapping the circumgalactic medium and the intergalactic medium in emission

STATE OF THE FIELD

Galaxies are open systems with extensive circulation of energy, gas, and metals between their stellar bodies and the surrounding circumgalactic medium (CGM), and, farther out, the intracluster and intergalactic media (ICM and IGM). The mechanisms that drive this circulation take the catch-all name of “feedback,” which describes the energy, momentum, and matter ejection driven by star formation (e.g., stellar winds, radiation pressure, supernova explosions) and by accreting supermassive black holes (SMBHs). The competition between accretion and these feedback processes regulates the growth of galaxies, ultimately accounting for both their large diversity and common trends. Much effort over the past decade has been devoted to investigating the “what, when, and where” of the regulatory mechanisms of galaxies. A relatively simple model in which a central galaxy’s gas accretion tracks its halo’s dark matter accretion, and the efficiency of conversion to stars depends on the potential well depth, can account for many observed galaxy properties over a wide range of redshift.

The First Structures and Reionization

The first baryonic structures—stars, black holes (BHs), and galaxies—arise within the first 0.5 giga-years (Gyr) of cosmic history (z >9). By z ~6 (age ~1 Gyr), the ultraviolet photons from these systems have reionized intergalactic hydrogen throughout the universe. Theoretical models predict a patchy reionization process in which ionized bubbles gradually expand and overlap, and quasar absorption spectra provide tentative evidence for this inhomogeneous structure at z ~6–7. For galaxies to be the principal sources of reionization, significant fractions of ionizing photons would need to escape low-mass (M_stars <10⁸M_☉), and currently unobservable, galaxies. Rare SMBHs with masses 10⁹M_☉ are already present by z ~6. Growing BHs to such high mass within a Gyr of the Big Bang requires either massive initial seeds (M ~10⁴–10⁶M_☉) produced by exotic physical conditions or highly efficient accretion from stellar mass seeds—or both. Understanding the origin of the early SMBHs and the contribution of radiation from accreting BHs to reionization remain open questions both observationally and theoretically.

The Growth of Galaxies and Black Holes

After the conclusion of reionization, structures appear to grow in a mostly self-regulated equilibrium mode, where the stellar mass, the molecular gas reservoir, and the star formation rate (SFR) of galaxies track each other across cosmic time. The fraction of mass in galaxies that is in cold neutral interstellar gas decreases steadily with time from >80 percent at z >3–4 to ~10 percent today, supporting the direct link between baryonic accretion onto galaxies and stellar mass build-up. However, the formation and evolution of the Hubble sequence of disks, bulges, and spheroids is still poorly understood. Early disk formation appears to be chaotic, followed by “disk settling,” where disk galaxies evolve from morphologically disturbed and clumpy systems with predominantly disordered gas motions to thin ordered disks over the past ~10 Gyr. Theory and observations suggest that secular disk instabilities, gas accretion, feedback, and galactic mergers all contribute to bulge formation, but their relative importance and contribution to the diversity of today’s bulges are still to be established. One striking puzzle is the emergence by z ~4 of massive evolved spheroids with little ongoing star formation. While the stellar mass of central galaxies is tightly correlated with the halo mass, star formation in satellite galaxies appears to become quenched after the stars enter the parent galaxy’s halo, which suppresses their stellar mass growth and reddens their colors relative to isolated systems. The observed correlation between SMBH mass and host galaxy stellar mass, well established for galaxies with M_stars >10¹¹M_☉, hints at coupled evolution, with BH growth regulating stellar mass or vice versa. The integrated rate of SMBH accretion over cosmic time tracks the cosmic SFR with a volume-averaged ratio that has remained broadly constant over the past 10 Gyr. However, beyond z ~2, much of our knowledge of SMBHs is confined to the luminous quasars, leaving a gap in our understanding of galaxy-SMBH coevolution at lower luminosity.

The Ecosystem: Flows in the Circumgalactic, Intracluster, and Intergalactic Media

Decoding the multiple processes that connect galaxy growth, star formation, SMBH accretion, and the CGM has been a central concern over the past decade, with advances on many fronts. Observations reveal ubiquitous outflows from typical star-forming galaxies at z ~2–4 and from rapidly star-forming galaxies at low redshift. It remains unclear, however, how much of the ejected material cycles back into galaxies and how much remains in the halo or escapes to the ICM/IGM. CGM observations—principally X-ray emission and UV absorption—show that hot (T ~10⁶–10⁷ K) and cool (T ~10⁴–10⁵ K) gas coexists in the halos of many galaxies and that the metal content of the CGM is highly inhomogeneous, with some metal-enriched pockets that could exceed the metallicity of the central galaxy’s stars and interstellar medium (ISM). Galactic winds contain co-existing gas phases from T ~10⁷ K plasma down to T ~10 K molecular gas. On larger scales, the deepest integral field unit (IFU) observations and “tomographic” absorption maps toward dense grids of background sources are providing the first hints of the filamentary IGM structures predicted so vividly by simulations.

The Role of Feedback in Powering and Feeding the Ecosystem

Explaining the stellar masses and the mass-metallicity-SFR relation of galaxies requires characteristic mass-loading factors (ratio of outflow rate to SFR) of order unity for Milky Way-like–galaxies with halo mass M_halo ~10¹²M_☉, rising to much higher values of 10–50 for dwarf galaxies in low-mass halos. Stellar feedback, which has been shown to correlate with the surface density of star formation, is considered the dominant mechanism for driving outflows in low- and intermediate-mass galaxies, and, by driving ISM turbulence, it regulates the efficiency of star formation on scales from individual molecular clouds to entire galaxies. For massive galaxies and halos, there is broad consensus that radiation and jets powered

by SMBH accretion play central roles in suppressing cooling from the CGM and quenching star formation, although much is still unclear on how these processes work. Observations of the past decade provide strong evidence for the effects of SMBH feedback on intracluster gas, connecting phenomena more than nine orders of magnitude in spatial scale.

Stellar and SMBH feedback, however, may not be sufficient to account for all observed galaxy properties. Questions exist on the galaxy-CGM-feedback coupling mechanisms, on the origin of transition points (e.g., the maximum baryon conversion efficiency at M_halo ~10¹²M_☉), and on the normalization, scatter, and evolution of galaxy scaling relations. For M_halo >10¹²M_☉, environmental quenching mechanisms that prevent the gas from accreting and/or cooling onto a galaxy and forming stars efficiently may need to be present in order to explain the presence of massive quiescent spheroids at high redshift. In low-mass galaxies, M_halo <10¹⁰M_☉, simulations show that stellar feedback needs to be coupled with the cosmic UV background to inhibit star formation and explain their low gas-to-star conversion efficiency.

Zooming into the Physics: The Milky Way and Nearby Galaxies

Nearby galaxies and the Milky Way offer testing grounds for understanding structure formation and evolution that are unparalleled in resolution and detail and for isolating the dominant physical mechanisms that drive them. Resolving the stellar populations of nearby galaxies and their satellites has enabled reconstructing much of their accretion histories. The ultra-faint dwarf satellites of the Milky Way and M31 probe the low-mass threshold of galaxy formation, with stellar populations that may have formed before reionization. Dynamical measurements of these galaxies show many orders of magnitude range in stellar mass over a narrow range of halo mass, M_halo ~10⁸–10¹⁰M_☉, suggesting that the final stellar masses are sensitive to intersections between star formation histories, feedback processes, dark matter assembly, and, possibly, dark matter physics, in ways that are still not fully understood. Large populations of “ultra-diffuse galaxies,” with low central surface brightness and large sizes, have been discovered and characterized in galaxy groups and clusters. They exhibit remarkable properties, from exceedingly low-velocity dispersions—suggesting little dark matter within their optical radii—to anomalous globular cluster (GC) systems.

Within the Milky Way, Gaia measurements of stellar distances and proper motions, multi-element spectroscopic surveys of hundreds of thousands of stars across all components of the Galaxy, and asteroseismic measurements that have opened entirely new routes to determining stellar ages are driving progress with samples of millions of stars with accurate radial velocities and metal abundances. These chemodynamical measurements provide increasingly strong evidence that many stars in the disk of the Milky Way migrate far from the radius at which they were born. They have also revealed that much of the Milky Way’s inner stellar halo was contributed by a single dwarf galaxy merger in the distant past, and that disk star kinematics are perturbed by the Sagittarius dwarf and other satellites.

The Backbone: Successes and Challenges for Theory and Simulations

Simulations of galaxy formation and evolution have made enormous progress over the past decade, and now routinely produce galaxies with global and structural properties that match many key observations. While these simulations begin with well-defined cosmological initial conditions, small-scale phenomena, including star formation, accreting SMBHs, and their feedback, are not resolved and are therefore implemented via sub-grid models. Despite significant differences in the sub-grid prescriptions, different simulation suites predict similar results for statistical properties, global scaling relations, and, qualitatively, some morphological features. But it remains unclear how the physical processes that occur on the parsec (pc) and sub-pc scales of gas clouds, individual stars, supernovae (SNe), and SMBHs couple across decades in time and space, leading

to the observed properties of galaxies and their surrounding media. There is also no consensus on the minimal or essential set of physical processes that must be included in galaxy evolution models. Furthermore, because the sub-grid models are tuned to match specific statistical galaxy properties (such as the stellar mass function and the mass-metallicity relation), their predictive power is currently limited. A key challenge for the next generation of simulations will be confronting them with observations that they were not tuned to reproduce, such as CGM observations and detailed galaxy morphologies and gas contents over a wide range of redshift.

D-Q1. HOW DID THE INTERGALACTIC MEDIUM AND THE FIRST SOURCES OF RADIATION EVOLVE FROM COSMIC DAWN THROUGH THE EPOCH OF REIONIZATION?

As the first sources formed, the universe emerged from the Dark Ages and became progressively transparent to photons. How this happened is a central question in modern astrophysics research. It is expected that a significant number of long-standing puzzles about the first billion years will be answered in the 2020s. However, the ultimate frontier of the ignition of the first stars and seeds of the first SMBHs will likely remain terra incognita. Beyond the next decade, the goal of the 2030s will be to discover, determine, and interpret the properties of the very first stars, galaxies, and BHs, together with detailed studies of typical Milky Way–progenitor galaxies at z >10. While the topic of reionization is covered in Appendix C as well, this appendix concentrates on the properties of the sources of reionization and the structures that emerge from this epoch.

D-Q1a. Detailed Thermal History of the Intergalactic Medium and the Topology of Reionization

Intergalactic gas remains cold and opaque during the Dark Ages. As soon as the first sources form, light from these sources, as high-energy photons, is expected to convert the cold gas to ionized plasma. The photons inject heat into the IGM and allow light to travel freely through intergalactic space. This reionization process is observed to unfold rapidly and is completed in the first billion years. One of the puzzles is how reionization occurs, including the identification of the dominant sources of ionizing photons during the epoch of reionization (EoR; from first light to z ~6). Because the thermal and ionization histories of the universe are intimately coupled, a deeper understanding can be gained through measurements of the timeline, thermal history, and topology of reionization. These measurements require improved neutral hydrogen line intensity maps (which measure the redshifted HI 21-cm transition) and deep wide-field optical/near-infrared (IR) imaging surveys covering the EoR. The temperature evolution from HI alone will help discriminate among the sources of heating, possibly distinguishing stellar-origin and heavy BH seeds. The topology, or distribution of angular sizes and clustering of the ionized bubbles, will test the nature of the sources of ionizing photons. Cross-correlation of HI maps with high-redshift sources (and their properties) will be a powerful diagnostic of how the heating/ionization of the IGM progressed and also of the nature of the dominant sources. The HI observations need to span tens of square degrees with noise levels of 0.2 mK² at z = 8 and 100 mK² at z = 15. Wide-field imaging surveys (e.g., with the Roman Space Telescope) will observe >10⁵ galaxies out to z ~10–12, down to K_AB ~26, suitable for cross-correlation with the HI signal. Accurate galaxy redshifts will be key for taking full advantage of the correlation signal and will require highly multiplexing near-IR spectrographs on telescopes that are about two orders of magnitude more sensitive than the 10 m class telescopes currently available. This increase in sensitivity can be achieved with a combination of larger size and higher angular resolution.

Targeting highly ionized patches during the reionization epoch provides a unique window into the process of reionization and the nature of ionizing sources. In the vicinity of quasars, intense radiation powered by the central SMBHs forms isolated ionized bubbles embedded in an otherwise mostly neutral IGM. High-resolution

absorption spectroscopy of the most distant quasars enables detailed density and temperature maps both inside and outside the ionizing bubbles around the quasars. Roman will detect ~2,600 z >7 quasars, and ~20 percent of them will be at z >8; Euclid is expected to discover ~150 bright (J_AB ≲ 22) quasi-stellar objects (QSOs) at z ~7–9. Medium spectral resolution (δv <50 km/s) near-IR spectroscopy is required to resolve the Lyman-alpha line in the ionizing bubbles; high-resolution spectroscopy (δv <10 km/s) will resolve transmission spikes that arise in low-density, more transparent patches in the neutral IGM. These require echelle spectrographs to observe objects that are two orders of magnitude fainter than achievable with current telescopes.

D-Q1b. Production of Ionizing Photons and Their Escape into the Intergalactic Medium

We do not know what the dominant sources of reionization are and how their relative importance changes at different stages of the EoR. Current evidence points toward low-mass galaxies at late stages (z ~6–9), although active galactic nuclei (AGNs) may play an important role as well. Establishing the role of low-mass galaxies in the EoR requires determining (1) the number of low-mass galaxies in the early universe, and (2) the amount of ionizing photons they produce that escapes into the IGM (i.e., Lyman continuum, or LyC, escape fractions). The first goal requires measurement of the faint ends of UV galaxy luminosity functions, which may extend into the regime of early GCs or proto-GCs, during the EoR. Pioneering studies have spectroscopically confirmed the redshifts of a few bright galaxies (H_AB ~25–26) at z ~7–9, but hundreds of fainter galaxies at these redshifts must be confirmed. Ultra-deep imaging surveys (with, e.g., the James Webb Space Telescope [JWST]) are needed, assisted by magnification from foreground galaxy clusters. A faint galaxy at z = 8–20 of M_UV = –12 corresponds to K_AB = 35–36.5. Detection and confirmation of such a galaxy requires both near-IR capabilities two orders of magnitude more sensitive than currently available and magnifications ~10–100.

Direct measurements of the LyC escape fractions of galaxies in the reionization era are not possible, owing to absorption by the intervening IGM, but are possible at lower redshifts. Therefore, the production and escape of ionizing photons from sources between first light and z ~6 need to be inferred by comparing the internal properties of the sources to those of galaxies at much lower redshift for which LyC escape fractions can be measured (see Question D-Q4c). This requires measurements of gas kinematics, gas conditions, geometries, and chemical compositions of the sources, as well as maps of gas inflows and outflows. JWST will reach depths of K_AB ~30–32, and therefore will detect bright galaxies to z ~14–16, albeit in small arcminute-size fields. However, most galaxies at z >10 are low-mass and extremely compact (<<1 kpc), and their lines are expected to be narrow (δv <<100 km/s). Therefore, near-IR integral field unit (IFU) spectroscopic capabilities that deliver spatially resolved (~100–200 pc) information at medium spectral resolution (R ~3,000–5,000) for K_AB ~32 are needed for mapping the low-mass galaxies. Even in the presence of lensing, this requires facilities that have at least two orders of magnitude higher sensitivity than the ones currently available combined with an angular resolution capable of resolving ~100 pc at z = 14–16. Measurements of several hundreds of galaxies across the luminosity function, down to Mstars ~10⁵–10⁶M_☉, are needed to understand LyC photon production and escape. This requires large fields of view (FOVs) or multiple pointings, as at K_AB = 30 the density of galaxies is ~30–3,000 per deg² between z = 12 and z = 8. The interpretation of this wealth of data will require detailed models and simulations of the properties of galaxies during the EoR in order to understand the conditions for the escape of LyC photons.

D-Q1c. Properties of the First Stars, Galaxies, and Black Holes

Population III (Pop III) stars are the first stars to form after the Big Bang, perhaps as early as z ~50–60 in the LambdaCDM model, and are expected to form until z ≲6 in isolated regions. Current expectations for Pop III stars are informed by models that include large unknowns such as their initial mass function (IMF), formation mechanisms, evolution, and the environments in which they form and that they impact.

This is an area where synergy with observations is expected to spur theoretical developments of models for the formation and evolution of the first stars, BH seeds, and galaxies.

Pop III stars are likely extremely faint and rare (possibly AB ~35, but more typically AB ~39; sky density ~1 Mpc^–3), and their direct detection will require more sensitive near-IR capabilities than offered by JWST. The possible exception may be rare but bright rapidly accreting supermassive, cool, red supergiant Pop III stars. Lensing may enable detection of Pop III stars that are extremely massive and/or form in dense clusters. Very rare caustic-crossing events may reach fainter Pop III stars. However, detection requires extensive monitoring of numerous lensing clusters to AB ≲ 29 mag. Long-lived low-mass Pop III stars ([Z/H] ~–5), or their direct Pop II descendants ([Z/H] ≳–4), may be observable in the Local Group (V 18). Potential targets for such near-field cosmological studies include extremely metal-poor stars in the Milky Way’s halo and bulge, old metal-poor GCs, and ultra-faint dwarf galaxies (~10³–10⁴M_☉). These observations require high-resolution UV spectra (R ~30,000) to detect weak metal lines, and dedicated surveys on more sensitive telescopes than currently available. Pushing such studies to statistically significant galaxy samples (~10 Mpc distances) requires even larger facilities (see Question D-Q4b).

Pop III may be observed indirectly via pair-instability supernovae (PISN). Recent models of PISN find that some reach AB ~31.5 at z ~20, detectable in JWST imaging. However, PISN are expected to be rare (10^–3–10^–2 yr^–1 arcmin^–2), and JWST may find only ~5–10 at z >15 unless a multiyear-wide survey is implemented. Wide-field surveys (e.g., with Roman) can potentially find up to 1,000 PISN candidates, but at lower redshifts. Last, models predict that PISN leave a distinctive pattern of elemental abundances, potentially observable in the earliest galaxies and in the ultra-metal-poor stars mentioned above. Abundance measurements may even shine light on the IMF of Pop III stars.

Gamma-ray bursts (GRBs) probe both early star formation, via their measured rates, and the metal enrichment of the IGM, by acting as background light sources for absorption line spectroscopy. GRBs require an alert system and prompt follow-up to identify the ones most likely to be at high redshift, to allow imaging and spectroscopy of the most interesting cases. JWST is limited in its ability to pursue rapid follow-up, although it is a powerful option to determine host redshifts. The path forward requires more agile and high-sensitivity facilities with near-IR imaging and spectroscopy.

D-Q2. HOW DO GAS, METALS, AND DUST FLOW INTO, THROUGH, AND OUT OF GALAXIES?

Pristine gas flows from the intergalactic environment surrounding galaxies, through the galactic halo, to fuel the growth of galaxies. The enriched gas is then returned by galactic winds back into the surrounding diffuse gaseous environment, which can then be accreted back onto the galaxies. The hot (10⁷–10⁸ K) gas in low-redshift groups and clusters and high-redshift massive clusters has been mapped in emission by X-ray telescopes, with the high angular resolution of Chandra revealing the intricate structures of cool fronts, internal shocks, and buoyant bubbles. However, our empirical constraints on the cooler phases (<10⁶ K) of the diffuse gas between and beyond galaxies come from absorption-line observations, which probe sparsely distributed individual sightlines, or from stacking and ensemble averages of observations. Given the complexity and multiphase nature of these media, progress in understanding the physical processes that shape the evolution of galaxies and their larger ecosystems hinges on securing multiwavelength maps of gas kinematics, chemical compositions, density, ionization, and thermal structures of individual galaxies, galaxy clusters, and their host halos across cosmic time.

D-Q2a. The Acquisition of the Gas Necessary to Fuel Star Formation

As galaxies grow, their reservoirs of fuel for star formation are expected to be replenished. However, little direct observational evidence is available for baryonic accretion at any mass scale or redshift, in

part because the accreting gas is expected to be kinematically quiet and is often confused with outflow signatures. Further progress in establishing the gas accretion history of galaxies requires spatially resolved imaging spectroscopy of both the diffuse ionized gas and the neutral atomic/molecular gas in galaxy halos. Because of its low column density, diffuse gas has been primarily traced through absorption in the UV and X-ray, although more recently emission line measurements of dense, metal-enriched gas as it cools have been possible in the UV/optical. Typical gaseous streams span several hundreds of kpc with velocity dispersion δv <<100 km/s, which sensitive imaging spectrographs would need to match down to sensitivities ~10^–20 erg s^–1 cm^–2 arcsec^–2 on spatial scales <1 kpc, to track accretion from halo to disk based on the gas kinematics. This sensitivity can be reached only over a small area with current 10 m class telescopes by allowing dedicated week-long integrations; thus, larger facilities are required to map halo-size areas. New sub-grid models calibrated from small-scale observations will need to be developed to provide genuine simulation predictions for all large-scale observable properties of diffuse gas. Such simulations will have sufficient overlap in spatial and temporal scales and in modeled physics with simulations of star formation and ISM to establish self-consistent physical models for different astrophysical phenomena over vast dynamical scales.

Within the disk, the gas is transformed into stars. Observations of a small number of high-redshift massive galaxies have revealed that the cosmic molecular gas content appears to broadly trace the SFR history, while the neutral atomic gas, a precursor of molecules, appears to have evolved more slowly. High-resolution imaging with the Atacama Large Millimeter/submillimeter Array (ALMA) of far-IR fine structure lines like [CII] is beginning to quantify the ISM in star-forming galaxies, but detailed studies of the cold molecular gas in typical, Milky Way–like, or lower-mass, galaxies beyond z ~2 require radio-to-mm observations at sub-kpc resolution and δv ~10–30 km/s in order to resolve and characterize the properties and kinematics of molecular gas clumps with masses ~ a few × 10⁸M_☉. The sensitivities at the required spectral and spatial resolution are at least an order of magnitude beyond those of present-day instruments like the Very Large Array (VLA) and ALMA.

D-Q2b. The Production, Distribution, and Cycling of Metals

Heavy elements are synthesized in stars and found in the low-density CGM/IGM far from star-forming regions since early cosmic epochs. However, a complete census of heavy elements in different environments and a robust understanding of the associated distribution mechanisms are both lacking. The presence of heavy elements is expected to alter both thermal and chemical states of the gas. Not only does the cooling efficiency depend sensitively on the gas metallicity, but the formation of molecules and dust grains also correlates strongly with gas metallicity. Tracking heavy elements from their production in stars, release to the ISM, and escape into the CGM/ICM/IGM provides a complete accounting of these elements, enables identification of dominant enrichment sources, determines the extent of feedback, and constrains the thermal properties of the gas.

Metallicity provides a quantitative measure of the enrichment level of different gas reservoirs. Metallicity measurements of ionized hydrogen (HII) regions require understanding of the excitation and ionization mechanisms and observations of weak, narrow (δv <100 km/s) diagnostic lines that appear primarily in the rest-frame UV and optical. Sensitive rest-frame UV absorption spectroscopy with δv <10 km/s provides a powerful probe of the metal content and homogeneity in the warm (10⁴–10⁵ K) diffuse CGM/ICM/IGM through observations of ionic absorption features across all redshifts; reaching UV_AB ~23 for background QSOs or bright galaxies would expand from single to multiple lines of sight through an individual halo. High-resolution (Ε/δΕ >1,000) spectroscopy of X-ray emitting gas would provide the necessary constraints for the elemental abundances and abundance patterns in the hot CGM/ICM. Last, wide-field IFUs in the

UV and X-ray would enable direct mapping of diffuse metal line emission, as described in the Discovery Area section below, and constrain the patchiness of metal mixing in warm phase and hot plasma. Together, these observations, which are beyond what is achievable with current and planned near-term facilities, would enable us to understand how heavy elements are dispersed, ejected, mixed, and redistributed during the lifetime of a galaxy.

D-Q2c. The Coupling of Small-Scale Energetic Feedback Processes to the Larger Gaseous Reservoir

A central puzzle of galaxy evolution is why the measured star formation efficiencies are low, around 10 percent or less, across the entire history of galaxies, even at the peak of their growth (z ~1.5–4.0) when their gas mass fractions are high, with M_gas/M_stars ~3–4. Furthermore, it is unclear why the efficiency depends on the halo mass, with the peak of conversion of baryons to stars in the central galaxy occurring at M_halo ~10¹²M_☉. Cosmological simulations require feedback from stellar winds/supernovae and from SMBHs at the low and high galaxy mass ends, respectively, to regulate the baryonic accretion onto dark matter halos and reduce star formation efficiency. Models predict that, depending on the halo mass, a catastrophic loss of cool ISM from the galaxy’s disk can occur, after which the galaxy remains quiescent until a new gas supply is accreted. At present, however, these effects of feedback have not been directly observed in situ.

While evidence for star formation–driven feedback is widespread, the principal physical mechanisms that drive the multiphase galactic winds are not understood. Establishing how small-scale energetic feedback processes are coupled to the larger gaseous reservoir holds the key for distinguishing between different energy and momentum injection processes, including thermal and nonthermal components such as magnetic fields and cosmic rays, in galaxies of different masses and redshifts. It also clarifies when and how the large-scale environment can become hostile to accretion and/or cooling of gas onto a galaxy. Spatially resolved observations of ionized gas within galaxies complement those of the cold neutral and molecular ISM (see Question D-Q2a) and enable a deeper understanding of the formation and survival of dust and molecules. High-resolution, sensitive UV/optical/near-IR spectral maps of galaxies, resolving HII region scales out to z = 10 down to line sensitivities of 10^–19–10^–20 erg s^–1 cm^–2, coupled with spatially resolved sub-kpc X-ray, UV, and optical spectral maps of the CGM will establish gas kinematics and chemical imprints from the ISM to the CGM/ICM/IGM and connect small-scale feedback to gas properties on scales out to and beyond the virial radius (~50–300 kpc, depending on redshift). This is beyond what is achievable with current and planned near-term facilities. The interpretation of these observations will require developments in theory and simulations aimed at understanding the physics of feedback and the role of magnetic fields and cosmic rays across multiple scales.

D-Q2d. The Physical Conditions of the Circumgalactic Medium

The CGM lies at the interface between infall and outflows, making it uniquely sensitive to the physics of baryonic flows. In particular, feedback affects the physical state of all baryons within a galaxy’s sphere of influence and ensures that a large fraction of both gas and metals associated with the dark matter halos remain in the CGM and local IGM. Theory suggests that, independent of redshift, hot atmospheres develop only in halos ≳10¹²M_☉, the expected mass threshold beyond which the supply of cool gas to the central regions of the halo is curtailed, slowing the rate of star formation. At higher redshifts, the physical conditions in the CGM record the concurrent global effects of the resultant feedback processes and predict future accretion activity. At low redshifts, the physical conditions in the CGM provide a record of the galaxy’s history of feedback.

The multiphase nature of the diffuse CGM/ICM and its complex dynamics, where all mechanisms are superposed, necessitate a multiwavelength and multiscale approach that includes imaging and spectroscopic studies of individual galaxy halos and clusters over the full spectral range from X-ray to radio. The important scales to probe span from large-scale halo environments of ~100 kpc down to star-forming clouds of ~100 pc, while the dynamic range in the gas density and temperature is significantly larger. Current instrumentation does not provide the necessary combination of high spatial resolution, wide field, wavelength coverage, and surface brightness sensitivity to enable such studies. In general, spectroscopic capabilities for resolving narrow (δv <100 km/s) kinematic features across the full spectral range are necessary. Wide-field, high-spatial-resolution IFUs with sensitivities more than 10 times better in X-ray, UV, optical, and near-IR than currently available are required for imaging faint, diffuse emission and resolving dense clumps over areas >300 kpc around Milky Way–type galaxies. High-throughput UV-optical spectrographs would enable absorption spectroscopy using QSO and galaxy background sources (see Question D-Q2b), sampling the diffuse CGM/ICM with a density of ~25 arcmin^–2, or an average grid spacing of ~100 physical kpc. Together, these capabilities would enable observations that constrain the density, ionization, metallicity, and velocity field of the CGM/ICM.

D-Q3. HOW DO SUPERMASSIVE BLACK HOLES FORM, AND HOW IS THEIR GROWTH COUPLED TO THE EVOLUTION OF THEIR HOST GALAXIES?

Investigations of SMBHs over the past ~20 years have underscored their importance for galaxy evolution but also the significant gaps that persist in our understanding of these objects. We still do not know how SMBHs form and grow, how they interact with and impact their host galaxy and the CGM/IGM, the full range of BH properties (e.g., the shape, form, and evolution of the BH mass function), and what role early BHs play in reionizing the universe. The coming decade promises answers to many of these questions.

D-Q3a. The Seeds of Supermassive Black Holes

The existence of luminous quasars at z >7 requires SMBHs to grow to M ~10⁹M_☉ in the challengingly short period, <1 Gyr, available since the Big Bang. Theorists have pursued a variety of ideas, broadly distinguished between “heavy seed” and “light seed” models. Examples of the former include the runaway collapse of early, ultra-dense stellar clusters, and the direct collapse of a primordial gas cloud into a >1,000 M_☉ BH, potentially as massive as 10⁵M_☉. Alternatively, distant quasars could grow from light seed BHs, such as those formed from the death of massive stars, either through suppression of feedback that modulates inflowing gas accretion rates (i.e., super-Eddington accretion) at early cosmic epochs, or through rapid merging of stellar-mass BHs accompanying hierarchical structure formation at early times. Besides probing the origins of the most distant BHs, understanding the birth of BHs will inform us about early heating of the IGM, and teach us about a potential major source of feedback in primordial and low-mass galaxies.

Both heavy and light seed models have theoretical challenges, and observations will be required to discriminate among them. Proposed observational tests include (1) measuring the high-redshift (z >6) quasar luminosity function (see Question D-Q1a for numbers and depth); (2) studying the occupation fraction of massive BHs in nearby low-mass galaxies; (3) detecting BH mergers down to 10³M_☉ at z ~10 and ~ a few × 10³M_☉ at z ~20 using the Laser Interferometer Space Antenna (LISA); and (4) detecting high-redshift, massive seeds in the X-rays. An actively accreting 10,000 M_☉ BH at z ~10 requires arcsecond or better X ray spatial resolution to avoid source confusion and enable unique host galaxy identifications, and will need to have that resolution at sensitivities <10^–19 erg s^–1 cm^–2 and over solid angles > 1 deg², the combination of which are far beyond Chandra’s capabilities.

D-Q3b. Existence and Formation of Intermediate-Mass Black Holes

There must be IMBHs in the gap between stellar origin BHs (≲100 M_☉) and SMBHs (>10⁵M_☉), as all viable paths to make SMBHs require a stage of “intermediate” mass. While IMBHs at the high end of the stellar mass range are now beginning to be observed, no BHs have yet been confirmed in the mass range 10³–10⁵M_☉ that bridges the stellar mass regime with the BHs in the center of spheroids. Finding a population of IMBHs would be transformative, providing an evolutionary link in the growth of SMBHs and constraining formation channels of seed BHs in the early universe. The next decade will be ripe for discovery of IMBHs. Time-domain surveys at X-ray, UV, and optical wavelengths will identify rare white dwarf tidal disruption events (TDEs), which probe BH masses <10⁵M_☉. Advanced Laser Interferometer Gravitational-Wave Observatory (LIGO) may detect IMBHs with hundreds of solar masses, and LISA is expected to find merging 10³–10⁵M_☉ BHs at 1 < z < 20. IMBHs in local dwarf galaxies will be probed using several techniques, including LISA for detection of mergers with IMBHs and sensitive, high-angular-resolution UV, optical, and near-IR telescopes for resolving the gravitational sphere of influence of IMBHs (0.01” [arcsec] for a 10⁴M_☉ BH at 5 Mpc; a few arcsec for Milky Way GCs) for both kinematic and integrated light studies. Measurements of integrated light profiles and proper motions will require tens of μ-arcsec relative accuracy, with samples of hundreds of stars within the inner arcsecond of GCs for proper motions. Next-generation radio interferometers and X-ray observatories may detect radiative signatures of IMBH accretion. For X rays, high source sensitivity (<10^–19 erg s^–1 cm^–2) at sub-arcsecond angular resolution in the ~0.5–10 keV band and, in the radio, sensitive imaging capable of resolving a few pc is needed to, for example, prove the existence of and locate IMBHs within local galaxies. Regardless of technique, key measurements include constraining the BH mass spectrum across its full range and measuring the fraction of halos harboring BHs as a function of mass. Such measurements will need to be coupled with advances in theoretical modeling of the underlying BH population and their hosts.

D-Q3c. Comprehensive Census of Supermassive Black Hole Growth

Recent National Aeronautics And Space Administration (NASA) missions have made great progress in understanding the demographics of BH growth. The cosmic X-ray background, which is dominated by accreting BHs, is ~90 percent resolved into discrete sources by the deepest Chandra and X-ray Multi-Mirror (XMM)-Newton surveys at <6 keV, although this fraction falls at higher energies. At 8–24 keV, which overlaps with the 20–40 keV peak of the cosmic X-ray background, only ~35 percent of the background is resolved by the deepest Nuclear Spectroscopic Telescope Array (NuSTAR) surveys. While a large population of obscured AGNs exists, how this population depends on redshift, luminosity, source of the obscuring material (e.g., torus versus galactic dust and gas), and environment remain open questions. Mid-IR missions—for example, Spitzer and Wide-field Infrared Survey Explorer (WISE)—have detected obscured sources in large numbers, although they are biased to the high-luminosity AGNs, where accretion luminosity dominates over stellar emission. The cosmic census of AGNs is currently patchy, which limits our understanding of the co-evolution of galaxies and their SMBHs. The highest redshifts remain largely unprobed, our knowledge of the most heavily obscured AGNs is incomplete, even at the lowest redshifts, and nuclear activity in the lowest-mass galaxies is poorly constrained. We need to fill these fundamental holes in our knowledge of AGN demographics in order to understand the mechanisms and importance of BH feedback in galaxy evolution and its interplay with star formation and star formation feedback and address deep questions about the actual physical structure of AGNs.

JWST will probe obscured AGNs by studying polycyclic aromatic hydrocarbons (PAHs; rest-frame 3–9 μm), ionized neon (12–16 μm), and silicate absorption (~10 and 18 μm) out to z >5 for the bluest features.

New, sensitive mid-IR/far-IR spectroscopic capabilities will be required for the longer-wavelength diagnostic features beyond z ~2. BH growth, particularly obscured BH growth, is likely enhanced during BH mergers, and our understanding of these mergers will grow with gravitational wave measurements by LISA and ground-based pulsar timing arrays, as well as with extremely high-resolution imaging by very-long-baseline interferometry (VLBI) and future optical/near-IR facilities. Because BHs are multi-messenger sources, progress will be enabled by many facilities, including deep X-ray surveys (e.g., Athena and hard X-ray, 10–30 keV, imaging surveys more sensitive than NuSTAR), deep optical/infrared surveys (e.g., Rubin Observatory, Euclid, and Roman), mid-IR observations (e.g., JWST), and time-domain surveys (e.g., Rubin Observatory).

D-Q3d. The Physics of Black Hole Feedback

The energy released by accreting SMBHs contributes feedback that helps regulate the growth of galaxies, although the magnitude and importance of that feedback is currently highly uncertain and likely varies for galaxies of different masses, environments, and evolutionary stage. Massive galaxies, groups, and clusters, which host the most massive BHs, provide a particularly promising way to distinguish proposed physical mechanisms (e.g., thermal, radiative, and/or magnetic processes) that accelerate outflows, because they allow study not only of the BH and its outflow, but also its impact on the surroundings. High-spectral-resolution X-ray kinematic measurements are a frontier scientific measurement for this question. Hitomi showed that the intracluster plasma of the Perseus cluster is astoundingly calm, with a turbulent gas velocity of ~164 km/s. Some yet-to-be-understood process is suppressing energy pumped into the gas by the nuclear activity at the cluster core and prevents it from inducing turbulence in its surroundings. Regrettably, Hitomi was lost after only a few weeks of observations, so the immense promise of high-resolution nondispersive spectroscopy must now wait for X-ray Imaging and Spectroscopy Mission (XRISM) and Athena. To probe AGN winds across all the relevant ionization states and phases, high-throughput, high-resolution spectroscopy from the hard X rays through the far ultraviolet (FUV) is needed. Ultimately, arcsecond angular resolution in the X ray with much larger throughput than currently available (i.e., >1 m²) is required for imaging—for example, shocks induced by outflows. Deep, spatially resolved infrared and millimeter measurements are required to probe the molecular outflows and outflow dust content. High-sensitivity radio interferometers in the 0.1–115 GHz range that reach below L_{1.4 GHz} ~10²⁴ W/Hz with ≤100 pc resolution would probe and resolve jet-gas interactions out to z ~1 and enable precision studies of the ISM in both molecular (low-J CO in emission) and atomic gas (HI in absorption) to learn how such interactions occur, as well as to study the mechanical effects of BH feedback via study of radio bubbles.

D-Q4. HOW DO THE HISTORIES OF GALAXIES AND THEIR DARK MATTER HALOS SHAPE THEIR OBSERVABLE PROPERTIES?

Observational, semi-empirical, computational, and theoretical studies of the past decade have sharpened our understanding of the relation between galaxies and their host dark matter halos. Many aspects of this picture, and the progress achieved over the past ~10–20 years, have been summarized earlier in the section “State of the Field”; yet, many aspects have not been strongly tested by observations, and the discovery potential is large for many areas in the field of galaxy formation over the coming decade. In particular, a plethora of data will parse our own Milky Way into its elemental constituents, enabling us to understand its physics and how singular or general it is as a system; inroads in the investigation of the lowest-mass galaxies in the local universe will clarify the process of galaxy formation close to its mass threshold; and the characterization of the physical components of low-redshift galaxies will provide the benchmark for describing and understanding the galaxies at higher redshifts, where we do not have the luxury of high spatial resolution.

D-Q4a. The Dynamical and Chemical History of the Milky Way

The Milky Way affords unique insights into the governing processes of galaxy formation: although it is a singular example, we can study it at a level of detail impossible for other galaxies. The goal of observational and theoretical studies of the Milky Way is to understand the assembly, star formation, and chemical enrichment histories of the thin disk, thick disk, bulge, bar, and stellar halo; the origin of the striking bimodality of element abundance ratios across the disk; the importance of gas accretion, radial gas flows, fountains, and outflows through time and at present; the impact of dynamical perturbations on kinematic structure; the baryon content and temperature-density structure of the gaseous halo; and the mass, density profile, shape, and substructure of the dark matter halo. The final data releases from ongoing spectroscopic surveys and from the Rubin Observatory and Roman in the coming decade will greatly advance our knowledge of the structure and substructure of the Milky Way’s stellar components.

A benchmark goal for the 2020s is to increase the number of Milky Way stars observed at medium and high resolution by an order of magnitude (to ~10⁸ and ~10⁷ stars for λ/δλ ~2,000 and λ/δλ >20,000, respectively) relative to current surveys, both by moving from grids of pencil beams to contiguous sky coverage and by reaching fainter spectroscopic targets. Contiguous coverage with multi-element spectroscopy, greater depth, and large numbers will decode the history of the disk and bulge, untangle the structure and merger history of the stellar halo, and measure perturbations to tidal streams that could reveal the impact of dark matter substructure. The thousands of very metal-poor stars ([Fe/H] <–2) secured by these new surveys will probe chemical enrichment during the earliest phases of the Milky Way’s formation. Taking full advantage of these data advances will require corresponding advances in calibration methods for abundance measurements, in statistical methods for interpreting enormous high-dimensional data sets, and in numerical simulations that resolve the intricate details of the formation of galaxies like the Milky Way.

At the same time, better characterization of the gaseous components of the Milky Way is essential for understanding the physics that governs the Milky Way today and for providing a template to interpret observations of other galaxies. Higher-sensitivity X-ray spectroscopy can greatly expand the set of AGN sightlines that can probe hot gas absorption in the halo, provided that the absorption lines can be separated from those of the Milky Way ISM and Local Group galaxies, while improved wide-field X-ray IFUs can provide emission-line detections. Simultaneous measurements of OVII and OVIII in absorption and emission provide strong diagnostics of density and temperature structure. In the Galactic Center, progress is still required to understand the extreme environments of the ~2 pc Circumnuclear Ring and the ~200 pc Central Molecular Zone—their energetics, motions, physical characteristics, and so on—and to provide comparison templates for more distant galaxies. The high dust content of this region requires use of infrared and longer wavelength probes, like those provided by Stratopheric Observatory for Infrared Astronomy (SOFIA) and (sub)mm/radio facilities.

D-Q4b. The Threshold of Galaxy Formation

The comparison between observed numbers of galaxies and predicted numbers of dark matter halos implies that the efficiency of galaxy formation plummets in halos with mass below M_halo ~10⁹M_☉, and observable stellar systems do not form in halos with mass M_halo <10⁸M_☉. The lowest-mass galaxies, M_star ~10³–10⁷M_☉ that inhabit halos close to this threshold are unique crucibles for galaxy formation theory. These are the most dark-matter-dominated and chemically primitive galaxies in the universe, and their shallow gravitational potential wells render their baryonic content sensitive to a variety of feedback processes. Open questions include the number of surviving satellite galaxies residing within halos, the number of destroyed low-mass galaxies that populate the stellar halos of their central galaxy hosts, and whether ultra-faint galaxies are fossils of the reionization era or can form in substantial numbers at lower redshifts. These

systems have been studied in detail only in the Local Group or, for the lowest-mass ultra-faint galaxies that now blur the boundary with GCs, within the inner tens of kpc of the Milky Way’s dark matter halo.

An area ripe for major advances is pushing to greater distances and to different environments. Initial results on the M31 satellite system have revealed intriguing differences in its population relative to that of the Milky Way, and the next decade will see dramatic improvements in our understanding of ultra-faint galaxies beyond the Milky Way. In particular, the Rubin Observatory sensitivity limit over 10 years of operations will reach 2,000 L_☉ galaxies out to 1 Mpc from the Milky Way (~3–4× the virial radius) and galaxies as faint as 200 L_☉ throughout the virial volume of the Milky Way; its lifetime sensitivity will detect classical dwarf galaxies (L_V ~10⁵L_☉) with half-light radii >1 kpc. This will open up the possibility of studying, in detail, ultra-faint galaxies that have never interacted with the Milky Way or M31—galaxies that have never been discovered but must exist if current CDM-based models of galaxy formation are correct—thereby providing important information about the effect of environment, reionization quenching, and the halo-galaxy connection at the lowest possible masses. JWST will enable observations of old main sequence turn-off stars (M_V = +4) to characterize star formation histories at the earliest epochs out to ~3 Mpc. Large fields, like those covered by the Rubin Observatory and Roman, will enable efficient observations of typical Milky Way dwarfs, with high target density for simultaneous observation of nearby stars. Wide-field, high-resolution, multi-object spectroscopy will bring progress with detection of weak and narrow metal lines in the oldest Pop II stars ([Fe/H] <–3) and velocity accuracy of 1 km/s to characterize the dynamics of the lowest mass systems (s ~4–5 km/s) and resolve stellar binary motions.

D-Q4c. Connecting Local Galaxies to High-Redshift Galaxies

Nearby galaxies serve as anchors for our interpretation of the physical and chemical histories of individual galaxies and galaxy populations at earlier epochs, including before, during, and soon after the EoR. Two of the key challenges for interpreting high-redshift observations are determining which galaxies produce and leak LyC photons and calibrating indicators of star formation, metallicity, and dust content.

Production of LyC photons is linked to the characteristics and evolution of the most massive stars, which are found in young, massive and supermassive (M_stars >10⁴M_☉) star clusters. Collecting statistically significant numbers of these relatively rare clusters requires probing crowded regions within galaxies in the local ~50–100 Mpc, using unique UV spectral signatures (e.g., P-Cygni NV and CIV, and broad HeII) to characterize their ionizing stellar content. Establishing the conditions, internal and/or external, under which galaxies leak LyC photons and calibrating the indirect UV and optical diagnostics to be used at z >4 will require mapping of LyC photon leakage from galaxies at 0.1 < z < 3 in sufficiently large numbers to discriminate among different conditions for escape.

Calibrating abundance measurements with UV nebular lines will be a top priority for the interpretation of the chemical build-up of galaxies at z >8 with JWST. Reconciling the metallicity scales of nebular emission lines, neutral gas absorption lines, and stellar photospheric lines will enable the interpretation of the chemical history, transport, and mixing, and the ionization structure of galaxies across cosmic times. These will require tracing the faint UV (HeII, CIII], OIII], SIII], etc.) and optical (HeI, auroral lines) lines within and across HII regions, and measuring abundances and depletion patterns of key elements in the neutral gas and in the photospheres of stars across the full range of metal abundances in nearby galaxies.

Metallicity calibrations and the quantification of LyC production and escape will require various combinations of sensitive, wide-field, high-spatial (~ a few pc at 100 Mpc) and low-to-high spectral (from δv ~500 km/s to δv ~ a few km/s) resolutions, UV and optical IFUs and multi-object spectrographs capable of detecting and characterizing faint photospheric lines and gas emission and absorption in local to medium-redshift galaxies. Recent large optical IFU and wide-field spectroscopic surveys (e.g., Calar Alto Legacy

Integral Field Area (CALIFA), Mapping Nearby Galaxies at Apache Point Observatory (MaNGA), Sydney-Australian-Astronomical-Observatory Multi-object Integral-Field Spectrograph (SAMI)) offer a roadmap for how the approach can be extended to the UV. Progress in the theory and modeling of massive star properties and evolution, including the role of multiplicity, rotation, and so on, will need to accompany observations in order to enable their interpretation.

Quantifying the SFR of high-redshift, dusty galaxies requires use of infrared tracers, such as the [CII] fine structure line. An accurate calibration of this tracer, both in intensity and line shape, as a function of local environment in the Milky Way and nearby galaxies can be accomplished with SOFIA, building on the legacy of Herschel.

D-Q4d. The Evolution of Morphologies, Gas Content, Kinematics, and Chemical Properties of Galaxies

Kinematics, metal abundances, and gas content of galaxies are unique tracers that connect evolving galaxy populations across time and help reveal how galaxies obtain their present-day structures. The past decade has provided us with the first measurements of the kinematics and resolved chemical abundances in galaxies at z = 1–3, showing well-defined trends but also more diversity than in the local universe. Observational capabilities have restricted such studies to M_stars ≳10^9.5M_☉, with limited spatial information. As a result, we have essentially no detailed information about progenitors of typical disk galaxies like the Milky Way at the peak of cosmic star formation rate density (z ~2) or earlier; this is also a major limitation in extrapolating the detailed observations being made in the Milky Way to the full population of similar galaxies at all cosmic epochs. JWST will begin to provide detailed kinematic and metal abundance maps at higher redshifts and lower masses and will reveal how galaxies transition from disordered kinematical and morphological states to more ordered systems (e.g., disks). The gain in sensitivity and wavelength coverage of JWST relative to current facilities will open up the high-redshift regime, but its limited spatial resolution ~0.1” (~0.7 kpc) at z ~4–7 will resolve only the largest, most massive galaxies. In addition, absorption-line-based work for spatially resolved stellar kinematics and metallicities will remain extremely challenging with JWST owing to both sensitivity and resolution limitations. Further progress in this area will require multi-object near-IR spectroscopy with HII-region-size resolution and δv ~50 km/s, with enough sensitivity to map galaxies in the stellar continuum below the knee of the stellar mass function (M_stars ~10¹⁰M_☉) from z ~1 to the end of reionization at z ~6. An important benchmark for testing the paradigm established over the past decade in which bursty star formation gravitationally heats the central regions of the host dark matter halos is to reach M_stars <<10⁹M_☉ systems, where this feedback-induced effect is predicted to be most efficient. Deep ALMA observations of the restframe far-IR [CII] line for large samples of galaxies out to z ~7–8 will secure the full census of the star formation in the dusty progenitors of today’s massive galaxies. Bulk metal abundance measurements in these obscured galaxies will require temperature-insensitive tracers—for example, [OIII] in the far-IR, that cannot be observed from the ground below z ~7.

DISCOVERY AREA: MAPPING THE CIRCUMGALACTIC MEDIUM AND THE INTERGALACTIC MEDIUM IN EMISSION

Imaging the CGM and IGM in emission out to z ~10 and beyond is a major opportunity for the next and the following decade, one that is just now coming into view with new instruments and observing strategies. “Imaging” is intended as contiguous sky coverage, over fields large enough to probe significant volumes around and between galaxies. This implies FOVs (or contiguous mapping) of several arcminutes or more, subtending ≥1 Mpc at z = 2–10. The goal is to detect line emission, from neutral hydrogen to ionized gas and metals across multiple spatial scales, which requires spatially resolved spectroscopy probing

from the kpc-size distribution of diffuse HI down to the ~100 pc sizes of HII regions within galaxies. These observations will address or contribute to addressing the science questions in this appendix, including (1) a full baryon and metal accounting in different gas phases at different redshifts, (2) how galaxies acquire fuel for sustaining star formation, and (3) a high-fidelity image of how energy and momentum from stars and SMBHs are transferred to the low-density CGM/IGM as a function of time (see topics D-Q1a, D-Q2a, D-Q2b, D-Q2c, D-Q2d, D-Q3d, D-Q4a, D-Q4c, and D-Q4d). Thus, it is worth exploring innovative designs of instruments for integral field emission-line mapping that accommodate needs for a range of spatial resolutions. Degree-size, low-angular resolution spectral maps offer complementary information to smaller-field, high-angular resolution spectral maps, by capturing the integrated emission from galaxies too faint to be detected individually and by averaging over cosmic variance, which enables spatial cross-correlations of multiple tracers to constrain physical processes.

We have learned an impressive amount about the IGM and CGM from absorption-line observations, in part because cosmological simulations have proven effective at creating a synthesized picture from the individual sightlines. However, simulation predictions are sensitive to numerical uncertainties and to unconstrained physics, including the geometry, kinematics, enrichment, and physical conditions of galactic winds, the impact of thermal instability, mixing instabilities, metal diffusion, and conduction, and the interactions between outflows and accretion. Emission-line maps across all spatial scales linking the galaxy, through the CGM, to the IGM can test all aspects of these predictions and ultimately provide observations of the cosmic baryon distribution and circulation that rival the level of detail that we currently have only from theory (see Figure D.1). This will place constraints on the physics of the mechanisms that regulate galaxies and larger structures, and drive their evolution.

The restframe UV-optical-IR provides a rich suite of emission lines, including HI and He recombination lines, together with strong metal lines, SiIV, CIV, OVI, and lower ionization metal species, such as OII, OIII, SiIII, and CIII, which typically trace cooler gas. These lines can be accessed from either space or the ground, depending on redshift. Wide-field IFU spectroscopy, with spectral resolution of a few thousand, affording velocity resolution δv ~50 km/s, is required to match typical velocity spreads in halos, and measure gas kinematics. The ISM in galaxies is clumpy over ~100 pc sizes, and metal line emission from shock-heated gas in the CGM and IGM is also expected to be clumpy, requiring ~0.01”–0.1” imaging to resolve, while diffuse HI will benefit from lower angular resolution. JWST will offer sensitive multi-object spectroscopy, but without contiguous spatial coverage, or small-field IFU capabilities. SPHEREx will provide groundbreaking all-sky coverage at 0.7–5.0 μm, but at low resolution, both spectral (R ~100) and spatial (6”–10”). At least an order-of-magnitude increase in both spatial resolution and spatial coverage relative to JWST IFUs are required for CGM/IGM imaging that informs theory. The greatest challenge, however, is sensitivity. Fluorescent Lyɑ emission from optically thick HI (N_HI ~10¹⁸ cm^–2) illuminated by the metagalactic UV background has a predicted line surface brightness of ~10^–20 erg s^–1 cm^–2 arcsec^–2, although this is higher in the vicinity of bright quasars where the UV background is boosted. Simulation predictions for metal-line emission are uncertain but detecting bright features in the outer regions of halos (r ~100 kpc) also requires sensitivity ~10^–20 erg s^–1 cm^–2 arcsec^–2 or better. This is deeper by up to an order of magnitude than what is routinely achievable on 10 m class telescopes. Thus, mapping the intricate morphologies visible in simulations of high-redshift structures will require IFUs on much larger telescopes than are currently available.

Moving to hotter gas, the E/dE ~100–1,000 resolution at <1 keV and 3 arcmin field of the XRISM micro-calorimeter will allow emission mapping of OVII, OVIII, and Fe lines (FeXXV and FeXXVI) in the ICM, and in the gas at the outskirts of local groups and clusters out to z ~0.3. The larger effective area of Athena allows maps of continuum and line emission to the virial boundary and beyond for clusters and groups, while also providing sensitivity to the hot gas halos of individual L^* galaxies. Ultimately, still larger effective areas over large fields of view, of order of tens of arcmin, and high spectral resolution are needed

to reach the shock-heated phases of the diffuse IGM. High spectral resolution (E/dE ~2,000) and 1” angular resolution are necessary for resolving the multiphase structure of the ICM/CGM and the discrete sources that contribute to the cosmic X-ray background.

Sunyaev-Zel’dovich (SZ) distortions of the cosmic microwave background (CMB) provide a complementary way to map ionized gas. The combination of X-ray and SZ maps is a much stronger diagnostic of density and temperature structure than either observable on its own. Next-decade facilities will push the limits to a few times higher angular resolution and higher sensitivity than the Planck satellite, with SZ

detection thresholds of individual halos down to ~10¹⁴M_☉ over large areas of sky, extending to 10¹¹–10¹²M_☉ (~10–100× deeper than current limits) for stacking analyses. These will provide new constraints on cumulative energy injection into the CGM. The combination of kinetic SZ measurements from these CMB experiments with new galaxy redshift surveys (e.g., DESI, SPHEREx, and Euclid) will enable cross-correlation of the CMB signal with the peculiar velocity field of galaxies and provide novel insights into the electron density distribution across many different circumgalactic and cosmological environments. New and more powerful CMB experiments will offer a rich new probe of the cosmic baryon distribution by extending to the denser and hotter phases of the IGM beyond halo virial radii.

Large, contiguous maps of cool gas (≲10⁴ K) complete the picture above. Existing maps of atomic hydrogen and molecular gas mostly probe the ISM of star-forming galaxies. However, observations of galactic winds show that they frequently contain atomic and molecular gas; whether this is entrained from the ISM or cools out of the hot flow is unclear. High-sensitivity observations over large fields could detect neutral hydrogen in the more distant CGM, where Lyman Limit absorption (N_HI >10¹⁷ cm^–2) is frequently observed, and molecular gas if it is present. Direct detection of HI or CO emission from the CGM, at distances of tens or even hundreds of kpc, would be a powerful diagnostic of thermal instability in the CGM and the geometric interleaving of cold gas accretion and hot gas outflows. Before reionization, most neutral hydrogen resided in the IGM rather than in galaxies. Mapping the z >8 era in the redshifted 21 cm line may lie beyond the capabilities realized in the 2020s, but such maps will eventually provide an extraordinary view of intergalactic gas at an epoch when few galaxies existed. The needs and expected progress in this area are discussed in Appendix C.

SUMMARY AND FINAL CONSIDERATIONS

The past two decades have firmly established the LCDM framework that sets the initial conditions for cosmological structure formation and for galaxy formation and evolution within and connected to these large-scale structures. In the coming decade, a suite of powerful facilities with unprecedented capabilities for studying galaxies will begin operations, priming the field of galaxy formation and evolution for a period of major advances.

The four questions and the discovery area outlined in this appendix represent key areas of exploration. The appendix identifies observations, theory investigations, and simulations necessary to either advance our knowledge or fully answer each of those questions in the next decade and to set the stage for the decade that follows. The progress required to address those questions will also greatly advance many other areas of research not explicitly mentioned in this appendix. A common theme throughout the appendix is the need for a synergistic multiwavelength approach and for continuous interaction between observations and theory. Theory and simulations will need to progress toward connecting multiple physical scales across many orders of magnitude in dynamical range: from detailed stellar models to stellar population synthesis, galaxy models, zoom-in galaxy simulations, and large box simulations of the universe. Curation of multimission/facility, cross-referenced data archives, quickly searchable across multiple dimensions and parameters, will also be required for the success of the program outlined in this report. The huge, petabyte-size data sets that will become available will require novel approaches to data handling and new statistical methods for data analysis and interpretation. Data volume represents a true challenge for the next decade, which will require innovative approaches in order to fully realize its promise.

A summary table (Table D.1) showing the flow from the science questions, sub-topics, and the discovery area to the observational and theoretical needs as presented in this report is given below.

TABLE D.1 Future Needs to Address Science Questions, Topics, and the Discovery Area