National Academies Press: OpenBook
« Previous: 2 RECOMMENDED PRIORITIES FOR ASTRONOMY AND ASTROPHYSICS IN THE 1980S
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 37
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 38
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 39
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 40
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 41
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 42
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 43
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 44
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 45
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 46
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 47
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 48
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 49
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 50
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 51
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 52
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 53
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 54
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 55
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 56
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 57
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 58
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 59
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 60
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 61
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 62
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 63
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 64
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 65
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 66
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 67
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 68
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 69
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 70
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 71
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 72
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 73
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 74
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 75
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 76
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 77
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 78
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 79
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 80
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 81
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 82
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 83
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 84
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 85
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 86
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 87
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 88
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 89
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 90
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 91
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 92
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 93
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 94
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 95
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 96
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 97
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 98
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 99
Suggested Citation:"3 FRONTIERS OF ASTROPHYSICS." National Research Council. 1982. Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee. Washington, DC: The National Academies Press. doi: 10.17226/549.
×
Page 100

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.

Frontiers of Astrophysics . LARGE-SCALE STRUCTURE IN THE UNIVERSE Probes of Large-Scale Structure As seen on photographs taken with the largest telescopes, galaxies appear to drift in the depths of space like motes in a sunbeam. Everywhere they are clumped in groups containing a few galaxies, and occasionally in clusters of a thousand or more. Some clusters clump in superclusters 50 megaparsecs or more across. On even larger size scales, however, groups and clusters of galaxies seem to be distributed nearly at random, the number in a given volume of space being about the same throughout the Universe. This uniformity of the distribution of matter on very large scales invites comparison between the observations and a simple model of the Universe, or cosmology, derived for a uniform distribution of matter from Einstein's General Theory of Relativity. According to this model, the geometry of space-time is curved by matter, and the curvature forces the matter to move: at any epoch the Universe must be either expanding or contracting. Hubble's discovery in 1929 that the Universe is actually expanding forces us to confront a bizarre implication of the theory: that an expanding Universe must have originated in a powerful explosion-referred to as the big bang before which neither time nor space had any meaning. In the half-century since Hubble's momentous discovery, astron 37

38 ASTRONOMY AND ASTROPHYSICS FOR TEIE 1980's omers have been probing the Universe in space and in time. Using radio and optical telescopes, they have found objects so distant that they are receding from us at 90 percent of the speed of light. With microwave antennas, they have discovered a faint radio noise that they interpret as the remnant of the big bang itself. From the theory of the nuclear reactions that must have taken place during the first 3 minutes, they have calculated the abundances of key elements and isotopes such as hydrogen, deuterium, and helium, which were produced in the big bang; with ground-based telescopes and ultra- violet spectrographs in Earth orbit they have verified that the actual relative numbers of these atoms in space agree surprisingly well with theoretical predictions. The big bang has become the standard model with which to com- pare observations. This is not to say that it is completely correct: the data are imprecise; their interpretation may be in error; and the theory could be wrong. A central problem for the future is the further development of the big-bang model and its testing against all avail- able observations. The big-bang model requires that matter is distributed uniformly on large scales. By using a variety of approaches, it is now possible to test whether this is true. Observers have used apparent magni- tudes as a rough measure of the distances of galaxies; plotting the directions of galaxies in various distance ranges, they have found that on scales exceeding 100 megaparsecs, galaxies are distributed rather uniformly. One can obtain the precise location of each galaxy in three dimensions by determining its red shift spectroscopically. Recording the spectrum photographically is time consuming, but the recent development of electronic array detectors has speeded up the recording of spectra so greatly that red-shift surveys of thousands of galaxies are now possible. The resulting three-dimensional distri- bution appears to be uniform on the largest scales. It is anticipated that red-shift surveys of much more distant galaxies will be com- pleted during the 1980's. X-ray and gamma-ray astronomy also tell us about the large-scale distribution of matter. A diffuse background emission not attributable to known sources appears in both spectral regions; its near isotropy proves that it cannot originate within the Galaxy but must instead originate at distances comparable with the size of the Universe itself. The High-Energy Astronomical Observatory-1 (HEAo-1) x-ray ob- servatory established that the x-ray background is highly isotropic and that its spectrum between a few and about 60 kiloelectron volts (keV) agrees closely with the radiation expected from a gas having

Frontiers of Astrophysics 39 a temperature of about 500 million degrees, leading to the suggestion that such gas is distributed uniformly between the galaxies. The Einstein (HEAo-2) x-ray observatory, on the other hand, discovered that individual quasars at large distances are powerful x-ray sources in the few-keV range powerful enough, in fact, that quasars at even larger distances than can be detected individually by the Einstein x- ray observatory must account for a substantial fraction of the ob- served x-ray background in the few-keV range. As some quasars have also been found to be powerful gamma-ray sources, the gamma- ray background may also be due to quasars. It is still not clear, however, how quasar spectra would sum up so as to mimic the spectrum of hot gas. The Advanced X-Ray Astrophysics Facility (AX~) recommended in this report can observe sources 100 times fainter than could the Einstein x-ray observatory and can thus determine whether faint quasars account for the observed background at ener- gies of a few keV. Measurements of faint quasars by the Gamma Ray Observatory (GRO) will give similar information for the back- ground at gamma-ray energies. If it proves that the x-ray and/or gamma-ray backgrounds are actually due to quasars, the fact that the background is highly isotropic requires that matter at great dis- tances is distributed very uniformly. If, on the other hand, inter- galactic gas is responsible for at least part of the x-ray background, one can infer that it is distributed uniformly; moreover, the amount of gas required is an important datum for the theory of evolution of galaxies. The cosmic microwave background radiation also gives information about the large-scale structure of the Universe. Precise measurements have revealed a smooth variation in its intensity over the sky that is attributable to the Earth's motion through the cosmos. The ob- served variation is unexpectedly large, corresponding to a velocity of 500 km/see for the Local Group of galaxies with respect to distant matter. The same measurements reveal no other certain variations larger than 0.03 percent, indicating that the Universe was highly uniform at the time the background radiation last interacted with matter. Ground-based experiments indicate that the spectrum of the microwave background radiation does not deviate significantly from thermal, as predicted by the big-bang model, but a balloonborne submillimeter experiment points to discrepancies that are difficult to explain. Both variations in intensity with direction and deviations from a thermal spectrum will be measured over the entire spectral range with improved precision (about 0.01 percent) by the Cosmic Background Explorer (COBE) mission planned by NASA.

40 Expansion Time Scale ASTRONOMY AND ASTROPHYSICS FOR THE 1980's The big-bang model predicts that galaxies should move away from each other with velocities that are proportional to their separations. Slipher and Hubble found evidence for such a relationship in exten- sive measurements of the brightnesses and red shifts of galaxies during the 1920's; the constant of proportionality between a galaxy's velocity and its distance is called the Hubble constant. According to relativistic models, the reciprocal of the Hubble constant (the "Hub- ble time") is roughly equal to the present age of the Universe that is, the time that has elapsed since the big bang. Determining the value of the Hubble time requires the measure- ment of the distances of remote galaxies, using a "ladder" of inter- connected distance scales determined by different methods; each step of the ladder reaches further into space. Hubble's own estimate for it was 2 billion years. It has since been revised several times to 5, then to 10, and then to 20 billion years; the latest estimates are between 10 billion and 20 billion years. Each revision has been the result of a major advance in understanding the properties of stars or galaxies that are used to construct the ladder of distance scales. The value of the Hubble time enters all cosmological calculations in a fundamental way. To find its true value, each step of the ladder of distance scales must be secure, and any contributions to the ve- locities of galaxies that are not due to the expansion of the Universe must be taken into account. An example of the latter effect is the motion of the Local Group of galaxies revealed by study of the cosmic background radiation; when this is taken into account, a more con- sistent set of data for the Hubble time emerges. Refinement of the distance ladder will take much more work. Development of more precise astrometric methods, as recommended in this report, will make possible a more accurate measurement of the distance to the Hyades star cluster, the first step in the ladder of cosmic distance scales. Because of its extremely faint limiting mag- nitude, Space Telescope (ST) will for the first time resolve Cepheid variable stars in the Virgo cluster, thereby eliminating an uncertain intermediate step of the distance ladder. The continued deployment of advanced optical detectors at ground-based telescopes will make possible the rapid measurement of red shifts of galaxies at moderately large distances, where the velocity field should be one of nearly pure expansion; ST can determine the distances of the same galaxies by comparing the brightness of their globular clusters with the bright

Frontiers of Astrophysics 41 ness of those in the galaxies of the Virgo cluster, whose distances are known accurately. The Early Universe The cosmic microwave background radiation carries information about the Universe before it was about 1/100,000 of its present age, so the COBE experiment is fundamental to studies of the early Universe. Other clues depend on the nucleosynthesis of various elements and isotopes in the first 3 minutes. Theoretical predictions of their abun- dances depend critically on the amount of ordinary matter present during that period. If the amount is low, the resulting deuterium abundance would be high and the helium abundance low; if the amount is high, the opposite would be the case. Present information on the abundance of deuterium and helium in interstellar space in our Galaxy, taken at face value, indicates that the amount of matter is too low by a factor of 10 for its gravitation to be able to halt the expansion of the Universe. However, helium has been produced and deuterium has been destroyed in stars, so present abundances in the Galaxy may not be the same as in the primordial gas that emerged from the big bang. Abundances in intergalactic gas, if it exists, should be primordial. Astronomers have discovered absorption-line systems in distant quasars that probably originate either in clouds formed by the out- ward ejection of thick shells of gas from the quasar itself or in in- tergalactic clouds lying along the line of sight. In the first case, the phenomenon would resemble the late stages occurring in the stellar outbursts known as novae. In the second case, the clouds should contain very little carbon or other medium-weight elements, which are telltale signs of stellar nucleosynthesis, because such gas would never have been inside a galaxy. The gas in such clouds would be a good candidate for the study of primordial helium and deuterium. Observations of helium and deuterium in such gas, however, must be made at much shorter wavelengths than are accessible to ground- based observatories; they require ST. With ST we can study helium lines in clouds of red shift greater than unity and deuterium lines in clouds of all but very low red shifts. Groups, Clusters, and Superclusters The grouping of galaxies on various size scales can be studied by calculating the statistical correlations between the observed positions

42 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's of galaxies. In earlier studies, the apparent magnitudes of galaxies were taken as measures of their distances, and their positions then follow from their observed directions in the sky. The calculated correlations between the positions derived in this way decrease as an inverse power of the distances between pairs of galaxies. A simple model to explain this is based on gravitational clustering of point masses, which are initially distributed at random but which then move under their mutual gravitation as the Universe expands. This model reproduces many of the features of the ob- served clustering of galaxies, so that galaxies may have formed early in the expansion of the Universe and clumped together later by gravitation. Recent observational work, however, has brought out an un- expected new feature in the distribution of galaxies. Aided by red- shift measurements, which furnish the distances of galaxies much more accurately than estimates based on their apparent magni- tudes, astronomers have found that groups of galaxies outside of clusters are not sprinkled at random through space but instead lie in great sheets between the clusters, leaving vast empty re- gions between. To explain this may require a new theoretical model. in which Galaxies formed rather late. At first, giant tur- bulent cells of gas collided, compressing the gas into sheets; only after the sheets formed did the galaxies condense from them and then begin to clump together as in the earlier model. Two kinds of data are required if we are to understand the formation and clumping of galaxies. First, red-shift surveys em- bracing a large number of galaxies are needed. For the nearer galaxies, it is feasible to obtain red sniffs w^tn currently ava'^A- able telescopes of moderate size, equipped with array detectors and fast spectrographs. To penetrate more deeply into space, however, large telescopes will be needed. Telescopes of the 5-m class will make important contributions, but only a new tele- scope of the 15-m class, such as the New Technology Telescope (NTT), can measure the red shifts of galaxies at large distances rapidly enough to accumulate the required number of galaxies. The raw speed of NIT, made possible by its order-of-magnitude increase in collecting area over the previous largest telescopes, is critical for this project. A A ~ ~ ~ ^ ^ ~ ~ ~ ~ ~ A _ A ~ ~-~ O ~ ~ -id. ·.1 .1 ·1 Hidden Mass and the Fate of the universe For the past 20 years, astronomers have been increasingly puzzled by the ,ihidden mass,, problem: the matter that constitutes most of

Frontiers of Astrophysics 43 the mass of the Universe is invisible. The spectra of galaxies indicate that, like our own Milky Way Galaxy, they contain normal stars; however, the internal motions in galaxies are so large that they would fly apart if the only gravitational attraction holding them together were that of the stars we see. There must be additional mass present in some form that is hidden from our immediate view- enough to supply the gravitational attaction required for stability. The rotational velocities observed in spiral galaxies demonstrate that the amount of hidden mass inside a given radius increases approximately linearly with radius out to distances of nearly 100 kiloparsecs. Similar results emerge from studies of groups of two or more galaxies: their masses must be at least 10 times greater than the masses of all the visible stars in them. Solution of the hidden-mass puzzle is a major goal of astronomy in the decade ahead. The first task is to find how it is distributed. The velocities of globular clusters in the outer reaches of galaxies reflect the strength of the local gravitational field and hence the distribution of mass in the parent galaxy. Since globular clusters in even relatively nearby galaxies are extremely faint, spectroscopic measurements of their velocities can be made only with a telescope as large as NIT. Galaxies themselves can serve as probes of the distribution of mass in clusters and superclusters of galaxies. Since galaxies are much brighter than globular clusters, work on clusters of galaxies is already proceeding with intermediate-sized telescopes. However, measurements of velocities of galaxies in distant clusters are essential to determine how the distribution of mass has changed with time; this will require observations with NIT. Various possibilities have been suggested to account for hidden mass: diffuse gas, massive neutrinos, collapsed stars (white dwarfs, neutron stars, black holes), and faint red dwarfs. Diffuse gas can be ruled out as a dominant component of either galaxies or clusters of galaxies through radio, optical, and x-ray ob- servations; although 100-million-degree gas exists in clusters of gal- axies, the amounts are not sufficient to hold the clusters together. Massive neutrinos, if they exist, might fall into clusters of galaxies, and possibly even into galaxies themselves, thus contributing to the hidden mass. Collapsed stars of various types could in principle constitute much of the hidden mass; however, such stars are the descendants of massive main-sequence stars and so would dominate the total mass only if, at early epochs of star formation, massive stars dominated the total mass of main-sequence stars. Just the contrary is observed to be the case for star formation in our Galaxy near the Sun: faint

44 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's red dwarfs, which are of low mass, are so numerous that they ac- count for most of the mass bound up in stars. One could speculate that there were many more massive stars in the outer parts of galaxies during the early stages of galaxy evolution, so that large numbers of collapsed stars would exist there today. However, if that were so, one would expect a higher concentration of heavy elements in the outer parts of the galaxies, since massive stars synthesize heavy elements and eject them into the interstellar medium; this is contrary to observation. Faint red dwarfs could also account for the hidden mass, as large numbers of them in the outer parts of galaxies would be consistent with both the lower concentrations of heavy elements and the lower light levels observed there. It may just prove possible to test this hypothesis by using the recent discovery that red dwarfs are rela- tively luminous sources of coronal x rays. AXAF will be able to detect such red dwarfs by observing their integrated coronal x-ray emission if they are numerous enough. The hidden-mass problem is intimately connected with the ques- tion of the ultimate fate of the Universe. According to the big-bang model, the Universe will continue to expand forever if the amount of matter in it is less than a critical value calculated to be between 0.5 x 10-29 and 2 x 10-29 g in each cubic centimeter. If the amount of matter exceeds the critical value, the present expansion will reverse at some time in the distant future, and the Universe will collapse back into a singular state similar to the big bang. The observations of deuterium and helium discussed earlier suggest that the amount of ordinary matter is only 10 percent of the critical value, so that only massive neutrinos could raise it above the critical value. A lower limit on the total amount of mass in all forms is obtained from the masses of clumps in the distribution of galaxies; current estimates suggest that the aggregate amount of matter in such clumps may be as much as 40 percent of the critical value. Since this is larger than the upper limit on the amount of ordinary matter obtained from observations of helium and deuterium, massive neutrinos may con- ceivably account for most of the matter in the Universe. Massive neutrinos are discussed further in the last section of this chapter. EVOLUTION OF GALAXIES The Study of Galaxies Like the Galaxy in which we live, the 100 billion or more galaxies in the visible Universe are fascinating systems in their own right. As

Frontiers of Astrophysics 45 the nuclear and gravitational energy stored in them is released, it is likely that galaxies evolve toward objects evermore structured and compact. Among the variety of forms that galaxies take, Hubble discerned several recurrent patterns spirals, ellipticals, lenticulars, and irreg- ulars; these patterns have still not been completely explained theo- retically. Ellipticals and lenticulars are nearly devoid of interstellar gas and dust, while spirals and irregulars contain gas and dust, as well as young stars formed recently from them. Until recently, the gas and dust in spiral galaxies other than our own could be studied with high angular resolution only at optical wavelengths, by imaging the dark interstellar dust clouds and the luminous gas clouds heated by bright young stars. Now the Very Large Array (VLA) radio tele- scope can image galaxies both in the 21-cm line produced by inter- stellar atomic hydrogen and in the synchrotron radiation produced by relativistic electrons gyrating in interstellar magnetic fields; it can thus trace the distribution and state of the interstellar medium with angular resolution comparable with that of optical telescopes. As in all fields of astronomy, spectroscopy is the key to deeper understanding. Ground-based optical spectroscopy of galaxies dem- onstrates that a major component of most galaxies is stars of various masses and ages, like those in our Galaxy. However, present ground- based telescopes are hard pressed to obtain the spectra of extremely faint subsystems of galaxies, such as individual giant stars, regions of ionized gas, and globular clusters; they are too small to permit collection of photons at a sufficiently high rate. NTT, with its order- of-magnitude increase in collecting area, can obtain the spectra of such objects, thus making possible a whole new range of studies related to chemical composition, distribution of stellar masses, and rotational and random velocities within galaxies. For a galaxy of a given red shift, NTT will make possible studies with much higher spectral resolution; for the same spectral resolution, it can carry out studies on galaxies of much higher red shift. The latter capability is crucial for analysis of objects of large red shift that will be discovered by ST. One of the most striking capabilities of the new instruments rec- ommended for the 1980's is the systematic exploration of the de- pendence of various galactic properties on red shift at greater and greater cosmological distances. Big-bang models of the Universe pre- dict such a dependence because the evolution of galaxies with time translates into changes with lookback time, and hence with red shift. ST and NTT will be able for the first time to observe galaxies with red shifts substantially exceeding unity, corresponding to lookback times

46 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's that are more than half of the Hubble time. ST can image such distant objects because the sharpness of its images makes them stand out against the background, and NIT can obtain their spectra because it has a much larger collecting area than present large telescopes. If the matter comprising the inner parts of galaxies has already settled into an equilibrium state within considerably less than a billion years after the big bang, the forms of galaxies would not depend sensitively on red shift out to red shifts of 10 or so. However, the evolution of stars and the conversion of interstellar gas into stars proceeds much more slowly and should be observable at much lower red shifts. The spectra of isolated elliptical galaxies should manifest subtle changes that reflect the evolution of the stars that they contain, while isolated spiral galaxies should in addition manifest the progressive depletion of interstellar matter, as well as its enrichment in heavy elements produced by supernova explosions. A major indirect effect will be the reduction in the number of short-lived massive stars as the gas required to form them is depleted. Failure to observe such basic predictions of big-bang theory would force major revisions in current thinking. Formation of Galaxies The first relativistic models of the big-bang Universe were derived by Friedmann in 1922. For simplicity, he assumed that matter is distributed absolutely uniformly. Although this assumption conflicts with the existence of stars and galaxies, the model is useful because matter is in fact distributed quite uniformly when averaged over large distances. Still, the origin of galaxies in a big-bang model is an unresolved problem. Many properties of galaxies can be explained at least qualitatively if it is assumed that they originated in small fluctuations in the amount of local matter in the early Universe. At that time, the be- havior of matter was governed by the pressure exerted by the cosmic background radiation. Two types of density fluctuations could have existed. One type, so-called isothermal fluctuations, would have led to gravitationally unstable clumps of matter if they had involved more than 105 to 106 solar masses; another type, adiabatic fluctua- tions, would have led to gravitationally unstable clumps if they had involved more than 10~3 to 10~4 solar masses. In both cases, instability would have set in about 100,000 years after the big bang, and as a result, the matter in the fluctuations would soon cease to participate in the cosmic expansion, would then become more dense as self

Frontiers of Astrophysics 47 gravitation drew the gas together, and would ultimately form discrete gas clouds of various masses. The Cosmic Background Explorer (COBE) satellite will yield impor- tant information on the proposed instability process by observing the disturbances in the background radiation that would accompany any density fluctuations in the early Universe. Adiabatic fluctuations, which involve variations in temperature and hence in the intensity of the cosmic background radiation, would result in intensity vari- ations on angular scales of a few degrees if the masses involved in the fluctuations are about those of clusters of galaxies. Complemen- tary information about fluctuations on the smaller angular scales corresponding to individual galaxies (less than a degree) will be obtained by the Large Deployable Reflector (LDR) in space. The theory of adiabatic fluctuations has been worked out in detail for the case in which there is a random collection of initial fluctuations of various sizes and masses. Fluctuations involving 10~3 to 10~4 solar masses, usually identified with groups and clusters of galaxies, should form clouds first; the formation of galaxies would have taken place later within these clusters and groups. Clusters and superclusters con- taining more than 10~3 to 10~4 solar masses must have formed through later gravitational clustering of the original mass aggregations of this size. Alternatively, isothermal fluctuations may have dominated the in- itial stages of galaxy formation. In this case, the first objects to form must have had masses from 105 to 106 solar masses, and galaxies must have been built up later by gravitational clustering of these smaller objects. The fact that globular clusters containing 105 to 106 solar masses are so common would be a natural result of isothermal fluctuations. If galaxies formed out of objects having 105 to 106 solar masses, then groups and clusters of galaxies must have formed sub- sequently through gravitational clustering of the galaxies themselves. This process can be modeled with computers by treating each galaxy as a point mass and calculating its gravitational interactions with its neighbors. Extensive simulations of gravitational clustering have been carried out in this way during the past decade; the results agree with observations in some respects, but they do not predict the large holes devoid of galaxies that have been observed between clusters. It is still uncertain whether galaxies or clusters of galaxies originated first. None of the existing computer simulations of either galaxy collapse or clustering addresses the origin of the fluctuations themselves. Current attempts to answer this important question, based on Grand Unified Theories of elementary particles, are encouraging.

48 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's The collapse of a gas cloud to form a disk galaxy like the Milky Way and the subsequent formation of the first generation of stars are both believed to have taken place during the first billion years of galactic evolution. These processes could therefore be observed only in galaxies so remote that the radiation now received from them was emitted in their early youth. Although presumably very faint, this radiation, red shifted by a factor of 10 or more because of the cosmic expansion, might be detected by the next generation of in- struments. Optical and ultraviolet radiation from young galaxies, if not absorbed by dust near the galaxy or in intragalactic space, would be red shifted into the infrared region, where the extreme sensitivity of the Shuttle Infrared Telescope Facility (SIRTF) would permit it to be detected. Overlapping shock waves from early supernova explo- sions could heat the interstellar gas to temperatures as high as 1 billion degrees, generating dust-penetrating x rays with energies up to 100 keV. These x rays, now red shifted into the few-kiloelectron- volt range, should be detectable by AXAF according to recent calcu- lations. If a galaxy like ours that had just been born were discovered, it would present a breathtaking opportunity for study. Evolution of Galaxies Once galaxies form, they evolve slowly as dying stars inject newly synthesized atomic nuclei into the remaining interstellar gas, from which new generations of stars are formed. Current models suggest that most of the matter entering the interstellar medium is ejected by stars of moderate mass, which evolve into red giants and then planetary nebulae, as red giants lose most of their outer layers in low-velocity stellar winds, while the deeper layers are thrown off in final outbursts that produce planetary nebulae. Most of the energy injected into the interstellar medium, on the other hand, comes from the explosions of massive stars, in which an entire star is disrupted to form a supernova. Supernova explosions are also the principal sources of heavy elements. Only stars whose mass equals or exceeds that of the Sun evolve significantly in a Hubble time, so that during the generations of stars that have occurred since the galaxies formed, much of the interstellar medium has found its way into low-mass stars that have evolved little over the lifetime of the Universe. Thus, interstellar matter has been continually enriched in heavy elements, while its mass has been continually reduced as more of it is converted into slowly evolv- ing stars.

Frontiers of Astrophysics 49 Star formation apparently proceeds at different rates in spirals and ellipticals, since at present spirals contain large amounts of inter- stellar matter, while ellipticals contain little. It has been suggested that early in the life of ellipticals, the star-formation rate was high enough that frequent supernova explosions were able to drive the remaining gas out of the galaxy in a so-called "galactic wind," thus quenching further star formation. Supernova explosions apparently occur frequently enough today to keep ellipticals swept clean of interstellar matter. In spirals, on the other hand, the initial supernova rate may not have been great enough to cause a catastrophic purging of the interstellar medium, and so a sufficient amount of interstellar matter still remains today to support active star formation. Thus, the different amounts of interstellar medium in ellipticals and spirals may be a consequence of different initial rates of formation of massive stars, which become supernovae; why the rates should have been different is an important unsolved problem of galactic evolution that will be addressed through the study of galaxies of large red shift. The existence of galactic winds is in accord with current x-ray observations of rich clusters of galaxies, in which most of the galaxies are either ellipticals or their close cousins, the lenticulars. In many such clusters there is a hot, x-ray-emitting intergalactic medium, whose total mass and chemical composition are consistent with the supposition that it has accumulated from galactic winds. The removal of gas from galaxies is aided by the intergalactic medium, once es- tablished, because it sweeps gas from galaxies as they move through it. There is evidence for this process in the x-ray and radio images Map of the sky in soft x-rays (0.15~.28 keV) from a series of rocket flights. (Photo courtesy of the University of Wisconsin x-ray group)

50 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's of certain galaxies that are being swept at present. One can under- stand the absence of gas-rich galaxies in rich clusters of galaxies along such lines, but a detailed understanding awaits x-ray observations with AXAF, which, because its sensitivity and angular resolution are greater than those of the Einstein x-ray observatory, can observe the process of sweeping at greater distances. Any realistic scenario for galactic evolution must take into account the effects of cosmic rays- the relativistic particles whose presence is inferred from experiments with charged-particle detectors in space. Constituting a prominent and permanent component of our Galaxy, cosmic rays have isotopic abundances that imply that they remain tens of millions of years in the Galaxy, most of which time is spent in the Galactic halo. In effect, cosmic rays constitute a relativistic gas with a pressure comparable with that exerted both by random mo- tions of the interstellar gas and by the interstellar magnetic field; the cosmic-ray gas therefore plays a critical role in the equilibrium infla- tion of the gaseous disk, in the fragmentation of the interstellar medium into molecular-cloud complexes, and, presumably, in the support and activation of halos of galaxies. Determining the origin and propagation of cosmic rays is therefore an important aspect of the overall effort to understand the course of galactic evolution. Cosmic-ray experiments on the Space Shuttle will advance these goals by supplying new information about the composition, energy spectra, and isotope distributions of the cosmic rays themselves. Verifying our ideas of galactic evolution will require intensive stud- ies of our own and nearby galaxies, as well as observations of galaxies so remote that their properties appear different from those of the more evolved galaxies nearby. It is now understood that stellar mass loss has a profound effect on galactic evolution. With its high sen- sitivity, STRTF will discover many more cool red-giant envelopes, in which the stellar light is degraded to infrared radiation by large quantitites of embedded dust; submillimeter, millimeter, and infrared telescopes, including the LDR in space, the 10-m submillimeter-wave antenna, the 25-Meter Millimeter-Wave Radio Telescope, and NIT, will permit spectroscopic observations of the spectra of molecules in such envelopes, leading to the determination of nuclear and isotopic abundances, which are clues to the nuclear processing that has oc- curred in the parent stars; they will also determine the velocity and mass of the outflowing gas, crucial parameters for calculating the rate at which mass is being ejected into the interstellar medium. Because of its exceptionally faint limiting magnitude, ST will be able to identify a much larger fraction of the low-mass stars believed

Frontiers of Astrophysics 51 to lie near the Sun than has been possible up to now, thus permitting a much more accurate assessment of the mass stored in these stars; this assessment will be aided by SIRTF, which will be much more sensitive to the extremely cool faint stars than are the infrared tele- scopes now available. Because of its ultraviolet sensitivity, ST iS ex- pected to find many new white dwarfs, hence improving our esti- mates of the number of moderate-mass stars that have already undergone evolution. NTT~ whose large collecting area will make it possible to obtain spectra of various subsystems of nearby galaxies, will reveal variations in abundances expected to develop as the result of different rates of evolution at different points within galaxies; the 25-Meter Millimeter-Wave Radio Telescope will permit determination of isotopic abundances of CO and other molecules with sufficient angular resolution to detect variations in abundances across the faces of galaxies; its beamwidth corresponds to 60 parsecs at M31. By yielding sharp images of large-red-shift galaxies, ST will permit them to be classified morphologically for the first time; this is essen- tial in order correctly to interpret observed correlations between the red shifts and other properties of galaxies. There have been tanta- lizing hints that certain types of galaxies are systematically bluer than their analogs nearby at lookback times as small as 3 billion to 6 billion years, but because morphological classification of these galaxies is beyond the capability of present ground-based telescopes, the inter- pretation of these results is unclear. NIT will play an important role in this research by permitting us to obtain optical and infrared spectra of galaxies at large red shifts, including those discovered by ST. interaction of Galaxies with Their Environment An important advance in studies of galactic evolution in the last decade has been the recognition of interactions between galaxies and neighboring galaxies and/or intergalactic gas. Exchange of mass, en- ergy, or angular momentum with the environment can modify gal- actic evolution in a variety of ways. It is widely believed that isolated spiral galaxies are formed with extensive halos, in which a substantial fraction of the mass is hidden in faint stars or in some other form, such as massive neutrinos. Galaxies located in binary pairs will in- teract with each other's halos in various ways if they are close enough together. For example, one galaxy can strip the halo material from the other by tidal forces; the material may either be redistributed within a common envelope or be entirely lost from the binary system. Since any energy or angular momentum lost in this way must come

52 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's from the binary orbit, the orbit must evolve. If one galaxy of the pair has a low mass, it may end up merging with the more massive galaxy, thus further increasing its mass; such a process is believed to account for the frequent occurrence of supermassive galaxies at the centers of rich clusters. The gravitational fields of such systems will be probed through measurements of the radial velocities of faint globular clusters with NIT, as well as through studies of the distri- bution of hot gas around galaxies with AXAF. There is evidence from x-ray observations of the Virgo cluster that the giant elliptical galaxy Messier 87 is accreting intergalactic gas. One model of the intense production of relativistic particles in the nucleus of M87 envisions the production of energy by the flow of matter into a massive black hole located in the galactic nucleus. Such accretion of intergalactic gas in clusters could provide a virtually unlimited source of matter to power the radio galaxies often found in clusters. Here AXAF, with its ability to observe the diffuse, hot gas with high angular resolution, increased sensitivity, and high spectral resolution, will be able to examine many radio galaxies at larger distances, for evidence of the M87 phenomenon. The whole problem of activity in the nuclei of galaxies poses a major puzzle in galactic evolution. Both elliptical and spiral galaxies (as radio and Seyfert galaxies, respectively) display activity that is highly localized in the nucleus of the galaxy. Possible physical ex- planations for such activity are discussed in the next section; here we mention those aspects of galactic activity that are relevant to galactic evolution in general. Do the extreme examples of galactic- scale explosions called quasars occur within galaxies? ST, with its high angular resolution, will be able to observe the parent galaxies if they are there; the light from a parent galaxy, which is lost in the glare of the quasar when studied with present ground-based tele- scopes, will be distinct from the quasar image when viewed with ST. Do many quasars occur within groups of galaxies, as recent ground- based observations of some quasars suggest? Again, ST, with its faint limiting magnitude, and NIT, with its ability to acquire spectra of faint objects, will answer this question even for distant quasars, thus determining whether the occurrence of a quasar in a galaxy depends on the environment of the galaxy. Is the density of stars in active galactic nuclei as high as required to explain such activity by stellar collisions? High-resolution pictures with ST will penetrate close to the centers of Seyfert nuclei and hence help to answer this question; its angular resolution of 0.05 arcsec corresponds to 25 parsecs at the distance of the nearest Seyfert galaxies. Is there an inward flow of

Frontiers of Astrophysics 53 hot gas in active ellipticals, as required by theories based on accretion by a massive black hole? The high-resolution images to be obtained by AXAF and ST, and studies of the time variability of high-energy emission to be carried out by AXAF, may help to answer this question as well. The instruments of the 1980's will have a major impact on the study of galactic evolution. Over the entire range of phenomena- from the earliest development of fluctuations in the Universe, through star formation in collapsing young galaxies, to the slow conversion of interstellar gas to stars and the emergence of active galactic nu- clei observations by the new instruments in the gamma-ray, x-ray, ultraviolet, optical, infrared, and radio regions of the spectrum will provide important new information. VIOLENT EVENTS Cosmic Rays, Supernovae, and Pulsars The Earth is constantly bombarded by cosmic rays, charged sub- atomic particles from space moving at relativistic speeds and with energies up to 102° electron volts (eV). The discovery of cosmic rays early in this century provided the first hints that the Universe is not a quiescent collection of stars and planets but, rather, the scene of violent events in which particles are accelerated to relativistic ener- gies. Until the 1950's, studies of cosmic rays were limited to those particles reaching the Earth; the bending of the trajectories of cosmic rays by interstellar magnetic fields precludes the identification of their sources by observing the directions from which they arrive at the Earth. A breakthrough in our understanding of the role of high-energy particles in cosmic processes occurred in the 1950's, when optical and radio astronomers discovered polarized emission from super- nova remnants and showed that it is synchrotron radiation from relativistic electrons accelerated and trapped in magnetic fields. Shortly thereafter, electrons were discovered in cosmic rays arriving at the Earth, in numbers agreeing with those required to explain as syn- chrotron radiation the nonthermal radio emission observed from the Milky Way. We now know that the acceleration of atomic nuclei and electrons to relativistic energies occurs in astronomical systems ranging in scale from the Sun and planets to giant radio galaxies. Particle acceleration results from motion of the magnetic fields embedded in the ionized

54 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's gases, or plasmas, that pervade interplanetary and interstellar space. In situ study of plasmas in the Earth's magnetosphere and in the interplanetary medium has shown that three processes accelerate charged particles: the passage of shock waves, the reconnection of magnetic fields of opposite polarity, and the compression of magnetic fields. Solar physicists have observed that magnetic fields emerging from the solar surface are occasionally forced into unstable configurations in which large amounts of magnetic energy are stored. Rapid recon- nection of the magnetic lines of force to form new configurations releases magnetic energy, much of which goes into accelerating large numbers of particles in a short time; the resulting high-energy par- ticles cause solar flares. Recently, the International Ultraviolet Ex- plorer (lUE) and the Einstein x-ray observatory have detected similar flare activity on other stars. Magnetic activity appears to be com- monplace, and in some stars it produces flares far more energetic than those on the Sun. Particle acceleration on a much larger scale accompanies the su- pernova explosions that mark the deaths of massive stars. Optical, radio, and x-ray observations of supernova remnants strongly sug- gest that most Galactic cosmic rays are initially accelerated in su- pernova explosions. Supernovae are also believed to be responsible for the synthesis of heavy elements, and thus they play a critical role in the chemical evolution of galaxies and, ultimately, in the origin of stars, of planets, and of life. There are at least two distinct types of supernovae. Type I super- novae, which may occur in low-mass binary star systems, are dis- cussed in the next section. Type II supernovae, which occur in mas- sive stars located in the disks of spiral galaxies, appear to be the natural consequence of the evolution of the cores of isolated massive stars. In successive stages of core contraction and heating, thermo- nuclear reactions produce heavier and heavier elements, until a core of about 1.4 solar masses of iron is accumulated. If an iron core contracts, because there are no further nuclear reactions to furnish energy that would stabilize it, it quickly goes into free-fall when the internal pressure drops as a consequence of the capture of electrons by atomic nuclei and the partial photodisintegration of iron-group nuclei into free nucleons and alpha-particles. If the core is sufficiently massive, no force is strong enough to reverse its huge inward mo- mentum, and the matter collapses to a point, forming a black hole. For cores of lower mass, as the central density approaches and even surpasses that of a typical atomic nucleus, the collapse may be

Frontiers of Astrophysics 55 halted by the repulsive component of the strong nuclear force. What happens then is still uncertain. One possibility is that the rebound of the collapsing core arising from this sudden "stiffening" of its inner regions initiates a shock wave that propagates out of the core and into the loosely attached envelope of the star, where it heats the material to such high temperatures that a sudden surge of re- actions occurs among the nuclei present. The elements heavier than iron thus created, together with other heavy elements produced by earlier nuclear burning nearer the surface of the star, are ejected by the shock wave into the interstellar medium. The process of core collapse releases about 1053 ergs, an amount of energy released if a star having a tenth of the mass of the Sun were completely annihilated. Most of this energy is lost in the form of energetic neutrinos and gravitational waves that cannot be cle- tected with current instrumentation. The visible light of the super- nova outburst comes from the shell of ejected material, which ex- pands outward at 10,000 km/see or more with a kinetic energy of about 1051 ergs. Some supernova explosions leave behind a compact remnant of 1.4 solar masses with a radius of only about 10 km-a star composed largely of neutrons, whose mean density is 10 times greater than that of a typical atomic nucleus. Conservation of angular momentum requires that a neutron star formed from the core of a star that is rotating, even if slowly, must rotate with a period measured in milliseconds. Compression of any magnetic field embedded in the original stellar core results in a field strength of 10~2 gauss or more. Some theorists have proposed that the electromagnetic energy available from the rapidly spinning, highly magnetized neutron star may be responsible for ejecting the outer layers of the star to form a supernova, but in any case, long after the explosion has taken place, intense electric fields are thereby generated, which can accel- erate particles to extremely high energies; they in turn emit beams of synchrotron radiation that sweep past the Earth each rotation period to produce the flashes that are characteristic of a pulsar. Hundreds of pulsars have been found with radio telescopes. The pulsar in the Crab nebula, which is the remnant of supernova 1054, has been observed over the entire electromagnetic spectrum from radio waves to gamma rays. The known origin of the nebula in a supernova explosion together with the indication from the observed synchrotron radiation that electrons are accelerated by the central pulsar strongly suggest that supernovae are a prime source of Gal- actic cosmic rays.

56 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's The stellar envelope ejected by the supernova explosion drives a shock wave hundreds of parsecs into the interstellar medium. Such shock waves appear to dominate the dynamics and thermodynamics of the interstellar medium, heating the gas to x-ray temperatures, accelerating interstellar clouds, destroying interstellar dust grains, and in some cases initiating the collapse of molecular clouds to form stars. They also compress the magnetized interstellar plasma and accelerate interstellar ions and electrons to relativistic energies. Radio and x-ray astronomers find evidence for this process in the fact that there is strong synchrotron radio emission from those regions where the x-ray emission shows that the gas must have been heated by a shock wave. Recent theoretical models based on scattering and ac- celeration of fast particles by turbulence in postshock plasmas can account for both the flux and energy spectrum of Galactic cosmic rays seen at the Earth. The isotopic composition of cosmic rays differs strikingly from that of materials in the solar system, suggesting that at least some cosmic rays originate directly in heavy-element-rich supernova envelopes. Measurements of the relative abundances of radioactive nuclei in cosmic rays, which have a variety of mean lifetimes, have shown that cosmic rays are contained for a few million years in the Galaxy; the constraints this puts on the required energy sources are consistent with acceleration in supernova shock waves. Binary Star Systems A large fraction of all stars is found in binary systems. Observations of the interactions between the two members of a binary system permit one to infer properties of the component stars that could not be determined otherwise. For example, precise timing of the radio pulses from a pulsar in a binary system has provided an accurate determination of the mass of the neutron star that causes the pulsar phenomenon; it has also revealed for the first time a slow decrease in orbital period, which theorists predicted to result from the emis- sion of gravitational radiation. Type I supernovae, whose optical luminosity decreases slowly with time, are probably descendants of white dwarfs in binary star systems of low mass. According to a currently favored model, they occur when the mass accreted from a companion star causes the mass of a white dwarf to exceed the Chandrasekhar limit of 1.4 solar masses, sending it into collapse and causing a thermonuclear explosion. The white dwarf is completely disrupted by such an event, which pro- duces about 1 solar mass of iron-group nuclei, principally the radio

Frontiers of Astrophysics 57 active nucleus 56Ni. The weak decays of 56Ni and of its daughter, 56Co, produce most of the luminosity of Type I supernovae and explain the exponential decline of their light curves. Observation of strong iron lines in the spectrum of supernova 1972e, indicating a large overabundance of iron, lends support to this idea; moreover, a feature attributed to radioactive 56Co has been tentatively identified. One of the most important developments of the 1970's was the discovery by the Small Astronomical Satellite-1, or Uhuru, spacecraft that compact, high-luminosity x-ray sources are located in binary star systems. Their x-ray emission results from the heat generated in the transfer of matter from the atmosphere of a nuclear-burning star to a close companion that is a collapsed star, such as a white dwarf, neutron star, or black hole. The infalling matter is believed to form an accretion disk a vortex of gas spiraling into the compact object having temperatures up to tens of millions of degrees. One such binary x-ray source, Cygnus X-1, has provided the best evidence for the detection of a stellar black hole a star that is col- lapsing to a gravitational singularity as predicted by Oppenheimer and Volkoff in 1939. The theory of stellar evolution predicts that collapse to a black hole is the likely fate of the core of a massive star, so that the Galaxy may well contain many millions of black holes. Although an isolated black hole cannot be detected with cur- rent techniques, a black hole that is accreting gas from a close binary companion star may be revealed by the x-ray emission of its accretion disk. The analysis of Cygnus X-1 shows that it is a binary system containing a compact object of about 10 solar masses; this object may be a black hole, as no satisfactory alternative interpretation of x-ray emission from the system has been found. Astronomers will continue to search for definitive evidence that Cygnus X-1 and possibly other binary x-ray systems actually do contain black holes. Many x-ray sources in binary systems were shown to be x-ray pulsars by satellite observatories operating during the 1970's. These systems contain rotating neutron stars with strong magnetic fields that channel the material accreting from a binary companion into narrow columns at the magnetic poles of the star, which are not aligned with the rotation axis. Thermal x rays from the hot polar regions are emitted in broad beams, which sweep around like those of a lighthouse, leading to the observation of x-ray pulses by distant observers. The interpretation of features in the high-energy spectra of some pulsars as electron cyclotron resonances implies that the magnetic fields involved exceed 10~2 gauss. Precise measurements of changes in the x-ray pulsation rates combined with optical measure

58 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's meets have yielded accurate values for the orbital parameters and the masses of the component stars; they have also provided insight into the dynamics of accretion flows and the properties of ultradense matter in neutron-star interiors. X-ray bursters form another major class of x-ray binaries. Found in the central regions of our Galaxy and at the centers of condensed globular clusters, these systems emit up to 1039 ergs of x rays in bursts lasting only 10 to 100 sec. These bursts appear to originate in weakly magnetic or nonmagnetic neutron stars that are members of close binary systems of low mass. Material accreted by the neutron star accumulates to a critical depth, then undergoes flash thermo- nuclear burning, releasing energy in a burst of thermal x rays; after several hours of renewed accumulation, a fresh layer of material produces the next flash. A dramatic discovery of the 1970's is the gamma-ray bursts bright, irregular flashes of gamma rays lasting only a few seconds. Although none of the sources of gamma-ray bursts has been firmly identified, their distribution over the sky suggests that most of them must be relatively nearby. However, the most intense gamma-ray burst ever recorded was observed on March 5, 1979, from the direction of a known supernova remnant in the Large Magellanic Cloud, an irreg- ular galaxy. The intrinsic luminosity of the burst, if it really originated at the 50,000-parsec distance of the Cloud, would be so great as to challenge explanation by any mechanism now known to astro- physics. Another startling discovery of the 1970's is the remarkable Galactic object SS 433, which apparently ejects more than an Earth-mass every year in two oppositely directed narrow jets of gas moving at ap- prox~nately one fourth the speed of light. Analysis of periodic changes in the optical emission-line spectrum indicates that the jets process like a spinning top, tracing out a complete cone every 164 days. This interpretation has been confirmed recently by high-resolution radio images of SS 433 obtained by the VLA and by very-long-baseline interferometry (WBl), which show the moving helical patterns sprayed into interstellar space by the processing jets. SS 433 is thought to be a binary system containing a compact object; however, the origin of the system, the mechanisms that accelerate and collimate the jets, and the cause of the precession are still unclear. Active Galaxies and Quasars The pioneers of radio astronomy discovered a source of unusual activity in the constellation Sagittarius. That source, Sagittarius A, is

Frontiers of Astrophysics 59 now known to be located at the exact center of our Galaxy. V~B! observations have shown that Sgr A is smaller than the solar system, although it is 70 times as luminous as the Crab nebula. Optical studies of the Galactic center region are prevented by interstellar obscuration, but observations of infrared emission, which penetrates the dust, show that it probably contains a massive cluster of red supergiant stars. High-resolution infrared spectra have revealed an emission line of ionized neon at 12.8-~m wavelength arising from clouds of ionized gas in the vicinity of Sgr A. If the Doppler broadening of this line is due to the revolution of the emitting gas about a central gravitating object, the region within about 1 parsec of the Galactic center must contain some 106 solar masses of material. While this may be ex- plained by a dense concentration of stars, its association with Sgr A makes it more likely that it is a single massive object, whose accretion disk produces the relativistic electrons responsible for the radio emis- sion of Sgr A. If so, the object may be a massive black hole. Gamma-ray detectors carried on balloons and on the High-Energy Astronomical Observatory-3 (HEAo-3) spacecraft have observed a strong emission line at 511 keV due to electron-positron annihilation at the position of the Galactic center. Recent HEAo-3 observations show that this line diminished in strength by more than a factor of 2 within a 6-month period, showing that the source of positrons must be extremely compact. Energetic phenomena similar to those at the center of the Galaxy are observed in other galaxies, but because of their much greater distances they are much more difficult to resolve spatially. For ex- ample, there is a spike in surface brightness at the nucleus of the Andromeda Galaxy, Messier 31, which is believed to be caused by a dense concentration of stars. Some spiral galaxies, called Seyfert galaxies, exhibit violent activity in their nuclei, including radio and x-ray emission, rapid motion of ionized gas, and powerful infrared emission concentrated within a region 100 parsecs or less across. The variability of the x-ray emission indicates that it originates in regions less than 1 parsec across. Some elliptical galaxies, called radio gal- axies, exhibit similar phenomena, and in addition display enormous radio lobes extending up to 500,000 parsecs into intergalactic space. These lobes appear to be fed by jets of relativistic particles emerging from an active galactic nucleus. For example, Messier 87, a giant elliptical radio galaxy in the Virgo cluster, has a bright, compact optical and x-ray source at its nucleus, from which a jet emerges. Optical studies indicate that there is a concentration of about 109 solar masses of material within its central 100 parsecs, a phenomenon that cannot be accounted for by a cluster of normal stars. The nucleus

60 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's of Centaurus A, another giant radio galaxy, emits a large fraction of its power as gamma rays. Quasars are starlike objects having large red shifts-the largest value so far observed is 3.5, corresponding to a recessional velocity of 91 percent of the speed of light. The first quasars were discovered through analysis of the spectra of optical counterparts of pointlike radio sources, but it is now clear that only a small fraction of quasars are strong radio sources. Many new quasars have been found re- cently through spectroscopy of the optical counterparts of faint x- ray sources discovered by the Einstein x-ray observatory. While their x-ray, radio, and infrared emission and their optical spectra quali- tatively resemble those of Seyfert-galaxy nuclei, quasars radiate much more power. VLBI radio images of the quasar 3C 273, showing expansion at velocities apparently greater than the speed of light. (Photo courtesy of M. H. Cohen, California Institute of Technology)

Frontiers of Astrophysics 61 The interpretation of the large red shifts of quasars has been de- bated for some time; however, the recent demonstration that many quasars lie in the same directions as groups of galaxies having the same red shift strongly supports the conclusion that quasars are really at the vast distances implied by a cosmological interpretation of their red shifts. At these distances, the energy released by quasars is equivalent to the complete conversion into energy of a solar mass of matter every year. Except for the big bang itself, quasars are the most powerful explosions in nature. Quasars resemble radio galaxies in that compact, active regions are connected by jets to distant radio lobes. The basic energy source of quasars must be located in the compact region (which may well be located in the nucleus of a galaxy); the relativistic particles gen- erated there stream out in jets to the outlying lobes. V~B! observations of the active regions of quasars reveal smaller, jetlike structures aligned with the much larger jets seen with conventional techniques; within these small jets, blobs appear to move outward with speeds that are close to that of light. In some quasars, the transverse velocity of these blobs apparently exceeds that of light; this observation can be explained if the blobs are moving nearly toward us at slightly less than the speed of light. Both the spectra and the time variation of the compact radio sources are consistent with impulsive injection of particles into, and subsequent rapid expansion of, magnetic trapping regions. X-rays have been detected from many quasars by the Einstein x- ray observatory. Many of the known quasars are strong x-ray sources, and new quasars have been discovered solely through their x-ray emission. The x-ray spectra are hard, implying that the x rays are either emitted thermally by hot plasmas or emitted nonthermally by relativistic electrons as synchrotron emission or inverse-Compton scattering. Some x-ray quasars vary on time scales as short as hours, so that the source regions can be no larger than the solar system, despite the fact that the power radiated in some cases exceeds 10~2 solar luminosities. The number of distant, x-ray emitting quasars is so large that they must account for a substantial fraction of the previously unresolved x-ray background radiation at energies of a few keV. At least one quasar, 3C 273, is known to be a strong gamma- ray source as well as an x-ray source. The optical spectra of quasars are rich in information, with power- law continua, many narrow absorption lines, and broad emission lines characteristic of rapidly moving clouds of hot gas. In many quasar spectra, the continuum varies with time but the emission

62 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's lines do not, suggesting that the continuum radiation originates in a much smaller volume. The emission-line red shifts are indicators of the cosmological distances to the quasars. Emission lines include the Lyman lines of hydrogen, which can be radiated by gas of high density, and coronal emission lines, normally radiated only by gas of low density. As the absorption-line systems usually have lower red shifts than the emission lines, they are attributed to intervening clouds of gas. In one class of absorption-line systems, whose red shifts are only moderately lower than those of the emission lines, heavy-element abundances are normal; such systems probably correspond to clouds of gas ejected by the quasar toward the observer. In the second class of absorption-line systems, the red shifts are considerably lower; as they contain no spectral lines other than those of hydrogen, these systems may arise in intervening intergalactic gas clouds having a primordial chemical composition. Theoretical models of quasars postulate a layered structure sur- rounding a compact central energy source. Furthest out are the clouds of cool gas that were ejected by the quasar at earlier epochs and that are now detected through observations of their heavy element-rich absorption lines. Inside is the region where the emission lines are formed. In order to explain the widths of these lines, the source plasma is presumed to be concentrated in clouds that orbit a central region containing a large mass. An intercloud medium, whose tem- perature is indicated by coronal emission lines to be a million degrees, prevents the clouds from expanding. At the center is a compact energy source responsible for the acceleration of particles and for the generation of gamma-ray, x-ray, optical, and perhaps infrared continua by a combination of thermal and nonthermal processes. Because of the enormous amounts of energy produced much of it in the form of relativistic particles-in such a small volume of space, the nature of the central energy source of quasars is of ex- ceptional interest. Among the possibilities are a compact cluster of neutron stars or stellar black holes or both undergoing frequent collisions, a massive plasma cloud stabilized by rotation, and an accretion disk formed from matter spiraling in toward a single mas- sive black hole containing 100 million solar masses or more. Do quasars and active galaxies really harbor supermassive black holes at their centers? If so, how did they form, and what is the source of matter that feeds the black hole? How is the gravitational energy of the accreting matter converted into the observed radiation? These are among the most exciting unsolved problems in astronomy.

Frontiers of Astrophysics 63 The Impact of Recommended Programs and Facilities on the Study of Violent Events Current theories of violent events in the cosmos, ranging from solar flares to quasars, present a variety of possible explanations for these events. As explained below, the data collected with the observational facilities recommended in this report will help to choose among the possibilities and thus to advance our understanding of these phe- nomena. The basic physical mechanisms that underlie the particle acceler- ation processes operating in solar flares-mechanisms believed to operate in many other astrophysical contexts will be investigated with a powerful assembly of instruments aboard the Advanced Solar Observatory (ASO) in space. The Solar Optical Telescope (SOT), which is at the heart of ASO, iS planned to be a 1-m-class diffraction-limited telescope that at 5000-A wavelength can resolve features as small as 0.1 arcsec, or 70 km on the Sun, thus yielding information of un- precedented resolution on magnetic-field configurations before and after solar flares. Extreme-ultraviolet (EUV) and x-ray telescopes of high spatial and spectral resolution on ASO will make measurements of temperatures, densities, and energetic particle populations in the flare plasma. Gamma-ray detectors on ASO will provide information on the numbers and interactions of high-energy particles accelerated by the flare; before ASO iS realized, the Gamma Ray Observatory (GRO) will be available to make measurements of the emission of gamma rays by solar flares. Solar flares radiate strongly in the ultraviolet (uv), the EUV, and the x-ray regions of the spectrum. Flares on other stars can be studied in these regions by ST, the Extreme Ultraviolet Explorer, and AXAF in the forthcoming decade. Elucidation of the nature of stellar flares and their relation to solar flares will require a concerted program of synoptic observations involving these satellites, as well as ground- based optical telescopes. Important questions regarding the sources, acceleration, and prop- agation of cosmic rays in the Galaxy will be addressed by new in- struments. The precise measurements of elemental and isotopic com- position of the cosmic rays as functions of energy, required to give insight into their origin and propagation in the Galaxy, will become possible with the deployment of sensitive detectors on long-duration flights as recommended here. The distribution and spectrum of cosmic rays in the Galaxy can be inferred from the detection by GRO of gamma rays resulting from the interaction of cosmic rays with the

64 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's interstellar gas and radiation fields. GRO studies of gamma-ray emis- sion from supernova remnants will provide crucial tests of theories for the origin of cosmic rays in supernova explosions and their ac- celeration by shock waves. Theories for the formation of heavy elements in supernova explo- sions will also be tested by GRO; the characteristics specified by the Space Science Board's Committee on Space Astronomy and Astro- physics (CSAA) in its 1979 report would permit it to detect gamma- ray emission lines from freshly synthesized nuclei in supernova ex- plosions out to the Virgo cluster, where a new supernova occurs about once every year. The dynamics of such explosions can be studied by observing the development of their x-ray and uv spectra with AXAF and ST, respectively. The phenomena displayed by compact stellar objects are so rich and varied that their study will require the concerted use of many of the proposed instruments. The VERB Array will observe the dy- namics of jets emerging from Galactic objects such as SS 433 with unprecedented precision. Because of its many independent baselines, it will observe the jets with an order of magnitude greater contrast than heretofore possible. The mechanisms for particle acceleration and radiation in pulsars will become clearer from observations of their x-ray and gamma-ray emissions with AXAF and GRO; for ex- ample, the observation of cyclotron lines leads to a direct determi- nation of the magnetic-field strength of the pulsar. AXAF and the X- Ray Timing Explorer (XTE) satellite will advance our understanding of accretion flows onto compact objects, the physics of matter and radiation in superstrong magnetic fields, and the interior structure of neutron stars by making possible studies of fainter objects, by yielding greater precision in determination of spectra, and by making possible millisecond timing studies of variable x-ray sources. Because of their much greater ability to detect faint sources of x rays and light, respectively, than previous systems, AXAF and ST will permit us to observe binary x-ray sources and optically identify them in nearby galaxies, providing clues to their origin and evolution. The phenomena of transient x-ray and gamma-ray sources will be observed with unprecedented sensitivity by GRO, as well as by XTE and other Explorer satellites. XTE will be able to detect x-ray bursts and transients quickly and to observe their spectral evolution in detail; this is a key clue to the mechanisms of x-ray bursts. An outstanding challenge is to detect and identify the optical and x-ray counterparts of the mysterious gamma-ray burst sources; this will require a concerted program of observations involving GRO, as well

Frontiers of Astrophysics 65 as XTE and other x-ray telescopes in space and optical telescopes on the ground. Many of the new instruments recommended in this report will have powerful capabilities for advancing our knowledge of quasars. Are they really located within galactic nuclei? As explained earlier, ST will answer this question by detecting the parent galaxies of qua- sars unambiguously. ST images will also permit classification of the quasar-associated galaxy as a spiral, elliptical, or other type, fur- nishing key information for theoretical quasar models. The increased sensitivity of ST will allow searches for small groups or clusters of galaxies surrounding quasars out to much larger red shifts; an optical telescope with the power of NIT will be required to measure their red shifts. It is not known why there are no known quasars with red shifts greater than 3.5. Did quasars simply not form until 2 billion or 3 billion years after the big bang, or are more distant ones being ob- scured by intervening dust? AXAF can detect quasars at red shifts greater than 3.5 in spite of any absorption by dust and determine their red shifts if the iron emission line is present in their x-ray spectra, and thus answer this puzzling question. Our understanding of the structures responsible for absorption lines in the spectra of quasars will advance greatly as a result of - observations of their uv absorption spectra with ST and the far- ultraviolet spectrograph in space. These instruments can detect very low column densities of gas in the outlying regions of intervening galaxies and clusters of galaxies or in intergalactic gas clouds. Such spectra of bright quasars will be good enough to determine physical conditions in the clouds. Knowledge of such conditions should de- termine whether the absorption-line systems arise from clouds ejected by the quasars or from intergalactic clouds; if the latter, observations in the far ultraviolet will yield the primordial helium abundance, a key datum for cosmology. ST and the far-ultraviolet spectrograph in space will yield a vast array of new information on the uv emission-line spectra of quasars, so far observed only in the brightest quasars using lUE and sounding rockets. It will be of great interest to find out whether the emission lines in highly red-shifted quasars are the same as those in nearby quasars observed in the optical region. If so, various lines that have been proposed as luminosity indicators can be confirmed and used to determine the distances of quasars at great red shifts, with im- portant applications to cosmology. The mystery of the central energy source that powers quasars and

66 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's active galaxies will be addressed in a new way by observations with GRO, SIRTF, and the 10-m submillimeter-wave antenna. For the first time, instruments operating in the gamma-ray region, the far in- frared, and the submillimeter spectrum will have sufficient sensitivity to determine the energy emitted in those bands by many quasars and galaxies. This is particularly important because much of the total emission may emerge primarily in those bands. If, as seems likely, the emissions at higher photon energies occur progressively closer to the central energy source, observations of rapid time variations of x-ray emission with AXAF will provide vital clues to the nature of the sources. Additional clues of a qualitatively different nature will be provided by observations at radio wavelengths. The VERB Array and its exten- sion into space, the first steps of which are recommended in this report, will provide the highest spatial resolution available at any wavelength, about 10-9 red. For objects at a red shift less than 0.1, this corresponds to only 1 light year, and hence there is a chance that one can directly image central radio sources, whose variability indicates sizes of that order. Such observations will provide direct evidence as to how particles are accelerated to high energy and ejected in beams that extend far into intergalactic space. FORMATION OF STARS AND PLANETS The Interstellar Medium Interstellar space is not empty. All along the Milky Way are dark clouds containing microscopic particles of interstellar dust. These clouds are often associated with luminous hot stars, whose lifetimes are so short that they must have formed from the clouds themselves. The interstellar medium is replenished by mass loss from stars as they evolve and die and is thus an essential link in the cycle of stellar death and rebirth that determines the structure and evolution of galaxies. Between the dark clouds is a relatively thin gas, which in the vicinity of hot stars is ionized and heated (forming H ~ regions), so that it emits a bright emission-line spectrum. This spectrum shows that the interstellar medium contains hydrogen and helium, along with traces of heavier elements like carbon, nitrogen, and oxygen. The 21-cm spectral line of neutral atomic hydrogen (H ~) has been used to map its distribution throughout our own Galaxy and nearby spiral galaxies. On a Galactic scale, the H ~ forms spiral arms that

Frontiers of Astrophysics 67 coincide with those outlined by dark clouds and young stars. When observers found that individual dark clouds often contain little H I, theorists suggested that the H ~ in such clouds has reacted on the surfaces of dust grains to form H2 molecules. Evidence for substantial concentrations of H2 molecules in dark clouds began to accumulate after 1968 through observations of the microwave spectra of am- monia, carbon monoxide, and other molecules. These data showed that the total density in dark clouds is much greater than that derived from 21-cm observations of H I, strongly implying that molecular hydrogen (H2) is the dominant material in these clouds. The Copernicus ultraviolet satellite in 1973 confirmed the wide- spread presence of H2, at least in regions translucent to starlight, by direct measurements of ultraviolet absorption lines formed by inter- stellar H2 in the spectra of distant stars. As predicted, the presence of H2 is correlated with that of dust. Roughly half of the interstellar hydrogen in our Galaxy (about 5 x 109 solar masses) is H I; the rest is H2. The far-ultraviolet spectrograph in space recommended in this report will permit extension of the study of H2 to much greater distances and to darker clouds. Instruments carried aloft by aircraft and balloons will soon make the first observations of the fundamental rotational transition of H2 at 28.2-~m wavelength. It has been cal- culated that the enormous sensitivity of a cryogenically cooled in- frared telescope in space, such as SIRTF, equipped with array detectors tuned to the 28-~m line, will permit the mapping of interstellar H2 just as interstellar H ~ has been mapped in the 21-cm line. Satellite observations have shown that a broad absorption band at 2200-A wavelength is present throughout interstellar space. This band is believed to be due to interstellar dust, perhaps carbon in the form of graphite; if so, its strength indicates that a substantial fraction of interstellar carbon is locked up in dust. Complementary observations by Copernicus determined the gas-phase abundances of many atoms and ions from their ultraviolet absorption spectra. Atomic carbon i somewhat depleted in the gas phase, consistent with the interpre- tation of the 2200-A band; other elements, including silicon, mag- nesium, and iron, are depleted by large factors, suggesting that they too are locked up in dust grains. That at least some of the interstellar dust is actually composed of these elements has been demonstrated by infrared observations, which reveal strong absorption bands at 10- and 20-~m wavelength, which are characteristic of magnesium- iron silicates. so and the far-ultraviolet spectrograph in space will extend these studies by covering virtually all of the resonance spectral lines of abundant atoms and ions. The great increase in sensitivity,

68 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's due to greater aperture, greater multiplexing capability, or both than that of previous instruments, will permit the study of interstellar matter along the lines of sight to more distant and more heavily obscured stars. With Copernicus, astronomers discovered a previously unknown hot (2 x 105 to 5 x 105 K) component of the interstellar medium, in which oxygen is ionized five times (O vie. This medium appears to form a halo around the Galaxy. By reaching stars at great distances, observations of O vat with the far-ultraviolet spectrograph in space will permit us to map the entire gaseous halo of the Galaxy. Since the late 1960's, radio astronomers have discovered a wide variety of molecules in the interstellar medium. Most of them are concentrated in relatively dense and dark clouds composed primarily of H2 molecules and dust, called molecular clouds. U.S. radio as- tronomers have led the world in the study of molecular clouds as a result of the availability of suitable radio antennas and good milli- meter-wave receivers; the 11-m millimeter-wave antenna on Kitt Peak operated by the National Radio Astronomy Observatory has played a particularly important role in such studies. The logical next step in this field will be the construction of a 25-Meter Millimeter-Wave Radio Telescope as recommended in an earlier form in the Greenstein report; its larger aperture, more accurate surface, and drier site will permit studies at higher sensitivity, at higher angular resolution, and at much shorter wavelengths (1 mm). Complementing this step, the construction of a 10-m submillimeter-wave radio antenna at an ex- cellent site, as recommended here, will permit the observation of interstellar molecules in selected atmospheric transmission windows down to 340 Em or less with angular resolution approaching 8.5 arcsec. The CO molecule has been mapped throughout the Galaxy; it is largely confined to spiral arms and is well correlated with interstellar dust. Photographs and CO maps reveal complexes of molecular clouds up to 50 parsecs long. These complexes, of which there are estimated to be about 1000 in the Galaxy, are usually elongated parallel to the Galactic plane and contain individual clouds 10 to 20 parsecs long. Within individual clouds are found regions of high density ("cloud cores") a parsec or less in diameter. It is in such regions that star formation is taking place. Interstellar dust is thought to originate in the ejecta of red giants and novae, for These objects often display emission bands charac- teristic of silicate dust. Apparently the dust condenses as the gas expands and cools; when it is subsequently heated by radiation from

Frontiers of Astrophysics 69 the nearby star, it emits the infrared radiation that we observe. Infrared surveys have discovered a variety of sources relevant to star formation, including some that have no optical counterparts but that emit microwave continuum radiation and extremely intense spectral lines of OH and H2O molecules. The energy source for these objects is a luminous hot star hidden from view by a nearby, dense molecular cloud. Its ultraviolet radiation is largely absorbed by a "cocoon" of dust, which re-emits most of the energy in the infrared. Some of the ultraviolet radiation heats and ionizes the accompanying hydrogen, which emits in the radio domain. The OH and H2O line emissions are produced by natural maser action, the molecular levels being inverted by collisions, infrared transitions, or both. Dust cocoons are invariably found in the densest parts of molecular clouds, supporting the view that the hot stars responsible for their heating were formed recently from the surrounding cloud. With newly developed infrared spatial interferometric techniques, it is now possible to map the formation of dust around many such stars. Studies of x-ray emission also contribute to an understanding of the interstellar medium. Rocket surveys have mapped the distribu- tion of nearby interstellar gas at million-degree temperatures, leading to a model in which the solar system is now surrounded by an ancient supernova shock wave. The Einstein x-ray observatory has mapped x rays from the gas heated by distant supernova explosions. In the future, the increased energy resolution available with AXAF will per- mit observations of individual emission lines, thereby permitting the detailed quantitative analyses of supernova shock waves and proc- esses occurring behind them, including the destruction of interstellar dust grains by the surrounding, shock-heated gas. The Small Astro- nomical Satellite-2 (SAS-2) and COS-B satellite have mapped the gamma rays resulting from the decay of pi mesons produced by collisions of cosmic rays with interstellar riatter, permitting the distribution of cosmic rays in the Galaxy to be surveyed and the masses of a few interstellar clouds to be determined. The higher sensitivity and an- gular resolution to be provided by GRO will permit many more of these clouds to be studied at gamma-ray wavelengths. Our current understanding of the interstellar medium is based on the fact that the energy radiated by interstellar matter must be sup- plied by the absorption of energy from supernova shock waves, from ultraviolet radiation from hot stars, and from interactions with cosmic rays. Diffuse clouds of atomic hydrogen are maintained at about 100 K by the ejection of energetic electrons from dust grains by ultraviolet stellar photons. Molecular clouds are much colder, on the

70 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's order of 15 K, because ultraviolet photons cannot penetrate the large amounts of dust present, leaving cosmic-ray interactions as the only source of heat, and because the molecules in such clouds are efficient radiators even at low temperatures. The hot intercloud medium re- vealed by Copernicus observations is maintained at a temperature of several hundred thousand degrees by shock waves from supernovae. The interstellar medium is thus composed of at least three phases having widely differing temperatures. What are the relations between these phases? How do they exchange mass and energy? The capa- bility of the optical, ultraviolet, radio, infrared, and x-ray telescopes of the future to observe all the relevant energy inputs and outputs will greatly increase our understanding of these questions. Molecular Clouds and Star Formation The fundamental rotational transition of the carbon monoxide mol- ecule CO at a wavelength of 2.6 mm is a widely used tracer of the distribution of molecular clouds in the Galaxy. An important frontier for future research is to understand the distribution of such clouds in other galaxies; the 25-Meter Millimeter-Wave Radio Telescope (at 2.6-mm and 1.3-mm wavelengths) and the 10-m submillimeter-wave antenna (at the wavelengths of higher CO rotational lines) will permit such studies with spatial resolutions better than 60 parsecs at Messier 31. A key theoretical problem is the mechanical equilibrium of molec- ular clouds. Observations of CO and other molecules lead to esti- mates of the total density and thus the gravitational forces within a cloud. The densities and temperatures derived from CO yield esti- mates of the gas pressure. In most clouds gravitational forces are much stronger than pressure forces, so that one would expect them to collapse nearly in free fall, a process that should take a few million years. If such collapses lead to star formation, the number of young stars in the Galaxy would be far larger than observed. It is therefore believed that molecular clouds are perhaps prevented from collapsing by internal turbulence, which is known to be present from the ob- served velocity broadening of molecular lines. Although in many clouds the force exerted by turbulent pressure is adequate to resist gravitational collapse, the turbulence must be highly supersonic and should dissipate rapidly through shock heating. Dissipation can be reduced if molecular clouds are actually composed of large numbers of subclouds, too small to resolve with present instruments. The 25

Frontiers of Astrophysics 71 Meter Millimeter-Wave Radio Telescope and the submillimeter-wave radio antenna may permit us to resolve such subclouds. The chemistry of molecular clouds determines their cooling rate and thus their ability to collapse. Chemical reactions among the elements C, N. O. and H under the nonequilibrium conditions that apply in interstellar clouds are of interest in their own right and also because of their possible bearing on the origin of life. Synthesis of molecules in space is thought to begin with the formation of H2 on the surfaces of dust grains and the subsequent ionization of H2 by cosmic rays to form H2 . Collisions of H2 and H2 lead to H3, which reacts with C, N. and O to produce species such as HCO+, CN, HCN, and H2CO. The complete chain of reactions will be much better understood by observing molecules that have been predicted theoretically, using the increased wavelength coverage of the 25- Meter Millimeter-Wave Radio Telescope and the 10-m submillimeter- wave radio antenna. Massive stars form in the dense cores of molecular clouds, as indicated by the observation of infrared sources, H2O and OH ma- sers, and microwave-continuum sources in these regions. According to one theory, dense cores of clouds form when the ultraviolet ra A 12.4-,um map of the central region of the Orion molecular cloud, including the Becklin- Neugebauer "protostar" and the Kleinmann-Low nebula, made with the Goddard Space Flight Center infrared CID array camera. (Photo courtesy of the National Aeronautics and Space Administration, the University of Arizona, and the Harvard-Smithsonian Center for Astro- physics)

72 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's diation from massive stars of a previous generation ionize that part of the parent molecular cloud nearest to them, raising the gas pres- sure and driving a shock wave into the cloud. The shock compression raises the density above the critical value for gravitational instability, and the gas contracts to form a core. Part of the core collapses to form a new generation of massive stars, and the cycle is repeated. In this way, a wave of successive star-formation events propagates through an elongated molecular-cloud complex. This picture is sup- ported by the fact that associations of young stars are observed outside the molecular cloud and aligned with it; the younger asso- ciations are nearer the cloud and the older ones further away, as expected from the model. Molecular-cloud cores containing massive young stars are often found immediately adjacent to H ~ regions. The increased angular resolution available with the 25-Meter Milli- meter-Wave Radio Telescope and the 10-m submillimeter-wave radio antenna will help to clarify the spatial relationships involved. The Solar System The NASA program of solar-system exploration by deep-space probes is one of the grand technological and scientific adventures of our generation. The opportunities to fly by or to orbit planetary bodies, to study their properties at short range, to make detailed measure- ments of their atmospheres and magnetospheres, and in some cases actually to land upon their surfaces, have given birth to the new field of planetary science, which draws on a number of scientific disciplines in addition to planetary astronomy. These studies relate strongly to the rest of astronomy and astrophysics. For example, the study of planetary atmospheres has led to the development of con- cepts and techniques that apply more broadly to the dynamics of stellar atmospheres and the interstellar medium; in situ measure- ments of planetary magnetospheres have provided new insights into the processes of magnetic-field reconnection and particle acceleration; the theory of the internal heating of Jupiter-like planets has conse- quences for more general studies of the collapse and structure of low-mass protostellar and protoplanetary objects; and studies of com- ets, meteorites, and other primitive bodies of the solar system have provided insights into the chemical composition of the original solar nebula that may well apply to interstellar clouds generally. The mechanisms that led to the formation of the Sun and its planets must be at work in the Galaxy today. Much of what we can learn about the formation of our solar system, both from the NASA plan

Frontiers of Astrophysics 73 etary-exploration program and from ground-based planetary astron- omy, may thus be applied to the more general problem of planetary formation. These studies have, during the 1970's, revolutionized our knowledge of the solar system; here we touch on only a few high- lights. Exploration of the Moon has shown that it is now inactive; its surface is the result of meteoritic cratering over its 4.5-billion-year lifetime. Mercury is also inactive, but Venus, like the Earth, shows the effects of tectonic activity great uplifts and a very dense at- mosphere, presumably of secondary origin. The Viking mission showed that Mars is at present inactive and that only small amounts of water are now present on the surface. On the other hand, images from the Viking orbiter demonstrate that Mars was once active and may have been washed by water at one time. It has recently become possible to classify asteroids by their in- frared reflection spectra. The evidence suggests that an important class of meteorites, the carbonaceous chondrites, must be derived from certain classes of asteroids, presumably through collisions that injected them into Earth-crossing orbits. Because of the large col- lecting area of NIT, it will extend studies of infrared reflection spectra to much smaller objects. The Voyager program has revolutionized our knowledge of Jupiter, Saturn, and their magnetospheres and satellites. With tremendous detail, it has revealed complex motions in Jupiter's deep atmosphere, verified its hydrogen-rich composition, and discovered enormous atmospheric lightning flashes. The Galilean satellites were found to be complex and completely different from one another; lo is heated to the melting point by Jupiter's tidal force and emits sulfur ions into the Jovian magnetosphere from a number of active volcanos. Future infrared telescopes on Earth and in Earth orbit will contribute to further understanding of all the planets by imaging and spectrally analyzing their infrared emission. NTT will speed up spectroscopy in the near infrared, while the LDR, because of its large collecting area and high angular resolution at 20-llm wavelength and longer (better than 1 arcsec) will permit high-resolution infrared spectroscopy of planets with many resolution elements over the disk. ST will have a unique capability to image Jupiter and its satellites regularly from Earth orbit with the same 150-km spatial resolution achieved only briefly by Voyager; in addition, ST will spectroscopically analyze plan- etary atmospheres using the reflected ultraviolet light from the Sun and the light emitted from upper atmospheric layers. We will gain a better knowledge of the complex chemistry occurring in comets by

74 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's studying their ultraviolet reflection spectra with ST, and their mo- lecular spectra with the 25-m Millimeter-Wave Radio Telescope and the 10-meter submillimeter-wave radio antenna. Current theories of the origin of the solar system postulate a ''solar nebula," a disk of gas and dust that formed together with the Sun nearly 5 billion years ago. It is believed that dust particles (either interstellar particles brought in when the solar nebula formed or particles that condensed later out of gases in a heated nebula) mi- grated to its midplane, where gravitational attraction drew them together to form the planetesimals, which, through collisions, ac- cumulated to form the terrestrial planets and the cores of the outer planets. This model has been severely constrained by the recent discovery of isotopic anomalies in meteorites. Not only must the whole process have occurred faster then previously thought possible, but 26A1 must have been injected just before the process started. Since both the supernova explosion responsible for producing the 26A1 and the origin of the solar system were rare events, it seems likely that they were causally related, perhaps because a nearby supernova explosion triggered the collapse of a molecular cloud. Observers have recently reported configurations of young stars that may have formed as the result of compression by such a supernova shock wave. Roles of Theory and Observation The formation of stars and planets is a formidable theoretical prob- lem. What mechanism stabilizes molecular clouds? Are supernova shock waves required to make them unstable toward collapse? Why do they fragment into objects of stellar mass as they collapse, and what determines the distribution of masses? How does the angular momentum observed in the solar system and in binary-star systems originate? What role do magnetic fields play in the collapse and fragmentation process? Why do some fragments become planetary systems, while others become binary-star systems or isolated stars? What determines the distribution of planetary masses and distances? Do satellites form at the same time as the planets, or later? Such questions are now receiving serious theoretical study. Full two- and three-dimensional hydrodynamic computer codes, taking proper account of the effects of turbulence and magnetic fields, are needed for satisfactory answers. Such codes, originally developed for other applications, can now be implemented only on supercom- puters, but the combination of array processors with the smaller

Frontiers of Astrophysics 75 computers recommended in this report may also be able to handle hydrodynamic codes for the formation of stars and planets. Future observational studies should aim to constrain theoretical models at every stage. The question of turbulent stabilization of molecular clouds requires the greatest possible angular resolution in order to determine whether the observed velocity broadening is ac- tually due to the overlapping velocity shifts of various subclouds. Both the 25-Meter Millimeter-Wave Radio Telescope and the 10-m submillimeter-wave radio antenna offer dramatic improvements in this area, providing spatial resolutions down to 0.02 parsec in the Orion molecular cloud. In the study of dense cores of molecular clouds, it is important to obtain data in the spectral lines of a number of different molecular species that depend differently on temperature and pressure and, because of differing saturation effects, arise in different parts of the core. The 25-Meter Millimeter-Wave Radio Telescope and the 10-m submillimeter-wave radio antenna will be able to detect many more lines because of the extended frequency range. It is also important to extend the search for young stars in these cores to much fainter limits than is now possible, in order to assess whether fragmentation results in simultaneous formation of a range of stellar masses. SIRTF, with its 1000-fold increase in infrared sensitivity, will be crucial in detecting much fainter stars. Because of its relation to complex molecules, the transformations of carbon are of particular interest; in diffuse clouds, carbon may be divided among neutral atoms, ions, and dust particles. The 10-m submillimeter-wave radio antenna will permit the mapping of neutral atomic carbon, the only atom other than H ~ for which this is possible. In molecular clouds, extension of the accessible frequency range using the 25-Meter Millimeter-Wave Radio Telescope and the 10-m submillimeter-wave radio antenna will bring many new organic mol- ecules, as well as excited states of known molecules such as CO and HCN, under study. In the vicinity of young stars and of shock waves formed in cloud cores, the temperature is high enough to excite submillimeter and infrared lines of molecules. The 10-m submilli- meter-wave radio antenna will permit spectroscopic observations of many such lines at wavelengths down to 340-~m wavelength, while optical-infrared telescopes, intruding NIT, will work up to 20-~m wavelength. The far-infrared range between these wavelengths can be studied with moderate-sized instruments on the Kuiper Airborne Observatory and on balloons, and the LDR will be able to do high- spectral-resolution work at arcsecond resolution on these objects.

76 ASTRONOMY AND ASTROPHYSICS FOR THE 1980 s A protoplanetary disk like that of the solar nebula, located at the 170-parsec distance of the Ophiuchus dark cloud, would subtend 0.3 arcsec if it is 25 astronomical units in radius. If its temperature is 100 K, it would radiate in the 20-50-~m wavelength range, with a total luminosity about equal to the Sun. Such an object is beyond current ability to detect but would be easily detected by STRTF. A LDR in the 10-m class would be able to study the spectra of such objects because of its large collecting area; it would achieve some spatial resolution on the object when operated with low spectral resolution. If current ideas are correct, planetary systems are ubiquitous. So far, however, attempts to confirm the astrometric detection of plan- etary-sized objects orbiting Barnard's star have failed, leaving us with no conclusive evidence for the existence of any planetary system but our own. This problem is of such importance that new efforts should be made to solve it, as explained later in this chapter in the section on Planets, Life, and Intelligence. SOLAR AND STELLAR ACTIVITY Activity on the Sun It has been known for a century that there are gigantic eruptions on the Sun, with conspicuous terrestrial consequences like the Aurora Borealis and radio blackouts. The solar atmosphere is the site of intensely energetic and extraordinarily complex activity, much of which is still not thoroughly understood. Insight into these phenom- ena is important for our understanding not only of the Sun but also of other stars, as well as of plasma processes in a wide variety of astronomical objects. The magnetic field of sunspots was discovered 70 years ago. The solar corona has long been known from observations during total eclipses of the Sun, but not until 40 years ago was it demonstrated that its temperature is a million degrees. Gradually, as instruments improved, additional features of solar activity were discovered, in- cluding eruptive prominences, small flares that appear in newly emerging active regions, and occasional large flares. It was found that flares are the source of highly energetic particles, called solar cosmic rays. The general magnetic field of the Sun was discovered, mapped, and followed throughout the entire 11-year sunspot cycle using the newly invented magnetograph. The intimate connection of magnetic fields with solar activity super-heated gases, fast par- ticles, violent eruptions, and flares-became clear.

Frontiers of Astrophysics 77 The motion of material in comet tails provided the first hint that the solar corona is expanding. Experiments on Earth-orbiting space- craft revealed that the expansion of the corona causes a supersonic solar wind that carries away a million tons of solar material every second at velocities between 300 and 600 km/sec. In the 1970's, in- struments aboard Skylab discovered that the corona erupts every few hours, and x-ray studies from that satellite outlined in dramatic detail the hot, dense regions of the active corona and their association with the solar magnetic field, as well as the cooler, more rarefied regions called coronal holes, from which high-speed solar-wind streams flow. Skylab also discovered x-ray bright points, from which most of the magnetic flux of the Sun seems to emanate. The solar wind extends far into space, forming a "heliosphere" within the surrounding interstellar medium, which is a vast sea of plasma activity perhaps several hundred astronomical units across. It is filled with fast and slow streams of solar wind, interacting violently to produce shocks and fast particles, and is swept constantly by blast waves from solar flares. Interaction of this plasma with planetary magnetospheres produces planetary magnetic storms and radiation belts and accelerates particles to high energy near the planets. The subject of intensive study through direct observations and the- oretical modeling, the heliosphere also serves as a guide to the com- plex phenomena that must also occur near other stars, although they are too distant for their tenuous plasma envelopes to be observed. In recent years, historical studies have revealed that the activity of the Sun may fall to a low level for a century at a time, as it did during the fifteenth and seventeenth centuries, or rise to a very high level, as it did in the twelfth century. The climate of the Earth appears to respond to these long-term variations, with the mean annual temperature in the Earth's North Temperate Zone declining in ex- tended periods of low activity. The causes of such long-term solar . . . var~ahons remain a mystery. The scope of solar activity is impressive; nothing like it is known in the terrestrial laboratory. How are magnetic fields produced inside the Sun? How do these fields cause flares, eruptive prominences, solar cosmic rays, and sunspots? How do they control and perhaps heat the active corona? And why does the corona expand to produce the solar wind? Considerable progress has been made over the past three decades in answering some of these questions, while others have evaded understanding to this day. Detailed observations of the magnetic fields at the surface of the Sun with the precision vacuum tower telescopes and high-resolution

78 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's magnetographs and spectrographs currently available have demon- strated that the solar magnetic field is concentrated into individual flux tubes in which the field is intense; incredibly, these tubes do not expand despite the huge internal magnetic pressure. The for- mation of sunspots, plages, and flares and the heating of the corona are profoundly affected by this seemingly unnatural behavior. Re- newed efforts with infrared, optical, ultraviolet, x-ray, and gamma- ray telescopes, together with determined theoretical studies, will be necessary for progress toward understanding some of the more com- plex aspects of solar activity during the 1980's. Instruments aboard the Solar Maximum Mission carried out the first precise, coordinated studies of flares in ultraviolet light and x rays, observing particularly the extremely compact and intense cen- ters of activity with temperatures above one hundred million degrees that lie at the heart of solar flares. The Solar Optical Telescope (SOT), a diffraction-limited meter-class telescope to be launched on the Space Shuttle during the 1980's, will resolve detail down to 0.1 arcsec, or 70 km on the Sun. For the first time, we will be able to see the structure of the most slender magnetic tubes and the details of the convection around them, as well as the glowing hydrogen filaments above. The SOT will be followed by the Advanced Solar Observatory (ASO) in space, which will provide better than 0.1-arcsec resolution images simultaneously at optical, ultraviolet, and soft x-ray wave- lengths. These observations should help to reveal the basic physics of a great variety of phenomena on the Sun, including plages, sun- spots and their penumbrae, the breakup of magnetic fields into dis- tinct flux tubes, and related activity at the boundaries of convective cells in the solar atmosphere, because many of these phenomena are expected to show important details as one exceeds the present ground- based resolution limit of about 1 arcsec. Because of the tremendous temperature range from the photosphere (6000 K) to the corona (2 x 106 K), simultaneous observations in all three spectral ranges are needed to establish the connections between phenomena seen in different atmospheric layers. Such insights will provide tests of present and future theoretical ideas that are in many cases much more stringent than can be furnished by the best observations from the ground. The ASO recommended in this report will be designed to obtain the most detailed observations of the solar surface that can be made from Earth orbit with a mission in the moderate-cost class. Spacecraft launched as part of the International Solar Polar Mission (ISPM) will add an important new dimension to our knowledge of the Sun and

Frontiers of Astrophysics of the heliosphere by making observations out of the ecliptic plane. The desirability of obtaining extremely detailed observations of the solar surface raises an exciting possibility a probe to the Sun itself. NASA'S proposed Star Probe mission, planned to carry out the first in situ exploration of any star, would come within 2 million kilometers of the solar surface; if it were to carry a 10-cm telescope, it would be able to resolve details of solar activity down to 10 km. The launch of such a mission could open up a new era of investigation in solar physics, much as NASA'S program of planetary exploration by deep- space probes has broadened and supplemented the traditional field of planetary astronomy. Studies of the solar core will receive renewed impetus during the 1980's. It was to probe the deep interior of the Sun that neutrino astronomy was born a decade ago with the assembly underground of a 37Cl neutrino detector in the Homestake Mine in South Dakota. The flux of solar neutrinos detected by this experiment is less than one third as intense as predicted by current theories of stellar energy 79 A large solar flare photographed in hydrogen light. The combed appearance of the solar gases is due to the strong magnetic fields in the flare region. (Photo courtesy of the Association of Universities for Research in Astronomy, Inc., Sacramento Peak Observatory)

80 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's generation and internal structure. This result has already led to the revision of models of the solar interior; in view of the magnitude of the discrepancy, however, a detector incorporating large quantities of gallium, which is sensitive to the full range of neutrino energies, will be needed to check our understanding of energy production in the Sun. More accurate laboratory measurements of the relevant nuclear reaction rates are also needed to refine the value of the solar- neutrino flux predicted by theory. The study of long-period oscillations of the surface of the Sun, which arise from waves trapped beneath the surface, provides insight into the properties of the solar layers between the atmosphere and the core. It was recognized during the 1970's that the solar 5-min oscillations are global phenomena that can be used to probe the structure and dynamical behavior of the solar convection zone hidden beneath the surface. Recent observations of these oscillations from a station located in long-term sunlight near the South Pole of the Earth have led to precise determinations of the oscillation frequen- cies. However, the study of solar dynamics over much longer time periods than the polar observations can provide is essential to the successful completion of such studies. This goal will require mea- surements from an Earth-orbiting spacecraft, such as a proposed Solar Dynamics Explorer satellite. Stellar Activity At about the same time that the magnetic field of the Sun was being mapped by magnetographs, similar instruments showed that many other stars also have magnetic fields; they range up to a hundred or a thousand times more intense than that of the Sun. Strong fields covering much of the surface have been detected directly only in stars more massive than the Sun, but recent evidence implies that less massive stars also have magnetic fields. Observations of chro- mospheric lines in less massive stars indicate that they, like the Sun, are surrounded by superheated gases, suggesting that they also have active regions and therefore localized magnetic fields. Many of these stars display magnetic cycles, during which stellar activity waxes and wanes with a regular period like the sunspot cycle. Ultraviolet and x-ray astronomy from space vehicles led to a break- through in stellar-activity studies. Observations from the Copernicus ultraviolet observatory showed that some stars possess stellar winds; in some cases, the wind is so strong that it must have profound effects on the evolution of the star. In addition, stellar winds carry

Frontiers of Astrophysics away so much angular momentum that the rate at which the stars rotate must be slowed, over hundreds of millions of years, to low values. The International Ultraviolet Explorer (lUE) satellite extended these studies to a much greater variety of stellar types. Observations with the Einstein x-ray observatory revealed that stars of nearly all types are surrounded by coronas at temperatures reach- ing millions of degrees, implying the existence of magnetic fields and hence active regions, stellar winds, and perhaps flares. Some stars exhibit such large "starspots" that one entire face of the star is dimmed; others exhibit flare activity thousands of times more intense than that on the Sun. The Sun serves as a local laboratory where detailed work can be done to understand the individual physical effects that contribute to 81 Einstein x-ray image of the young star cluster The Hyodes. (Photo courtesy of R. Stern and J. Underwood, Jet Propulsion Laboratory, and M. Zolcinski and S. Antiochos, Stanford University)

82 ASTRONOMY AND ASTROPHYSICS FOR THE 1980 s this richly varied activity. The exotic and unanticipated properties of magnetic fields in convective fluids are legion; we have only begun to discover what they are and how they fit together to produce solar activity. The Sun offers a close look at the complex detail of magnetic activity, whereas the surfaces of distant stars cannot be studied in detail because stars appear only as points of light in the largest telescopes. On the other hand, recent observations show that activity can be identified and analyzed in other stars using the radiation from the whole star. While a coordinated attack on problems of solar activity is being mounted, observations of activity in other stars should also be pressed forward with ground-based telescopes, as well as ultraviolet and x- ray telescopes in Earth orbit, including ST, AXAF, and the far-ultra- violet spectrograph in space. Each of these instruments will have far greater sensitivity than its predecessors and, therefore, will be able to study far more stars than have been possible with earlier instru- ments. The goal is to reach stars with many different characteristics; theoretical predictions of the dependence of activity on the age and rotation period of a star can be tested much better with studies of many stars than they can if only the Sun is studied. \ The Role of Magnetic Fields Magnetic fields lie at the heart of solar and stellar activity. They are somehow produced within most stars, as well as within several of the planets of the solar system. In the Sun, the field floats up to the surface, where it produces sunspots. In dissipating, the field heats the plages and the corona; if it does so rapidly enough, it causes explosions or eruptions, sometimes accompanied by superheated flares. The origin of magnetic fields in the interior of the Sun and other stars poses a fundamental problem to physics. It has been suggested that magnetic fields were trapped in the interiors of stars at the time they formed from interstellar gas. According to this suggestion, the field slowly rises to the surface of the star over its lifetime. However, the regular 11-year reversal of magnetic field in the Sun, as well as the cyclic variations observed in other stars, appears to require an- other explanation. Any primordial magnetic field in the Earth must have dissipated long ago; the Earth's magnetic field that we now observe must be continually regenerated by the flow of electrically conducting fluid in the core. Just as the convection and nonuniform rotation of the

Frontiers of Astrophysics liquid iron core is thus responsible for the maintenance of the field of the Earth, convection and nonuniform rotation in the solar con- vective zone can maintain the field of the Sun. The generation of magnetic fields is described by the equations of magnetohydrodynamics. Mathematical solutions of these equations incorporating reasonable estimates for the fluid velocities within the convective zone predict magnetic fields remarkably like the actual magnetic fields of the Sun. On the assumption that the angular velocity of solar gases increases with depth, the equations predict that an east-west magnetic-field component develops first at middle and high solar latitudes and then migrates toward the equator, where the oppositely directed fields from opposite hemispheres meet and cancel each other. The north-south fields are predicted to reverse direction roughly when the east-west field reaches tropical latitudes and maximum strength, as is observed. Although this theory is evidently on the right track, it is based on an untested model of solar convective motions. Little is known about the actual motions of fluid beneath the solar surface or about the complicated hydrodynamics of the internal convection, circulation, and nonuniform rotation of the Sun. Observations of the north-south circulation at the surface, together with determinations of the angular velocity at various depths through observations of the 5-min surface oscillations, are essential for a better understanding of solar convec- tion and magnetic-field maintenance. The generation of magnetic fields in other stars will remain a puzzle until we have a solid understanding of the circulation in the Sun. The stronger fields in younger stars, implied by their higher levels of activity, may be a direct consequence of their higher rates of rotation. However, it is not at all clear how the very strong magnetic fields of stars more massive than the Sun are produced. Not only are such stars slow rotators, but it is also believed that they have no internal convection. The origin of stellar magnetic fields thus remains obscure. 83 The often explosive dissipation of magnetic fields and the accom- panying acceleration of particles to high energies is a major problem in astrophysics. Laboratory and theoretical work, as well as obser- vations of magnetospheric phenomena in the Sun and planets, shows that reconnection of magnetic-field lines to form configurations of lower energy is a key mechanism. Reconnection requires diffusion of magnetic fields through the gases in which they are embedded; however, the rate of diffusion is negligible according to the equations of magnetohydrodynamics. Apparently, it is necessary to invoke

84 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's plasma instabilities not included in those equations; once established, they may also facilitate the conversion of magnetic energy into the energy of fast particles. Understanding magnetic reconnection and particle acceleration is essential not only for magnetospheric physics and solar physics but also for other violent events in astronomy, including those associated with x-ray sources and quasars. Only in the Sun and planets can these processes be observed both on the scale and in the detail required for a clear understanding of many of their features. Stellar Mass Loss The importance and ubiquity of strong stellar winds became apparent during the 1970's through advances in space ultraviolet astronomy and ground-based infrared astronomy. It is now clear that both the hot, blue giant stars and the cool, red giant stars have stellar winds and that these winds are sometimes prominent features of regions of star formation. The most luminous stars appear to lose mass at rates up to a billion times the mass lost in the solar wind. A luminous giant star may thus lose a substantial fraction of its mass even during its relatively short lifetime of a few million years, with profound effects not only for its subsequent evolution but also for its interstellar environment. The luminous, hot blue giant stars have by far the strongest stellar winds. Observations of their ultraviolet spectra with telescopes on rockets and on the Copernicus and lUE satellites have shown that these winds flow at speeds up to 3000 km/see and are characterized by temperatures below 105 K. The mass-loss rates inferred from infrared and radio observations approach and sometimes exceed 10-5 solar masses per year. Such winds cannot be driven by gas pressure alone, as is the solar wind, but must be driven instead by the stellar ultraviolet radiation pressure, scattered from ions in the wind; the same mechanism is thought to play a role in the ejection of gas from galactic nuclei and quasars. These winds are so powerful that they hollow out enormous cavities in the interstellar gas, push- ing outward expanding shells of interstellar material that resemble those generated by supernova explosions. The mechanism for driving the winds in hot stars is not well understood. Time variations in ultraviolet spectra of the stars suggest that the wind is unstable, and theorists are investigating possible instabilities in an effort to nail down the characteristics of the underlying flow.

Frontiers of Astrophysics 85 Observations with optical, infrared, and radio telescopes show that cool, red-giant stars have winds comparable in strength with those of the hot, blue-giant stars but with temperatures less than 3000 K and with much lower velocities, about 30 km/sec. No ultraviolet or x-ray emission is seen. These winds are rich in dust grains and molecules; since nearly all stars more massive than the Sun even- tually evolve into red giants, such winds provide a major source of new interstellar gas and dust, furnishing a vital link in the cycle of star formation and galactic evolution. As for hot stars, the mechanism for driving these winds is not well understood. Radiation pressure appears to be inadequate to sustain such a high-mass-loss rate, and the gas pressure is too low. Perhaps instabilities, turbulence, and/or magnetic fields in the stellar atmosphere are responsible. Recent observations of H2 and CO molecules in the Orion nebula with infrared and radio spectrographs have revealed clouds of gas expanding outward at velocities approaching 100 km/sec. VUB} radio observations disclose expanding knots of H2O maser emission as- sociated with the region of star formation in the Orion nebula, sug- gesting that strong stellar winds are associated with protostars. The cause of these winds remains unknown. The new instruments for the 1980's recommended in this report will have powerful capabilities for investigations of strong stellar winds and their consequences. AXAF will carry out sensitive surveys and spectroscopic observations of x-ray emission from the winds of luminous hot stars, permitting studies of the properties of the coronal gas and the mechanisms for heating it. ST and the far-ultraviolet spectrograph in space will greatly increase the number of stars that are observable over the number observable with Copernicus and {UK and furnish a better understanding of the physical conditions in the winds of hot stars, of their driving mechanisms, and of their inter- action with their interstellar environments. There is also much to be learned about the winds of both hot and cool giant stars and the winds associated with regions of star for- mation from infrared observations with new instruments. Because its large aperture will make possible higher spectral-resolution mea- surements, NIT will permit us to determine mass-loss rates and ve- locity profiles of the winds of hot and cool giant stars. Because of its high sensitivity, SIRTF will be able to observe stellar winds and their associated dynamics in much more optically obscured regions of star formation, and the LDR, because of its large aperture, can measure their spectra. The VERB Array will provide more reliable ob

86 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's servations of the motions of the radio maser sources in such regions, because its enhanced sensitivity will allow the study of fainter, more numerous sources. PLANETS, LIFE, AND INTELLIGENCE Molecular biology has given penetrating insight into the nature of life. Complex DNA molecules in the nucleus of every cell en- code all the information required to determine how the cell de- velops and functions. All the rich diversity of life on Earth is coded in strands of DNA, which have evolved from primitive forms that apparently arose through a series of transformations in our corner of the Universe. As we consider the myriads of stars in the Universe, we wonder what other genetic systems may exist. Are we alone in the Universe? Questions concerning the ori- gin and fate of life on Earth have been pondered since ancient times. Astronomy has shown that there are enormous numbers of stars like the Sun and that the abundances of chemical ele- ments are much the same everywhere. It seems possible, there- fore, that there are habitats for life scattered throughout the Universe. Life on Earth evolved from primitive forms by muta- tion and natural selection into new species of ever-increasing complexity. The assumption that such processes operate wher- ever the conditions are right remains speculative until other ex- amples of life in the Universe are found. Life in the Solar System Life on Earth is so intertwined with the chemistry of the atmosphere and oceans as to form a single ecosystem, in which each part is affected by the others. Based on carbon, an abundant element, and dependent on liquid water, the ecosystem is sustained by the low- entropy energy of sunlight, captured through photosynthesis. Be- yond these basic considerations, it is not clear what other properties of planet Earth played an essential role in the origin of life. Solid surfaces? Gravity? The daily cycle of light and dark? Several planets in the solar system are similar enough to the Earth that a search for life on them can shed some light on these questions. Mars has long appealed to scientists and the public alike as an in- triguing target for exploration. The 1970's have seen a remarkable effort in that direction, culminating in the orbiting spacecraft and

Frontiers of Astrophysics 87 landers of the Viking mission. Although the best-publicized results were the photographs from the landing sites arid the chemical searches there for microorganisms and organic molecules, large amounts of other data were also returned that relate to the Martian atmosphere and temperature and the geological processes that have shaped the Martian surface. The conclusion that emerges from these studies is that the Martian atmosphere, now dry and cold, was perhaps quite different in the past; probably large amounts of water precipitated at one time, shaping the surface in torrential floods. Conditions at that time could have been more favorable to life, so that further studies are required to tell us whether life ever existed on Mars. Venus has been the subject of intensive study by both the United States and the Soviet Union. Its surface is hot enough to melt lead; this and the fact that the atmosphere is topped by sulfuric acid clouds Old stream beds on Mars, west of the Viking landing site, as seen from the Viking Orbiter. The terrain slopes upward to the left. (Photo courtesy of the National Aeronautics and Space Admin- istration)

88 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's indicate that Venus is inhospitable to life. Mercury is also too hot during its long days, and it has no significant atmosphere. Beyond Mars, Jupiter has been the subject of speculation because its dense atmosphere contains organic molecules that could constitute the first steps toward the origin of life. The temperature is extremely low at the visible cloud tops, but it increases inward, so that organisms would be quite comfortable at an appropriate depth. However, there is no evidence of a solid or liquid surface for life to cling to if, indeed, that is essential. The remarkable properties of Jupiter's sat- ellites were revealed for the first time by the Voyager photographs. Most appear to be too cold and to lack sufficient atmospheres. A1- though lo is internally heated, it is overrun by hot lava and subjected to sulfurous volcanic fumes, so that its surface would be hostile to most forms of terrestrial life. Beyond Jupiter, most bodies are too cold for life to be credible, although Titan, the largest satellite of Saturn, is now known from Voyager observations to have an atmos- phere composed primarily of nitrogen, with a temperature at the surface near the triple point of methane; it is conceivable that inter- esting organic molecules, of types believed to be precursors to prim- itive life, could exist there. In all likelihood, however, life in the solar system has been con- fined to the Earth and possibly Mars. On Earth, the development of intelligence has led to the explosive development of technology and thus to radio communication and powerful radars; the earliest radio signals have by now propagated 50 light-years into space, past sev- eral thousand stars more or less like the Sun. In principle, then, intelligent beings far from our solar system could learn of our exis- tence by studying radio waves from space, just as we could discover any such beings by using similar means. Conditions for Life in the Universe The discovery of complex interstellar molecules shows that chemical processes possibly relevant to the origin of life are ubiquitous. Among them are formaldehyde and hydrogen cyanide, both of which play an important role in laboratory experiments directed at recreating the chemical evolution that preceded the origin of life. More than 50 different molecules have now been found in molecular clouds, in- cluding long-chain molecules such as HCgN. The abundances of the simpler molecules observed have been explained by ion-molecule exchange reactions, but it is not yet certain whether the more complex molecules require other processes, such as surface catalysis on dust grains, for their explanation.

Frontiers of Astrophysics 89 When a molecular cloud collapses to form stars and planets, its complex molecules may be destroyed. On the other hand, molecules might survive the extreme conditions encountered during the col- lapse by condensing onto dust grains. The complex organic molecules found in carbonaceous-chondrite meteorites may have formed in this way. Such a process could be critically important for the origin of life in the solar system, since carbon would otherwise be locked up as either carbon monoxide or methane gas, which are too light to be gravitationally bound by planets as small as the Earth. By con- densing on grains, carbon can be bound into planets, where it is available for life. The synthesis of organic molecules in space will be studied more intensively in the coming decade. By enabling studies of new mol- ecules and higher energy states, with increased sensitivity and an- gular resolution, the millimeter, submillimeter, and infrared instru- ments recommended in this report offer an opportunity to pursue the chemistry of carbon in interstellar clouds, and particularly in the collapsing cores of dark clouds, where stars and planets are believed to form. Furthermore, they will offer additional opportunities to study the molecular composition of comets, which are believed to be samples of the most primitive material in the solar system. Planets like our own may be the most likely habitats for life outside the solar system. However, in view of the lack of confirmation of a planet orbiting Barnard's star, there is as yet no certain evidence that any planets exist beyond our solar system. Current theories of star formation suggest that a fragment of a dark interstellar cloud destined to become a star would have substantial angular momentum, which would set into rapid rotation any gas and dust that accompanies the cloud-to-be-star on its inward gravitational collapse. The accompa- nying centrifugal force would resist the inward pull of gravity and result in a rotating disk of gas and dust that would provide a natural breeding place for planets, arrayed as ours are, in a great flattened and rotating system. So much is speculation; to prove that planets exist near other stars, we must observe them. This is a formidable task, given the fact that even the largest planet in the solar system, Jupiter, has only a tenth of the size and a thousandth of the mass of its parent star, the Sun. One way a distant observer could detect Jupiter is through its grav- itational pull on the Sun. This effect would cause a variation in the radial velocity of the Sun of a few tens of meters per second and, as observed from a distance of 5 parsecs, a 1-milliarcsecond displace- ment on the solar position. Could we detect a Jupiter-sized planet around a nearby star? There already exist several techniques

9o ASTRONOMY AND ASTROPHYSICS FOR THE 1980's for obtaining radial-velocity measurements of the required precision; sustained programs of observations of many candidate stars are now required. Astronomers have expended great effort to make milli- arcsecond-position measurements on nearby stars, but until recently the required precision has not been available. With the development of the optical astrometric techniques recommended in this report, however, it should be possible to observe many stars with a precision exceeding 1 milliarcsecond and by this means to detect Jupiter-sized planets around nearby stars, if they exist. Space astrometry should ultimately yield much higher positional accuracy, leading to the still more interesting prospect of detecting Earth-sized planets. Far-in- frared interferometric observations from space could also reveal planets around nearby stars. Search for Extraterrestrial Intelligence Even if other planets are detected, it will still be difficult to infer whether life is present; to do so directly would require imaging the planet itself with as yet undreamed-of resolution. Only if the planet harbors intelligent life capable of producing electromagnetic signals detectable at Earth is there at present any hope of finding life outside our solar system. It is a remarkable fact that radio and television signals generated copiously on Earth could be detected at distances of many parsecs by civilizations that, like ours, would otherwise have no way of knowing of the existence of details of the planet that is our home. Should the human race search seriously for signals from other possible civilizations? Much has been written about this question, both on a technical and a philosophical level. Reception of intelligent signals from space could have a dramatic effect on human affairs, as did contact between the native peoples of the New World and the technologically more advanced peoples of Europe. The effects would be beneficial, if the information could be deciphered and should prove generally useful; on the other hand, they could be harmful if humanity is not ready to use the information wisely. The technology is now available to make significant searches of this kind. The 300-m radio telescope of the National Astronomy and Ionosphere Center at Arecibo, Puerto Rico, is capable of receiving a message beamed at us from any of the hundreds of billions of stars in our Galaxy, provided the civilization sending the message were transmitting with a facility similar to that at Arecibo. Several searches for such extraterrestrial signals have already been undertaken, so far

Frontiers of Astrophysics 91 with negative results, but the rate of improvement of communica- tions technology is so rapid that each search has been far more sensitive than its predecessors. We are entering an era when it is technically possible both to detect planets around nearby stars and to detect signals from intel- ligent life on planets immensely farther away, even if we cannot detect the more distant planets themselves. Both investigations would bear directly on important scientific questions. Our interest in the tiny fraction of the matter in the solar system that condensed into planets is heightened by the fact that life has developed on at least one of them. Have condensations to planets and the origin of life occurred elsewhere as well? And has that life evolved into com- municative intelligence, with which we human beings might be able to enter a conversation about life in the Universe? These questions reach far beyond astronomy, and even beyond science as we currently think of it. Yet astronomers, who are in a sense commissioned by the public to keep an eye on the Universe, feel bound to ask them and to point out how we might begin to try to answer them. It is for these reasons that the Committee recom- mends that in the 1980's an astronomical Search for Extraterrestrial Intelligence be initiated as a long-term effort. ASTRONOMY AND THE FORCES OF NATURE Energy Sources in the Universe In the 1970's, physicists have made substantial progress toward re- alizing an age-old dream-the understanding of all the forces in nature as different aspects of a single fundamental force. A theory that unifies electromagnetic and weak nuclear forces has been suc- cessfully developed along with a comprehensive theory of the strong nuclear force; new theories aimed at unifying both of these theories are now being proposed. Astronomical data have played a role in these developments and may play an even greater role in the future. Newton's law of gravitation, formulated in precise mathematical terms, set the stage for the investigation of the forces of nature that continues today. We now realize that chemical energy, such as that released in the burning of fossil fuels, is due to the action of electrical forces within atoms. Holding electrons in orbits around nuclei just as gravitation holds planets in their orbits around the Sun, these forces release energy whenever an electron drops into a lower orbit. Magnetic forces result from the motion of electrically charged par

92 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's ticles. In the 1860's, Maxwell unified electrical and magnetic forces in a single theory, called electromagnetic theory, which also explains electromagnetic radiation as a wave that sustains itself through a constant interplay between electrical and magnetic energy. By the enr1 of the nineteenth century, both gravitational and electromagnetic forces were well understood at a certain level. O Early in the twentieth century a series of important experiments revealed that the orbits of electrons are qualitatively different from those of planets. The position of a planet can be predicted precisely from a knowledge of the gravitational force acting on it, but the best one can do with an electron is to predict its probability of being at various possible positions. The impossibility of doing any better, embodied in Heisenberg's Uncertainty Principle, is an essential fea- ture of what is now known as quantum theory. Today, the melding of electromagnetic theory and quantum theory, called quantum elec- trodynamics or QED, is unchallenged in its ability to describe elec- tromagnetic phenomena. A shining goal of contemporary physics is to bring the understanding of all the forces of nature up to the standard of QED. Sunlight is electromagnetic radiation, and the form in which the energy of sunlight is stored by plants is chemical energy; both forms of energy are embraced by QED. What about the energy stored in the Sun, which it emits as sunlight? Early suggestions included elec- tromagnetic radiation trapped within the Sun, chemical energy stored in its atoms and molecules, and the energy due to the gravitational attraction between all of its atoms. However, none of these forms of energy is adequate to keep the Sun shining for its known age of 4.5 billion years. The solution to this problem was reached in the early 1920's, when it was recognized that a new form of energy discovered in the laboratory, nuclear energy-which is released, for example, when the nuclear force between hydrogen nuclei (protons) draws them together to form helium nuclei-could keep the Sun shining for many billions of years. Nuclear interactions come into play only at very high temperatures; only then do nuclei have sufficient speeds to overcome their mutual electrical repulsion. Thus, nuclear forces play a role in astronomy only where matter is extremely hot, as in the interiors of stars or in the searing heat of the big-bang explosion. Laboratory studies of nuclear reactions show that there are actually two types of nuclear force, strong and weak; the latter is associated with an unusual particle called the neutrino.

Frontiers of Astrophysics Two Puzzles: Solar Neutrinos and Hidden Mass 93 Neutrinos can penetrate the entire Sun, so weak is the force with which they interact with matter. Detectors placed beside nuclear reactors, which are copious sources of neutrinos, can record only a minute fraction of those emitted. Despite the great difficulty of de- tecting them, the role of neutrinos in astronomical research has be- come increasingly important. The current theory of stellar energy generation predicts that large numbers of neutrinos are produced in the fusion of hydrogen to helium in the deep interior of the Sun. Because this theory is critical to our understanding of stellar structure and evolution generally, it is important to test this prediction by measuring the flux of solar neutrinos at the Earth. The observed flux of neutrinos is less than one third of that predicted from the most carefully constructed models of the solar interior. Among various proposed explanations of this discrepancy is the possibility that neutrinos behave differently from what has been assumed until recently. In a completely different area of research, it has been proposed that the problem of hidden mass in galaxies might be resolved if the rest mass of neutrinos were not zero, as usually assumed. From calculations of the number of neutrinos produced in the big bang, one finds that neutrinos could supply the hidden mass in galaxy clusters if they possess a rest mass about 1/10,000 that of the electron. There are thus two astronomical problems that might be resolved if neutrinos prove to have properties not previously known. Theo- retical physicists have recently suggested a resolution of both of these problems. The recently developed unified theory of weak and elec- tromagnetic forces is based on a principle called gauge invariance and is therefore referred to as "the gauge theory of weak and elec- tromagnetic interactions." So far it has succeeded in explaining all the various phenomena involved with both electromagnetic and weak nuclear forces. The gauge theory of weak and electromagnetic interactions in its original form says nothing about the problems of solar neutrinos or hidden mass. However, pursuing the principle of gauge invariance behind it, physicists have constructed a theory of the strong nuclear force, called quantum chromodynamics, or QCD. This theory pos- tulates the existence of elementary particles that combine to form protons and neutrons, called quarks. The success of QCD in explain- ing the results of experiments in elementary-particle physics gives

94 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's increasing confidence that it is the correct theory of the strong force that binds neutrons and protons into atomic nuclei. Spurred by the success of the gauge theory of weak and electro- magnetic interactions and of QCD, physicists are now trying to find an even more general gauge theory, called "Grand Unified Theory," that incorporates both. Some theories of this type predict that there should be the three types of electrons that are actually observed, as well as three corresponding types of neutrinos, called e, mu, and taut In some versions of the theory, e, mu, and tan neutrinos are regarded as three aspects of the same basic neutrino, which has a finite rest mass and which oscillates back and forth among its three aspects. Although the nuclear reactions in the Sun emit only e neu- trinos, according to some Grand Unified Theories neutrino oscilla- tions would be expected to occur long before the neutrinos reached the Earth, so that at the Earth one would observe a random mixture of e, mu, and tan neutrinos. Since the Homestake Mine apparatus is sensitive only to e neutrinos, a factor-of-3 discrepancy would thereby be explained. Oscillations can occur only if neutrinos have a finite rest mass. If the value of the rest mass were in the right range, it would have a dramatic bearing on our understanding of the hidden-mass problem and of the ultimate fate of the Universe. Theories involving several different types of neutrinos are con- strained by calculations of the properties of the early Universe. If there were more than about four types of neutrinos, their contri- bution to the gravitational acceleration in the early Universe would have been so great that there would not have been sufficient time for primordial neutrons to decay; there would then be more helium in the Universe than is actually observed. Thus, current astronomical observations eliminate some versions of Grand Unified Theories. A critical experiment endorsed earlier in this report will help to shed light on the true nature of neutrinos. The gallium solar neutrino experiment will be sensitive to neutrinos of much lower energy than those measured by the 37C1 detector in the Hamestake Mine. The flux of such lower-energy neutrinos can confidently be calculated from the observed luminosity of the Sun, independently of the details of solar models. If there is a discrepancy between the predicted and observed values of the solar neutrino flux in the gallium experiment, it could be an indication that neutrinos oscillate and have a finite neutrino rest mass. There may also be powerful sources of high-energy neutrinos among the many sites of violent activity observed to occur on both stellar

Frontiers of Astrophysics 95 Chlorine solar-neutrino detector deep ire the Homestake Mine, Lead, South Dakota. (Photo courtesy of R. Davis, Jr., Brookhave~z National Laboratory) and galactic scales. Despite the difficulty of detecting such neutrinos and the weak fluxes to be expected because of the distances to the sources, the study of energetic-neutrino detectors with possible as- tronomical applications is appropriate for the coming decade. An interesting possibility for such study is the proposed observation of neutrino-induced reactions in seawater employing arrays of photo- multipliers to detect the associated Cerenkov radiation. Before the First Three Minutes Although astronomical data now available appear to be in agreement with the predictions of big-bang cosmology, the big-bang model cannot yet be considered conclusively proven, so that it is of the greatest importance to test its predictions however possible. In par- ticular, the model predicts that the cosmic microwave background originated as high-temperature radiation in the first few minutes of time. As the Universe expanded, according to this view, the radiation cooled to its present observed temperature, about 3 degrees above absolute zero. When the Universe was about 1/10,000 of a second old, its temperature was a trillion degrees, so hot that the radiation

96 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's present created about 100 million proton-antiproton pairs for every proton now observed in the Universe. As time passed, these pairs annihilated, leaving behind only the very small fraction by which the number of protons exceeds the number of antiprotons. Had this excess not existed, the number of protons in the present Universe would have been 10 billion times smaller, and there would not have been sufficient matter in the Universe for the formation of galaxies, stars, and planets. What caused the excess of matter over antimatter implied by this big-bang scenario? Until recently, physicists had regarded the excess as a fact as inexplicable as the existence of the Universe itself. Re- cently, it has been suggested that Grand Unified Theories provide an explanation: very heavy particles present in the first 10-38 sec of the history of the Universe decayed, creating in the process slightly greater numbers of protons than antiprotons. This prediction can be tested in a straightforward way, for if protons can be created they must also decay. As the lifetime of the proton estimated from Grand Unified Theory is 100 billion billion times the age of the Universe, physicists are not concerned that the Universe will soon evaporate. On the other hand, the predicted proton lifetime is sufficiently short that one such decay will occur in a ton of material each year. Ex- periments are now in progress to detect such events. The Limits Of Gravitation Gravity keeps us on the Earth, binds the Earth to the Sun, and slows the expansion of the Universe. Newton described it as a force, while Einstein, in his General Theory of Relativity, interpreted gravitational forces in terms of the curvature of space-time. Einstein's theory, unlike Newton's, is believed to be valid for very strong gravitational fields and for bodies moving close to the speed of light; it is therefore crucial for an understanding of systems such as neutron stars, black holes, and the expanding Universe. The General Theory of Relativity predicts that when any non- spherical body collapses to form a compact object or a black hole, it emits a new form of energy called gravitational radiation. Although this radiation is predicted to be extremely difficult to detect, several research groups are now building detectors thousands of times more sensitive than those available during the 1970's. Parallel efforts to calculate the amount of gravitational radiation emitted by collapse indicate that, if the planned development of new instrument concepts succeeds, we might hope to detect an event within two decades-

Frontiers of Astrophysics 97 even earlier if there should be a new supernova within the Galaxy. The recently confirmed, slow decrease in the orbital period of the binary pulsar has already been interpreted as the result of gravita- tional radiation from a close pair of neutron stars. While efforts to develop a quantum theory of gravitation have not yet succeeded, there is reason to believe that quantum effects should occur near black holes, where space-time curvature is high. The quantum theory of elementary particles predicts that even in vacuum, particle-antiparticle pairs are constantly being produced and anni- hilated in an interval of time too short to observe. If this effect should occur near a black hole, one member of the pair may fall into the black hole before the pair annihilates. Zero-mass particles, including photons, are created similarly; the black hole thus appears to the outside world as a source of radiation, ultimately evaporating as a result of the energy lost. Black holes of all sizes could have been created in the big bang; in particular, those having masses about of 10~5 g (the mass of a small mountain on Earth) would just be evap- orating now, giving rise at the ends of their lives to bursts of gamma radiation. Such radiation from evaporating black holes has been searched for, and, although the Gamma Ray Observatory will con- tinue the quest, so far none has been found. It thus appears that primordial black holes with masses less than that of a mountain cannot make up a significant fraction of the mass of the Universe. The theory of black-hole evaporation depends on the quantum nature of strong nuclear forces but not on the quantum nature of gravitation. Although no convincing theory of gravitation that in- corporates the quantum principle has yet been produced, it is con- jectured that the quantum effects must become important whenever the radius of curvature of space-time becomes less than the so-called Planck length, 10-33 cm. Such conditions are thought to have occurred in the Universe at times before 10~3 sec and at temperatures above 1032 deg. Because the energies and temperatures characteristic of Grand Unified Theories are remarkably close to these values, some physicists believe that a theory should be possible that incorporates all four forces in nature into one "Super-Grand" force at energies only slightly higher than those relevant to Grand Unification. A prime hope for such a theory is that it will yield, almost as a by- product, the correct theory of quantum gravitation. Attempts in this direction have so far met with little if any success, but the devel- opment of such a theory could be considered to be the ultimate challenge to physics at present. The notion of force, as a law governing matter once created, fails

98 ASTRONOMY AND ASTROPHYSICS FOR THE 1980's to take account of the process of creation itself. Is it possible, as astrophysics pushes the frontiers of time back to the moment of cosmic creation, that the existence of the Universe will be recognized as a consequence of the nature of the fundamental force? Is it possible that the potential existence of the world somehow calls it into exis- tence? Such questions, once believed outside the range of science, are now arising in scientific thought.

The primary mirror for the Space Telescope being inspected after figuring. Photo courtesy of the National Aeronautics and Space Administration)

Next: 4 APPROVED, CONTINUING, AND PREVIOUSLY RECOMMENDED PROGRAMS »
Astronomy and Astrophysics for the 1980's, Volume 1: Report of the Astronomy Survey Committee Get This Book
×
Buy Paperback | $60.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!