Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Planets, Star Formation, and the Interstellar Medium Understanding the formation of stars and planets has long been a major astronomical goal, and important progress has been made in the recent past. The most important development is the detection of planets around normal, nearby stars. These planets, with masses ranging from a few tenths to a few times that of Jupiter, were detected by the small wobble they produce in the motion of their parent stars. In addition, the first substellar object the companion to the nearby star GL 229 has been imaged, and its spectral characteristics have been identified as being similar to that of a giant planet. The detection of extrasolar planets represents the culmination of more than two centuries of speculation and observation that began with the hypothesis of Laplace and Kant that a rotating gas cloud would collapse into a flattened disk to form a star and an associated planetary system. Another important development has been the ability to observe directly the protostellar disks out of which planets appear to form. Such observations have shown that the formation of stars and protostellar disks begins in our galaxy's tenuous interstellar medium. Much has also been learned about the structure of the interstellar medium and how it affects and is affected by star formation. The same process of star formation observed in the Milky Way is seen in other galaxies as well and is a crucial factor in their formation and evolution. However, astronomers' knowledge is very fragmentary at present; it is not yet possible to predict how the environment of a galaxy affects its interstellar medium and star formation, how the elements created within stars are dispersed into the interstellar medium, the rate at which stars form, what types of stars will form, the nature of the planetary systems that will result, or whether the planetary systems will be hospitable to life. KEY THEMES The fundamental scientific goal in studies of star formation and the interstellar medium is to understand how stars and planets form and how habitable environments might arise from the evolving interstellar medium in galaxies. This goal can be subdivided into three themes that link the formation of planets like Earth to the formation and evolution of galaxies: · Formation and evolution of planetary systems; · Formation of stars from the interstellar medium; and · Evolution of the interstellar medium in galaxies. 10
PLANETS, STAR FORMATION, AND THE INTERSTELLAR MEDIUM FORMATION AND EVOLUTION OF PLANETARY SYSTEMS 11 The recent discovery of extrasolar planets is enormously exciting to the general public. It is also of great scientific interest, since we now have within our grasp the ability to answer age-old questions about the formation and evolution of planetary systems, the existence of nearby solar systems, and the possibility that there are other habitable planets in the universe. The study of such planets represents an entirely new field of astronomy that will almost certainly experience explosive development in the coming decade. There are two complementary ap- proaches to this subject. One is inferential: researchers can study the process of star formation so as to understand the physical conditions in which planets form as part of the birthing process of stars like the Sun. The second is more immediate and involves searching for planets themselves via a host of techniques that are just now being perfected. These techniques range from the study of perturbations to the velocity, position, or brightness of parent stars, to the detection and spectral analysis of light from the planets themselves. Key Questions About the Formation and Evolution of Planets Important questions about the formation and evolution of planets include the following: Frequency of Occurrence and Physical Characteristics · How many planets are there around the nearest 1,000 stars, and what are their masses and distances from their parent stars? · How do the properties of planetary systems vary with stellar type, age, and metallicity? · How does the incidence of planetary systems depend on stellar multiplicity? · What is the frequency of terrestrial planets? Suitability for Life · What are the atmospheric temperatures, densities, and chemical compositions of extrasolar planets? · How do these conditions depend on a planet's distance from its parent star? · Are possible signatures of a habitable planet, such as H2O, O2, or O3, present? Evolution of Disks and Formation of Planets · What is the initial star/disk mass ratio, and how does this ratio evolve with time? · When and how do planets form in a disk? · How does planet formation depend on the mass of the planet? · How does the formation of planets affect the evolution of the disk? · How does planet formation depend on the number of stars in the system? Evolution and Stability of Planetary Systems · How long are planetary systems stable? · How does the stability of a planetary system depend on the masses of its constituents? · How did the solar system's Kuiper Belt originate, and what is its relationship to planetary debris disks? Recent Progress in Understanding the Formation and Evolution of Planetary Systems The past 2 years have seen the culmination of a decades-long search for planets around normal, nearby stars. Nearly one dozen stars are now known to have planets with masses ranging from a few tenths to a few times the mass of Jupiter and with orbits at distances ranging from a few hundredths of an astronomical unit to a few
2 A NEW SCIENCE STRATEGY FOR SPACE ASTRONOMY AND ASTROPHYSICS astronomical units from their parent stars. They were detected through very precise measurements of the radial velocity or positions of their parent stars. With current techniques, many more planets, some as small as Saturn or even Uranus, will probably be discovered by ground-based telescopes over the next decade. Along with indirect detection of planets, direct detection of light from a planet-like object outside the solar system has been achieved for the first time. The "brown dwarf" or "super-Jupiter," GL 229B, lies about 7 arc sec (~50 AU) from its parent star; its spectral energy distribution implies a temperature of ~1,000 K, and its spectrum in the near infrared is dominated by methane absorption like that seen in the spectra of gas-giant planets in the solar system. The luminosity and temperature of GL 229B, coupled with theoretical models, suggest a mass roughly 50 times that of Jupiter. The formation of planets around normal stars appears to be intimately tied to the formation of the stars themselves. Millimeter interferometers have resolved a few disks around nearby T Tauri stars, revealing gas in Keplerian orbits, but the best angular resolution now attainable with visible and infrared telescopes and millimeter interferometers can barely resolve them. Current understanding of planet-forming disks is limited to crude estimates of their size (less than a few hundred astronomical units), mass (a few tenths of a solar mass), and lifetime (less than a few million years). Future Directions for Understanding the Formation and Evolution of Planetary Systems To achieve an understanding of the formation and evolution of other planetary systems, it is essential to develop a capability for making infrared observations of unprecedented sensitivity, dynamic range, and angular resolution a long-range goal beyond the time horizon of the scientific goals considered by TGSAA. Nonethe- less, development should commence on the technology that will enable both infrared spectroscopy of terrestrial planets within 10 parsecs (pc) of the Sun and mapping of protostellar disks within about 150 pc of the Sun at sufficient resolution to clearly resolve both the inner disk (~0.1-AU resolution at 5 to 10 Em) and the outer disk (~1-AU resolution at 50 to 100 lam). For the coming decade, the most important future directions identified by TGSAA for the study of extrasolar planets are, in priority order, as follows: Performing a census of planetary systems around nearby stars. This sample should include enough stars (~ 1,000) so that the frequency, separations, and masses of planets can be investigated as functions of the most important parameters, specifically stellar mass, multiplicity, metallicity, and age. Precision astrometry is a critical part of the goal of obtaining a census of nearby planets; for the 1,000 nearest stars, observations with an accuracy of 5 to 10 microarc see are required to detect planets with a mass similar to that of Uranus and with an orbit a few astronomical units in radius. The required duration of the census-taking mission depends on the orbital periods of the planets and the complexity of the planetary systems, but a decade (comparable to the orbital period of Jupiter) seems reasonable. 2. Detecting extrasolar terrestrial planets. It is preferable to attempt detection of extrasolar terrestrial planets around nearby stars by astrometry at the 1-microarc see level, so that follow-up observations can be made of the planets and their associated planetary systems. However, observations of more distant planets, by photom- etry or gravitational microlensing, might provide statistical evidence regarding the frequency of occurrence of terrestrial planets. 3. Imaging protostellar disks with 10-milliarc see resolution. In nearby star-forming regions, imaging at 10 milliarc see would reveal features as small as 1 AU and would permit direct observation of the disk-stellar wind interface region, gaps in disks owing to the presence of planets, and hot, young protoplanets. 4. Obtaining images and spectra of gas-giant planets around nearby stars. Crude spectroscopy (R~ 100) would enable characterization of the atmospheres of these objects. Such planets might be detected in reflected starlight or, if they are massive, as self-luminous infrared objects.
PLANETS, STAR FORMATION, AND THE INTERSTELLAR MEDIUM Ground- vs. Space-Based Observations 13 The critical capability for obtaining the desired census of planets is astrometry with a positional accuracy of <10 microarc sec. Although ground-based astrometry will play an important role, particularly in detecting large planets on long-period orbits, space-based observations offer the greatest potential for finding planets of low or moderate mass. Similarly, direct imaging of large planets close to their parent stars could be accomplished by means of a space-based interferometer with a 10- to 20-meter baseline; low-resolution spectroscopy of such planets could be done with a high-dynamic-range coronagraph on a space-based telescope. Ground-based observations will also contribute to attaining these goals. Both radial velocity measurements and astrometry can contribute to obtaining a census of extrasolar planets. The amplitude of the stellar radial velocity induced by a planet depends on the angle of inclination of the orbit, which is unknown; it decreases with the size of the orbit, so that the radial velocity technique is most sensitive to planets close to their parent star. Astrometric observations are complementary to studies of radial velocities: they are independent of the angle of inclination and therefore provide unambiguous measurements of planetary masses, and they are most sensitive to planets far from their parent star. Although space-based observations still appear to offer the greatest accuracy, particularly over wide angles, substantial improvements in ground-based astrometry are possible. Terrestrial planets might also be detectable through gravitational microlensing; surveys of 30 million to 50 million stars toward the galactic bulge should result in the observation of ~100 tensing systems per night, and careful monitoring of these systems might find the small increase in brightness over the nominal microlensing signal induced by a planet in the tensing zone. High-resolution imaging of protostellar disks in the mid- and near- infrared should be possible if the two Keck telescopes are equipped with outrigger elements. Finally, advanced adaptive optics on large ground-based telescopes offer the possibility of obtaining low-resolution spectra of a few tens of planets of Jovian mass around the nearest stars. Theoretical Studies Theory is important for all of space astronomy and astrophysics and will continue to play a leading role in guiding and interpreting observations of extrasolar planets. Theory is particularly important in the study of the stability of planetary systems. The recent detections suggest that planetary systems may be not only ubiquitous, but also remarkably diverse. For example, the planet orbiting 51 Pegasi is as massive as Jupiter but is as close to its star as Mercury is to the Sun. Did it form that close to 51 Pegasi, or is it the product of unstable long-term evolution? Theoretical analyses indicate that the solar system itself is marginally unstable. The long-term dynamical evolution of planetary systems may also ' account for the debris disks observed around 15 to 20% of main sequence A-K stars, and for the 'formation and evolution of the Kuiper Belt in the solar system. ~O ~O FORMATION OF STARS FROM THE INTERSTELLAR MEDIUM Knowledge of how stars form is critical to understanding both the origin and evolution of systems of stars, such as galaxies, and the origin and evolution of planetary systems. Although star formation is an ongoing process in the Milky Way, observing the creation of stars from interstellar material has proved to be a formidable challenge for astronomers. Stars are born deep within the cold, dust-enshrouded cores of dark molecular clouds where the opaque dust renders newly forming stars invisible to even the most powerful optical telescopes on the ground and in space. However, it is possible to penetrate the veil of obscuring dust with observations in the infrared, submillimeter, and millimeter wavelength bands of the spectrum. Unfortunately, Earth's atmosphere is itself opaque over a large portion of this wavelength range, making such long-wavelength observations difficult or impossible from the ground. The direct investigation of star formation is therefore a recent development of astronomical science, which has required both space-based and ground-based observational facilities at the fore- front of technological capabilities. Over the last 20 years, great progress has been made toward understanding how stars form. Major milestones include discovery of the following:
4 A NEW SCIENCE STRATEGY FOR SPACE ASTRONOMY AND ASTROPHYSICS · Giant molecular clouds, the largest, most massive, and coldest objects in our galaxy, and the sites of all present star formation; · Bipolar molecular outflows, very energetic jets of cold molecular gas that appear to be a natural bv-oroduct of the formation of all stars; and lion. -, ~ · Circumstellar disks surrounding young stellar objects, which are believed to be the sites of planet forma Astronomers' current understanding suggests that the gravitational collapse of a dense cloud core is the basic mechanism by which stars are formed, but the most fundamental aspects of how this happens remain unknown. Observers still have not achieved an unambiguous identification of a protostellar object. Key Questions About Star Formation important questions about the formation of stars from the interstellar medium include the following: Core Collapse · What are the properties of protostellar cores in giant molecular clouds? · How do the cores affect the kind of star and planetary system that emerge from the collapse process? · What are the properties of infalling protostellar material over the range of scales from 0.1 to 1,000 AU? · What are the size and mass distributions of protostellar disks, and how are they related to protostellar luminosities and evolutionary states? · How luminous are protostellar objects, and how does their luminosity evolve with time? Role of Clusters . Do neighboring young stars have an important influence on the formation of a protostar? · Do neighboring stars play an important role in determining the spectrum of initial stellar masses or the rate of star formation? · How do the conditions in protoclusters affect the survivability of circumstellar disks? · How do supernovae and spiral density waves influence star formation? Masses of Stars · What factors determine the mass of an individual star? Bipolar Flows · What is the origin of bipolar outflows, and what is their role in star formation? · How do bipolar outflows change the structure of the ambient molecular cloud? Molecular Clouds and the Interstellar Medium . . . . . . . How do molecular clouds evolve to form dense cores? What is the physical and chemical structure of a giant molecular cloud? How do giant molecular clouds evolve, and do they have a well-defined age? What is the mass spectrum of dense cores, and what determines it? How do magnetic fields and turbulence influence the structure and evolution of molecular clouds? How does star formation within a cloud influence the cloud's structure? How does the rate of star formation depend on the properties of the ambient medium?
PLANETS, STAR FORMATION, AND THE INTERSTELLAR MEDIUM 15 Recent Progress in Understanding Star Formation Spectroscopic observations at millimeter wavelengths recently provided evidence for protostars with infalling material. Three such sources have now been identified, and these are among the coldest and most deeply embedded young stellar objects known. Millimeter-wave interferometry has produced the first resolved images of disk-shaped structures around protostellar objects. These observations have also provided spectroscopic detection of disk rotation, confirming the nature of these objects as circumstellar disks. With the detection of disk rotation around a number of young stellar objects, near-infrared spectroscopy has provided additional evidence for circumstellar disks. Hubble Space Telescope (MST) observations have also detected disks and gaseous globules around low-mass stars in regions of massive star formation, providing a new approach to the study of such disks (Figure 2.1~. Deep-infrared imaging of giant molecular clouds has revealed that star formation occurs exclusively in dense molecular cores and that a significant, and perhaps dominant, fraction of all stars form in rich stellar clusters generally embedded in the most massive molecular cores. These observations have also suggested that the outcome of such star formation is an initial spectrum of stellar masses that is universal in form and similar to that of field stars in the neighborhood of the Sun. Millimeter-wave, optical, and infrared imaging have established a close relationship between optical jets and molecular outflows, indicating that these two types of flows have a common physical origin. HST observations have resolved optical jets and have revealed that they become collimated very close to the star, on scales compa- rable to the size of the solar system. Millimeter, submillimeter, and far-infrared observations have established that the most collimated and energetic bipolar outflows are closely linked with the most embedded and least evolved young stellar objects, suggesting that the most active outflow phase occurs simultaneously with the most active infall phase of protostellar evolution. Future Directions for Understanding Star Formation Improved angular resolution and sensitivity at infrared and submillimeter wavelengths are essential to future progress in all areas of star formation research. High-resolution imaging and spectroscopy of the disks, jets, and inner envelopes of the nearest protostellar systems are particularly important. Because such a capability is crucial for understanding star formation as well as planet formation, the development of space-based infrared interferom- etry to achieve the capability should be emphasized. In the near term, three scientific goals given below in priority order can be addressed within technological capabilities existing now and projected during the next decade: Characterizing the very earliest stages of star formation by observing the structure and dynamics of protostellar regions. High-resolution far-infrared and submillimeter spectroscopy at a velocity resolution of 0.3 km/s or better will enable the study of collapsing regions of dense cores on scales between 100 and 1,000 AU in spectral lines that are blocked by Earth's atmosphere. Spectroscopic observations of strong cooling lines from OH and H2O in the 50- to 350-,um band could enable the first identification and study of protostellar disk accretion shocks and of the cooling regions of molecular outflows. Observations of protostars in the 100- to 800-,um band could provide accurate determinations of protostellar luminosities and the protostellar luminosity function, thereby directly testing and refining predictions of star formation theory. Surveys in the 150- to 350-,um band could provide a complete inventory of protostellar activity in all nearby molecular clouds. 2. Determining the luminosity functions of embedded stellar populations over a range of differing environments. The best techniques to achieve this goal are high-resolution (0.1-arc see), near-infrared (2- to 5- ,um) imaging, and modest-resolution spectroscopy (R~3,000) of young embedded star clusters within 5 kpc of the Sun. Such observations will test the universality of the initial stellar mass spectrum, particularly in regions where high-mass stars form, and they will determine the rate of star formation as a function of physical environment and location in the galaxy. 3. Measuring the frequency, separations, and orbital motions of binary companions of protostars on
16 A NEW SCIENCE STRATEGY FOR SPACE ASTRONOMY AND ASTROPHYSICS scales of 0.5 to 5 AU. Observations of the orbital motions of protostellar binaries would give the first determina- tions of protostellar masses, measurements fundamental to the development of a theory of protostellar evolution. In addition, knowledge of the frequency of companions is vital to understanding the process of fragmentation in star formation and the survivability of protoplanetary disks. Although space-based observations are required to attain most of these objectives, it is possible to achieve a major scientific goal of star formation studies from the ground: of particular importance are imaging and spectros- copy of nearby protostellar regions at extremely high angular (10-milliarc see) and spectroscopic (R~106) resolu- tion at 350-,um wavelength. Such observations will enable resolved imaging of protostellar disks and inner envelopes in the nearest star-forming regions. This goal could be achieved with a ground-based interferometer array operating at 350 ,um at a high-altitude site with a minimum baseline of at least 15 km. EVOLUTION OF THE INTERSTELLAR MEDIUM IN GALAXIES The evolution of the interstellar medium in galaxies is central to galactic structure and star formation on a galactic scale. Astronomers wish to understand the processes that control the efficiency of star formation and the nature of the stars that form. On larger scales they need to understand the rate and extent of the dispersal of hot gas, ionizing photons, and nucleosynthetic products throughout the galactic disk and halo. Most generally, investiga- tions of the interstellar medium of galaxies may answer fundamental questions such as the following: What triggers star formation in a galaxy? What determines how star formation propagates and evolves in time? What is the feedback of star formation on the interstellar medium? What are the key extragalactic environmental effects on galaxy evolution? The birth sites of new stars are the giant molecular clouds (GMCs), which together occupy a small fraction of the volume of a galaxy but contain substantial mass (several times 109 solar masses). A typical (105 to 106 solar masses) GMC is probably built by assembling diffuse atomic clouds from a volume extending hundreds of parsecs on a side. These diffuse clouds are heated primarily by the interstellar ultraviolet radiation field and are cooled primarily by EC II] (158 lam). The formation of a GMC from diffuse clouds may be accompanied by shocks that emit copious tC II] (158-,um) and tO I] (63-~m) radiation. One of the greatest obstacles to understanding the cycles of star formation and galactic evolution is a lack of knowledge about the distribution on all scales of the material composing the diffuse interstellar medium. This information is critical for understanding key aspects of star formation, such as its efficiency, the energy dissipated, and the mechanisms that trigger it and shut it off. The Sun resides in a "local hot bubble"; over much of the volume within 100 pc, the density of interstellar matter is less than 1% of the mean interstellar value, and the temperature is extremely high: greater than 106 K. Astronomers do not know whether such low-density gas occupies 10% or 80% of the disk, nor do they know its structure and topology or those of the denser gas that surrounds it. Even locally, astronomers have no consistent model for the location of the million-degree gas that produces the very bright, 0.25-keV x-ray background. These factors can be central in determining the mechanisms that eventually control star formation and galactic evolution. Attempts to observe the spectrum of this hot gas have been frustrated by the low spectral resolution available, so that astronomers do not understand the physical conditions in the hot gas. Observations of 0.5- to 1-keV x rays suggest that there is a large amount of even hotter gas at several million degrees in the inner galaxy, which may drive a strong outward wind. At the outer limits of a galaxy, the gaseous galactic halo is an equally dynamic but poorly understood entity. The halo acts as a reservoir for the stellar and gaseous debris blown out of the disk, and for the cosmic rays that diffuse out. Galactic halos may also be replenished by the infall of gas from the intergalactic medium; observa- tions of quasar absorption lines suggest that galactic halos exist at high redshift and show evidence for massive- star nucleosynthesis. The origin, structure, and evolution of these halos are not understood.
PLANETS, STAR FORMATION, AND THE INTERSTELLAR MEDIUM Key Questions About the Evolution of the Interstellar Medium below: 17 Major unresolved issues concerning the evolution of the interstellar medium in galaxies include those listed Formation of Molecular Clouds · What are the relative roles of gravitational and thermal instabilities, magnetic fields, and spiral density waves in the formation of GMCs? · What determines the relative amounts and spatial distribution of atomic and molecular gas in galaxies? Physical Properties of the Diffuse Interstellar Medium · What are the physical properties of He different phases of the interstellar gas: the cold and warm neutral media, the warm ionized medium, and the hot ionized medium? · What is the spatial distribution of these components within the galaxy, on both the scale of clouds and the much larger scale of spiral arms? · What is the composition of interstellar dust, and what is the size distribution of its grains? - What is the ionization source for the warm ionized medium? · How do radiative losses and thermal conduction regulate the temperature of the hot gas? - How are cosmic rays accelerated? Interstellar Medium and Galaxies · How is star formation organized in galaxies? · What are the physical mechanisms that regulate the rate of star formation in galaxies? · What is the role of supernova remnants in determining the structure of the interstellar medium? · How are the nucleosynthetic products from supernovae and evolved stars mixed into the interstellar medium? · What causes starbursts? · How do globular clusters form? Galaxies and Their Environment · Are gaseous galactic halos the relics of galaxy formation or the products of violent events within galaxies? · How does gas from the nucleus and disk of galaxies mix into the halo? medium? · Under what circumstances is there significant mass exchange between galaxies and the intergalactic Recent Progress in Understanding the Evolution of the Interstellar Medium Significant advances in our knowledge of the diffuse interstellar medium during the past 5 years have come from a number of space-science missions, including the International Ultraviolet Explorer (IUE), Infrared Astro- nomical Satellite (IRAS), Cosmic Background Explorer (COBE), Roentgensatellit (ROSAT), Extreme Ultraviolet Explorer (EUVE), Kuiper Airborne Observatory (KAO), Infrared Space Observatory (ISO), and HST. Many of these studies have focused on the physical properties of the diffuse interstellar medium. The far-infrared imaging of the galaxy by IRAS has highlighted previously invisible regions, both the portions of dark clouds heated by star formation and the diffuse "infrared cirrus" ubiquitous throughout the disk and halo. IRAS has also clarified the importance of dust reradiation of starlight, in which small particles absorb ultraviolet and optical starlight and reemit it at mid- and far-infrared wavelengths. COBE images have tantalized us with coarse maps of the [C II]
18 A NEW SCIENCE STRATEGY FOR SPACE ASTRONOMY AND ASTROPHYSICS (158-,um) and [N II] (205-pm) emission lines, which trace the gaseous energy budget and the warm ionized medium, respectively. Similarly, mid-infrared spectroscopy from ISO shows the great value of measuring astro- physical conditions in diffuse atomic and molecular gas. Probes of the hot gas in the local interstellar medium have been provided by ROSAT, which measures soft x- ray emission from gas at 106 to 107 K. ROSAT has also observed "x-ray shadowing," in which a foreground interstellar cloud absorbs x rays emitted by distant hot gas. Such observations provide estimates of the tempera- ture, pressure, and structure of the hottest regions of the interstellar medium. Images and spectra from the KAO have enriched current knowledge of star-forming regions, supernovae, and galaxies in the far infrared. In addition to probing the coolest regions of the interstellar medium, mid- and far-infrared observations allow astronomers to study the reprocessing of starlight by dust grains and abundances of ionized nebulae from fine-structure lines. Spectroscopic observations by the IUE and the HST have provided high-precision gas-phase elemental abundances for a range of interstellar environments, including gas in the local medium, the galactic disk, and distant halo clouds. These studies have yielded new information about the composition of interstellar dust, the origin of the galactic high-velocity clouds, and the processes that transport gas between the galactic disk and halo. EUVE has provided valuable information on helium in the local medium. HST images of gaseous matter in star- forming regions have also revealed the complex interactions between stars and their surrounding environments, beautifully shown in the giant pillars of gas and dust in the Eagle nebula (Figure 2.2~. Future Directions in Understanding the Evolution of the Interstellar Medium Tackling the fundamental questions about the evolution of the interstellar medium in galaxies requires observational capabilities in space ranging from ultraviolet and x rays to the far infrared. In this section, TGSAA discusses specific scientific initiatives that would provide answers to many of the important questions discussed above by probing diffuse gas over the wide range of temperatures ~ 10 to 107 K) present in the interstellar medium. Although optical instruments are sensitive to ionized nebulae at 104 K, observers require spaceborne, infrared and submillimeter instruments to study cool gas (10 to 100 K), ultraviolet spectrographs to study the resonance lines that trace the atomic and ionized gas (102 to 106 K), and x-ray spectrographs to observe the hottest components (2106 K). Since observations of the interstellar medium in other galaxies can address many of the same scientific questions, most of these instruments can be used to study nearby galaxies as well. The most important future directions identified by TGSAA for research in this area over the next decade in priority order-are as follows: 1. Determining the large-scale three-dimensional structure of the interstellar medium. Spectroscopy of the [C III (158-,um), tO I] (63-,um), [N II] (205- and 122-pm), and (of lower priority) H2 (28-, 17-,12-,um) infrared emission lines at high spectral resolution (<10 knits) and moderate angular resolution (<30 arc min) are particu- larly important. These lines trace the energy budget, the warm ionized and neutral medium, and warm molecular gas throughout the disk and halo. Not only would a EC II] and tO I] survey locate the diffuse clouds in the galaxy, but comparison with H I (21-cm) surveys and infrared continuum surveys would also reveal the density and temperature structure of the clouds. The mapping of IN II] will trace the leakage of ionizing photons from massive stars in the disk into the halo. The first two H2 rotational lines at 28 Am (I = 2~0) and 17 ,um (I = 3~1) will be observed primarily in photodissociation regions, the warm outer envelopes of GMCs illuminated by intense far- ultraviolet fluxes from regions of active star formation. 2. Determining the connection between the Milky Way's disk and its gaseous halo. High-resolution ultraviolet absorption-line spectroscopy of stars and quasars is the relevant technique to employ. Observations are needed of faint background targets at moderate resolution (15 knits) and of brighter stars within 1 to 2 kpc at high resolution (1 knits). In the long term, it is important to extend this capability down to 91.2 nm in order to measure key absorption lines of heavy elements, H2, deuterium, O VI, C III, and S VI for a much wider variety of sources than will be possible with the Far-Ultraviolet Spectroscopic Explorer (FUSE). 3. Mapping the soft x-ray background with a spectral resolution of 1 eV and, in stages, with an angular resolution approaching 1 degree. Spectroscopy of the diffuse x-ray emission above 0.1 keV provides powerful diagnostics of the hot phase of the interstellar gas through the richness of its emission line spectrum. Understand
PLANETS, STAR FORMATION, AND THE INTERSTELLAR MEDIUM 19 ing the distribution of this hot gas is central to determining its origin and its role in controlling the evolution of the interstellar medium and galactic halo. 4. Mapping the spatial distribution of transition temperature gas (T ~ 1 to 3 x 105) in the local disk and halo. This can best be achieved by mapping the ultraviolet emission lines of C IV (154.8 and 155.0 nm) and O VI (103.2 and 103.8 nm) at an angular resolution of 1 degree and a velocity resolution of 15 knits. These emission lines are tracers of energetic processes in the interstellar medium. 5. Mapping the physical conditions and composition of the hot gas in supernova remnants, super- bubbles, and the galactic center. This undertaking will require the development of high-throughput, high- resolution, x-ray spectroscopy. Large-scale maps can be obtained in prominent x-ray emission and absorption lines to measure the ionization and chemical composition of the hot gas. (The same techniques can be exploited to study young stars embedded in GMCs.) The spectral resolution should be adequate to study the dynamics of supernova remnants (velocity resolution of ~100 km/s). _ ~ ~_A ~ _ _ J ^ ~ _~ ^ A.~ _ V _ ~ 6. Mapping the coldest parts of the interstellar medium in selected atomic and molecular emission lines and in broadband thermal emission from dust grains. This task will require the development of spectroscopy in the submillimeter and far-infrared bands. CONCLUSIONS For many years, astronomers have speculated that there should be planets around other stars, but only recently have extrasolar planets actually been detected. We now stand at the threshold of discovering both the nature of extrasolar planetary systems and how they came about. Are there other planetary systems such as the solar system, and if so, how did they form? The formation of planets is intimately tied to the formation of the stars they orbit; how do stars form? The formation of stars in turn is regulated by the structure of the interstellar medium; how does this medium evolve in the Milky Way and in other galaxies? Answering these questions presents a major technological challenge owing to the enormous range of scales involved from the size of galaxies to the size of planets and the range of wavelengths over which observations must be made, from x rays for the hot gas in galactic disks to millimeter and submillimeter radiation for the cold gas and dust in protostellar disks. The greatest gap in current knowledge is at the smallest scales, and motivates the long-term goals of develop- ing systems that can obtain spectra of terrestrial planets around nearby stars and image protoplanetary disks in nearby star-formin~ regions. The first goal requires the ability to detect 5- to 2()-m radiation from ~ planet within ~ ~ ~ 1 -- -a -a rig - ~~~~~~ ~ rat ~ . ~ . .. . · · .. ~ . ~ . . . . ~ . .. ~ . . _ _ u.1 arc see of a star that Is over a million times brighter. Since the nearest star-forming regions are about 150 pc away, the second goal requires a resolution of about 1 milliarc see at 5 to 10 ,um to image the inner parts of the disk and about 10 milliarc see at 50 to 100 ,um to image the outer parts. In the long run, it will be essential to achieve these extremely challenging goals if we are to observe terrestrial planets around other stars and understand how they formed. Initiation of the technology development necessary to meet these long-term goals is of prime importance to the astronomical community. Over the time scale considered by this report (the next ~10 years), the top three scientific priorities for the study of planets, star formation, and the interstellar medium are, in rank order, as follows: 1. Obtain a census of planetary systems around enough stars (~1,000) so that the frequency, separa- tions, and masses of planets comparable to or larger than Uranus can be investigated for a range of types of stars and stellar systems. This activity will require complementary approaches of radial velocity measurements (better than 10 m/s) and high-precision astrometry (better than 10 microarc see) over at least a 10-year period. Higher astrometric precision would allow the survey to be carried out to lower planetary masses and is clearly desirable. 2. Characterize the very earliest stages of star formation by observing the structure and dynamics of protostellar regions. Imaging and high-resolution spectroscopy in the far-infrared and submillimeter regions of the spectrum should reveal how the properties of the accreting gas, the circumstellar disks, and the molecular outflows depend on the evolutionary status and mass of the protostar.
20 A NEW SCIENCE STRATEGY FOR SPACE ASTRONOMY AND ASTROPHYSICS 3. Determine the large-scale three-dimensional structure of the interstellar medium and the star-for- mation regions within it. This project could be done by mapping the galaxy at high spectral resolution (<10 km/s) and moderate angular resolution (<30 arc min) in the infrared lines of [C II] (158 ~m), 1° I] (63 am), EN II] (205 and 122 am), and possibly H2 (28, 17, and 12 am). At somewhat lower priority, TGSAA rated the following projects as of comparable importance for the next decade: 1. Detect indirectly terrestrial-mass planets; 2. Perform ultraviolet spectroscopic studies of the connection between galactic disks and halos with a sensitivity significantly higher than that now provided by HST; and 3. Conduct near-infrared imaging and spectroscopic studies of young embedded star clusters.