Strategy for the Detection and Study of Other Planetary Systems and Extrasolar Planetary Materials: 1990-2000

7

Relation to Other Astronomical and Astrophysical Studies

The effort to detect objects near other stars will open up new observational capabilities that will have a decisive influence on the development of knowledge in related astrophysical research. Some of the fundamental questions of astronomy, such as the formation and evolution of stars and the formation, dynamics, and evolution of galaxies, require for their clarification much more accurate data than are available at present. In the following sections a few examples are given that connect the study of extrasolar planetary materials with other areas of investigation.

FUNDAMENTAL PROPERTIES OF STARS IN THE SOLAR NEIGHBORHOOD

The search for planetary systems will naturally involve an intensive study of the stars in the solar neighborhood, out to a distance of perhaps 100 parsecs. The fundamental properties of most of these objects, particularly those of low mass relative to the Sun, are not accurately known; they are crucially important, however, both for comparison with theories of star formation and stellar evolution and for establishment of the extragalactic distance scale. These properties include distances, masses, radii, luminosities, the mass spectrum, and the binary or multiple system frequency. An astrometric telescope with precise measurement capability would have a substantial impact on the accurate determination of distances by direct trigonometric parallax and therefore, indirectly, on most of the other quantities.

In the area of star formation, two important pieces of information would influence ideas on its relation to planet formation. First, what is the nature of the stellar mass function at the low-mass end? It is thought to peak at about 0.25 M; the number of stars per unit volume per unit mass interval declines for lower masses (see Figure 5.1 in Chapter 5). Observational data in this range are very uncertain, however, as there are large selection effects. Do substellar companions and planets lie on a continuum extending to the lowest masses, or are there fundamental discontinuities in the mass spectrum?

Second, what is the statistical frequency of binary systems of various mass ratios? Is the frequency of binary or multiple systems different among systems in which the primary is at the low-mass end of the main sequence? How common are substellar companions vis-a-vis planetary companions? Does the number of stars per unit mass interval really drop off sharply below 0.25 M? Such data should provide clues regarding the fundamental question of whether low-mass stars and substellar objects form by the same process as do giant planets. Are most substellar companions products of binary star formation, perhaps due to high angular momentum? Are most planets products of the accretion of dust?

In addition, information about main-sequence stars in the solar-mass range is extremely important, because the evolutionary lifetime is comparable to or shorter than the age of the galaxy. Theoretical models of stars and stellar evolution are compared with observed properties of stars, including their masses, luminosities, and surface temperatures. The purpose of the comparison is to obtain fundamental knowledge about the energy source, age, composition, and detailed structure of stars at various stages of their evolution. At present, masses of only a small number of stars have been determined with reasonable accuracy. Astrometric programs and radial velocity programs to search for extrasolar planets would undoubtedly yield a number of cases in which stellar masses could be measurable by the classical method of determining the orbits of a binary system. Furthermore, the accuracy of existing measurements of mass as well as binary statistics could be greatly improved. It is interesting to contemplate the detection of a black hole by the same astrometric technique as that used for planetary detection. It would also be desirable to have accurate luminosities of nearby stars; for these, distances are the greatest uncertainty.

An astrometric facility with the accuracy needed for planet searches could result in dramatic improvements in fundamental distance measurements for nearby stars. A further consequence of improved distance measurements would be increased accuracy of observational determinations of stellar radii, for example, by fitting infrared measurements to model stellar atmospheres. Thereby the accuracy of the empirical mass-luminosity relation and the empirical temperature-luminosity relation on the main sequence could be substantially improved and the observations extended to a much greater volume of space. Even with a resolution of 10^-4 arcsec, stars within 100 parsecs could have their distances measured to 1 percent accuracy, a great improvement over the measurement capability of the European Space Agency's Hipparchos satellite.

As one example, the luminosities of subdwarf stars in the solar neighborhood could be much more accurately determined. These objects lie below the ordinary main sequence in the H-R diagram, have low heavy-element abundances compared with those of the Sun, and are thought to be among the oldest stars in the galaxy. The determination of their precise location in the H-R diagram is an essential step in determining the absolute location of the main sequence of globular clusters. The H-R diagrams of these clusters, in connection with the theory of stellar evolution, are analyzed to give the age of the galaxy, one of the most important parameters in cosmological theories.

PROPERTIES OF STAR-FORMING REGIONS

The investigation of star-forming regions for evidence of protostellar objects, protoplanets, and nebular disks is likely to yield additional information on some of the fundamental questions about star formation. Examples include observational identification of protostars actually in the gravitational free-fall collapse phase; observation of bipolar outflow and its relation to star and planet formation; clarification of mechanisms for binary star formation and the relationship between the formation of binaries and disks; the role of magnetic fields, turbulence, and thermal pressure in initiation of star formation and during the collapse phase of protostellar condensations in molecular clouds; the lifetime of nebular disks, which determines the time available to form planets; and detailed studies of the radiation coming from protostars at all stages of their evolution.

Questions related to the dust in star-forming regions are closely connected to questions about planet formation. Does the presence of circumstellar dust automatically imply accretionary growth toward subplanetary masses? Is the dust usually blown away by stellar winds before it has a chance to accrete, or can the dust just stay in orbit around the star without forming planets? Has the dust observed in young stellar objects been injected into the interstellar medium by evolved stars, or does it nucleate and grow during the collapse phase? Do the infrared excesses recently discovered by IRAS around seemingly ordinary main-sequence stars like Vega imply that dust has already aggregated, perhaps yielding subplanetary masses resembling our asteroid belt, or are these places where planet formation has failed? In the latter case, Poynting-Robertson drag effects would result in dissipation of the dust disk in ~10⁶ to 10⁷ yr, and some mechanism for replenishment would be required. Is there evidence for the formation of halos of dust and gas, where cometary material could be accreting, in disklike young stellar objects?

More generally, what are the statistics of the presence, the structure, and the evolutionary time scales for such disklike objects, and how might high-precision astronomical measurements from the ground and in space best provide the basis for diagnosing these statistics? Is the formation of disks with sizes comparable to the solar system and masses comparable to the minimum mass solar nebula a common outcome of the star-formation process? A further question involves the star-formation rate in molecular clouds, which is believed to be limited by turbulence within the cloud. Bipolar outflows may provide the energy source that sustains the turbulence; in such a case, star formation would be self-regulated. Or is the magnetic field the main mechanism for limiting star formation?

Studies of the interaction between the outflows and the surrounding molecular cloud might be expected to provide crucial tests for any one of these hypotheses. With orbiting telescopes such as HST and SIRTF, it should become possible to study several hundred systems with ages between 10⁵ and 10⁶ yr to determine such properties as rotation, flattening, ringlike structures, and the role of winds in disk evolution and dissipation.

THEORY OF STELLAR EVOLUTION

The existence of planets, particularly massive ones, could generate observable effects in certain advanced phases of stellar evolution. When a normal star evolves from the main sequence to the red giant region, its radius expands by several hundred times. A large planet in orbit would be encountered by the expanding star and would begin to spiral in through the envelope of the giant as a consequence of frictional drag. The planet could accrete mass from the envelope, and at the same time the stellar wind from the giant would begin the process of ejection of the envelope.

The possible outcomes of this evolution range from complete evaporation of the planet to its spiraling down into the interior of the giant. If the combination of circumstances were favorable, the planet, having gained mass by accretion and having lost angular momentum by frictional coupling with the envelope, would end in a short-period orbit about the dense core of the giant. At the same time the low-density envelope of the giant would have been expelled.

This type of system, composed of an object of about 0.1 M (the augmented planet) in orbit about a white dwarf (the core of the red giant), has characteristics similar to those of the so-called cataclysmic variables, which are deduced to be binary systems whose outbursts in light can be interpreted as a consequence of the transfer of mass from one component to the other. The buildup from planetary mass to stellar mass in this process could represent a new and unusual mode of star formation.

Many fundamental questions regarding the processes of stellar evolution can be addressed through studies of dust. Dust is believed to carry the heavy-element products of stellar evolution through the cycle from the interstellar medium (ISM) into star-forming systems and then back into the ISM in the ejecta of evolved stellar systems. For example, silicate and carbon grains have now been observed in the ejecta of supergiants, novae, and planetary nebulae, in the material producing the general interstellar extinction, in molecular cloud cores, in pre-main-sequence objects embedded in these clouds, and in the comae and tails of comets in the solar system. Do grains that form in the ejecta of evolved systems survive to be incorporated into young stellar systems? What is the rate of destruction by interstellar shocks? Do meteoritic inclusions with anomalous isotopic compositions represent condensations formed in the ejecta of such transient energetic nucleosynthetic events as nova and supernova explosions? Do cometary dust grains and the zodiacal dust grains resemble the dust grains in the ISM, in circumstellar shells of evolved stars, and in regions of star formation? How do grains participate in such processes as the onset of stellar collapse, planetary accretion, and the coupling of outflow energy to molecular clouds?

Studies of dust in connection with the formation of planetary systems would help to illuminate elements of the larger processes that involve the generation of heavy elements in stars, the ejection of this material into the ISM and its incorporation into grains, and the subsequent cycling of the material into new stars. Because of their general astrophysical importance and their particular relevance to investigation of precursor planetary systems, the committee sets out specific recommendations concerning dust-related studies (see Chapter 6).

DISKS IN ASTROPHYSICS

The observation and theoretical interpretation of disk structures are critical elements in the study of extrasolar planetary materials. The improved understanding of disks that could be obtained by intensive investigation of relatively nearby systems, of the T-Tauri-type and others, could influence studies in other areas of astrophysics. Disk systems play an important role in stellar binary systems in which mass is transferred between the stars. Examples are cataclysmic variables and novae. Other systems whose theoretical interpretation involves disks include the unusual galactic object SS433 and quasars. One of the crucial elements in all discussions of disks is the mechanism for mass and angular momentum transport; some kind of turbulence is indicated, but its origin is completely unclear in most systems. In the specific case of the presolar nebula, there is a theoretically suggested physical basis for it.

It is evident that theoretical studies of the "solar" nebula could result in considerable input to studies of other disk systems, including consideration of such questions as evolutionary time scale, the physics of turbulence, the stability of the disk to both thermal and dynamical effects, and tidal effects. A particular area of interest appears to be the interactions between disk systems and the planets that may form within them. Recent studies of ring systems in our solar system have revealed that density waves produced by orbital resonances profoundly affect any structure in, for example, the Saturn system. This phenomenon could have wide application in other systems, such as disks in binaries.

As a further issue, the existence of jets in extragalactic objects has been linked to the presence of disks. In this connection the jets observed near pre-main-sequence objects as part of the complex of observations known as bipolar flows are also thought to be influenced by the presence of a disk. Is there a common mechanism involved, which can operate on widely different scales? Improved theoretical and observational understanding of both the generation and collimation of bipolar outflows near young stars could therefore have wider implications in the study of a vast range of other astrophysical phenomena.

PROBLEMS IN GALACTIC STRUCTURE

Progress on a number of problems in galactic structure, such as the distribution of mass in the galaxy, the metal content of the various components of the galaxy, and the origin of the globular clusters with respect to the origin of the galaxy itself, is critically dependent on knowledge of stellar distances and motions. One of the instruments proposed in this report, a large astrometric telescope with the capability of measuring displacements small enough for the detection of planets, could also be used fruitfully in the investigation of such problems.

For example, through measurement of their trigonometric parallaxes, the distances of Cepheid variables could be measured directly for the first time. These objects, through the relation between their oscillation periods and absolute luminosity, provide a fundamental "standard candle" for establishment of the distance scale in the local group of galaxies and in the universe as a whole. The accurate measurement of the distances to even a few of the nearer Cepheids would provide the long-sought after "zero point" for the period-luminosity relation. An astrometric telescope with 10-4-arcsec precision could determine the distances to at least 20 Cepheids within 5 percent. Distances could also be measured for other interesting classes of stars such as the RR Lyrae variables.

Accurate distances to these and other types of stars, combined with measurements of stellar motions at different distances from the galactic center, would contribute to the determination of the density distribution of stars and the galactic gravitational potential in both the spheroidal and disk components of the galaxy. The rotation curve of the galaxy within a few kiloparsecs of the Sun could be more accurately determined. An astrometric telescope, if used at near-infrared wavelengths, could measure the motions of stars and other features in the vicinity of the galactic center, improving our knowledge of that fascinating region and testing the hypothesis of the existence there of a black hole.

The study of globular clusters is critical for the understanding of galactic structure. The determination of their space motion with respect to the galactic center would be possible for the first time with precise astrometric observations. For instance, a cluster at a distance of 1 kiloparsec from the Sun with a transverse motion of 100 km sec-1 would show a displacement on the sky of 0.2 arcsec in 10 yr. These measurements would lead to the distances of the globular clusters, would provide a determination of the galactic gravitational potential at large distances from the Sun, and would allow the mass-to-light ratio of the galaxy to be deduced. This observed quantity is crucial for the establishment of the properties of the cold, dark component of the galactic mass. One would also be interested in measuring globular clusters of various metal contents, in order to determine if there is a correlation of metal content with orbit. In addition, the membership of a globular cluster could be determined, and the individual orbits of stars in the cluster would be measurable, giving an indication of the dynamical evolution of the cluster.

Some measurements of the relative motions of other nearby galaxies with respect to our own galaxy would also be possible—the proper motions of the Large and Small Magellanic Clouds, for example, correspond to displacements of 4 milliarcsec in 10 yr. With these results one would be able to determine the dynamics and masses of the systems, deduce the history of past interactions, determine the direction of the angular momentum, and possibly probe the gravitational potential of our galaxy at very large distances.

IMPLICATIONS REGARDING THE SEARCH FOR LIFE

This report does not deal with various attempts to search for extraterrestrial intelligence, such as radio telescope "listening" surveys, as a mode of searching for planetary material around other stars. Yet the two endeavors have an obvious connection. For example, if artificial radio signals were detected, it would presumably indicate the evolution of planetary systems, as well as life, somewhere in the galaxy; the habitable planets would probably, but not necessarily, be in the star system from which the radio signals emerged.

Perhaps more realistic than an imminent discovery of extraterrestrial life is a scenario in which objects in the planetary-mass range are detected without any concomitant evidence of life. Detector systems are evolving with such rapidity that we can anticipate positive identification of extrasolar planets, if they exist, within a decade or so of the initiation of a comprehensive search. Current theories lead us to suspect that other planetary systems, habitable planets, and perhaps even life forms are likely, but as yet we have no direct confirmation that even a single extrasolar planet exists. An actual discovery of planets would give the search for extraterrestrial life more impetus and a somewhat firmer scientific basis than it has today.

If planets were detected, an immense and immediate popular and scientific interest would be generated. People would want to know if the planets are Earth-like, or habitable. A spectroscopic search for oxygen or radio listening would suddenly be given specific targets. Viewed in this way, a successful search for evolved planetary systems around other stars can be seen as stimulating the search for life elsewhere in the universe.