Astronomers believe that the atoms of our bodies were created in the primordial explosion that marked the beginning of the universe and in the nuclear fires burning within stars. Some of these atoms, scattered throughout space, ultimately formed planets, soil, and organic molecules. Thus to understand our human beginnings we must understand the life history of stars and galaxies and even the whole universe. Whereas astronomers of the past were concerned more with charting the stars in a permanent cosmos, astronomers today study evolution and change.
Astronomical findings in this century have raised questions about life cycles in space. We have found that the Great Red Spot of Jupiter is not a fixed birthmark, but results from a continuing struggle of gases in motion. We have witnessed stars in the act of formation, still wrapped in the gas and the dust out of which they condensed. We have seen other stars exploding, having first spent their nuclear fuel and then collapsing under their own weight. And in the debris from stellar explosions, we have found oxygen and carbon and other elements essential to life. We have discovered huge streams of matter propelled from the centers of galaxies at nearly the speed of light. We have found that galaxies evolve. We have learned that galaxies are not scattered evenly in space, but are bunched together in filaments and sheets and other large groupings whose origins have not yet been explained. Finally, we have gathered evidence that
the entire mass of the universe began in a state of fantastic compression, some 10 billion to 20 billion years ago.
How did the universe come into being in the first place? What determined its properties? Will it keep expanding forever or instead collapse on itself? A century ago, such questions were considered to lie outside the domain of science. Today, they are central to the field of astronomy.
Progress in astronomy is driven by advances in technology. In the 1930s, new communication devices led to the reception of radio waves from space. For thousands of years before, visible light had been our only way of seeing the universe. Since the 1950s, rockets and satellites have recorded infrared radiation, ultraviolet radiation, and x-rays emitted from space. Such radiation, invisible to the human eye, has revealed completely new features of many astronomical objects and has announced some objects not before known. New electronic detectors have replaced photographic plates, resulting in a 100-fold increase in sensitivity and a broader range of available wavelengths. Highspeed computers have revolutionized theoretical astronomy by permitting the simulation of millions of interacting stars or galaxies. Electronically linked to combine the data from different antennas, large arrays of radio telescopes can work together as if they were one giant eye.
As discussed in Chapter 1 and Chapter 4, astronomical exploration in the 1990s will take advantage of novel technologies to make new instruments of startling power. In equipping ourselves for the future, diversity must accompany precision. We theorize and forecast as well as we can, but if the past is a guide, some of the next decade's discoveries will catch us off guard. In astronomy, the frontiers surround us.
OUR SOLAR SYSTEM AND THE SEARCH FOR OTHER PLANETS
The Formation and Evolution of Our Solar System
In the middle of the 18th century, the German philosopher Immanuel Kant proposed that our system of planets and sun condensed out of a great rotating cloud of gas and dust. This proposal, called the nebular hypothesis, has gained enormous observational and theoretical support and is favored today. Kant suggested that a primitive gaseous cloud slowly contracted under the inward pull of its own gravity. The central, densest regions formed the sun. The outer regions collapsed along the axis of rotation, because of gravitational forces, but did not fall directly toward the nascent sun, because of centrifugal forces pushing outward. Caught between these two forces, the material formed a flattened shape, called a protostellar or protoplanetary disk, in orbit about the sun.
In processes still under investigation by astronomers using computer simulations, material in the disk formed into dense condensations that eventually coalesced into planets—some small and rocky, like Mercury, Venus, Earth, and Mars, and others further from the sun and composed primarily of hydrogen gas, like Jupiter, Saturn, Uranus, and Neptune. Almost all the remaining material in the disk was swept out of the solar system, except for trace amounts of dust and small bodies like comets and asteroids.
Two different phases in this evolutionary scenario received observational support in the 1980s. Astronomers working with data from an orbiting telescope in space, the Infrared Astronomical Satellite (IRAS), found young stars forming in the nearby constellation of Taurus that appeared to be surrounded by disks roughly the size of a nascent solar system. But IRAS alone lacked the ability to discern the small structures and small velocities necessary to make a conclusive judgment. Astronomers using arrays of millimeter wavelength telescopes linked together to act as a single large telescope demonstrated that the material around the young stars was in orbit around the stars and was present in a quantity sufficient to make solar systems. IRAS found supporting evidence for a later phase in the life history of solar systems when it discovered that as many as one-quarter of all nearby stars are surrounded by disks of orbiting particles that may be the debris left over from the formation of planets (Plate 2.1).
Understanding how stars and planets form will be one of the major scientific themes of astronomy in the 1990s. Progress will come from increases in sensitivity and spatial resolution at radio, infrared, and optical wavelengths. The Space Infrared Telescope Facility (SIRTF) will be a thousand to a million times more sensitive than its predecessor IRAS, enabling scientists to search for and characterize protoplanetary disks at all stages of evolution. Observations with the Millimeter Array (MMA) and with the infrared-optimized 8-m-diameter telescope to be built on top of Mauna Kea, Hawaii, will measure features as small as a few times larger than an Astronomical Unit (or an AU, as the earth-sun separation is called by astronomers) at the distance of the nearest star-forming regions. All of these telescopes will help determine the temperature, density, and composition of protostellar disks. A technique called interferometry will be used at infrared wavelengths to discern angles as small as a thousandth of an arcsecond. 1 Images with this ultrahigh resolution may reveal traces of the formation of Jupiter-sized planets, such as gaps in the distribution of disk material.
Angles are measured in degrees, which can be divided into 60 arcminutes and further subdivided into 3,600 arcseconds. The full moon subtends half a degree, and a large lunar crater an arcminute. An arcsecond is the angle subtended by a penny at a distance of two and a half miles. A thousandth of an arcsecond is the angle subtended by a penny seen from a continent away.
The Search for Other Planets
The Italian philosopher Giordano Bruno argued that space is filled with infinite numbers of planetary systems, inhabited by a multitude of creatures. For this and other indiscretions, Bruno was burned at the stake in 1600. Yet the question remains: are there other planetary systems in the universe? In the 1980s, for the first time, some evidence was found for disks of material surrounding other stars. However, as yet, no definitive observations have been made of a planet in orbit around another star. Some of the instruments proposed for the 1990s have the capability for detecting the Jupiter-sized planets of other stars.
Direct imaging of a distant planet is not easy, because the light from the parent star is billions of times brighter than the planet and because the star and its orbiting planet appear so close together as to be almost inseparable. The problem is analogous to trying to find from a distance of 100 miles a firefly glowing next to a brilliant searchlight. Fortunately, there are other ways to find planets. The position of the central star of a planetary system should wobble slightly in response to the changing gravitational tugs of its orbiting planets. At the distance of Alpha Centauri this shift in position amounts to a shift in angle of only a few thousandths of an arcsecond for a planet the size of Jupiter. Yet optical and infrared telescopes linked together as ground- or space-based interferometers will be capable of measuring such small angles and of surveying hundreds of stars within 500 light-years for the presence of distant planets like our own Jupiter. The orbital wobble of parent stars should also produce small velocity shifts that should be detectable with sensitive instruments on the large ground-based telescopes to be built in the 1990s.
Comets and the Origins of Life
Our sun and its solar system were formed about 4.5 billion years ago, as deduced from the proportions of uranium and lead measured in meteorites. In the intervening eons, life formed, and, most recently, humans began to explore the universe. What molecules were present to make the amino acids, proteins, and DNA that formed the first living creatures? What was the origin of life?
Weather, volcanoes, and bombardment by asteroids have erased from the planets almost all traces of the initial conditions in the solar system. Comets, however, spend most of their time far from the sun and may show us the pristine material of our world. Astronomers were surprised to find that the carbon in Halley's Comet (Figure 2.1) exists in the form of “tar balls” of complex organic molecules, rather than in simpler methane and carbon monoxide gas. The biological significance of this discovery is still hotly debated.
Comets will be closely studied in the coming decade. In addition to the close-range exploration planned for NASA's Comet Rendezvous Asteroid
Flyby spacecraft, observatories on the ground such as the MMA and new 8-m-diameter optical and infrared telescopes, in airplanes such as the Stratospheric Observatory for Far-Infrared Astronomy (SOFIA), and in orbit about the earth such as SIRTF will determine the size, distribution, and composition of comets as far away as the orbit of Jupiter.
Weather and Volcanoes
What forces create the climate and weather on the earth? What forces make the continents drift and volcanoes erupt? Is our climate changing because of the actions of humans? Some information on these questions comes from
measurements of the variability of the sun's output and of how other stars like the sun vary. Different perspectives come from observing other planets and their satellites with robotic spacecraft sent on voyages of discovery, and from telescopes on, or in orbit around, the earth. Observations of the atmospheres of the other planets made possible by infrared and radio telescopes have revealed the constituents of those atmospheres, as well as variations in their temperature and density with height. Careful examination of the dimming of the light from a star as it passed behind the planet Pluto was used to probe the atmosphere of this distant planet. Future observations will probe the climatology and meteorology of the planets and their satellites to give a comparative basis for understanding our own environment.
The Voyager spacecraft found sulfur-spewing volcanoes on Jupiter's satellite Io that appear to be driven by the grinding tides raised by the giant planet. These volcanoes are now monitored by earth-bound telescopes using cameras sensitive to the heat, or infrared radiation, from the volcanoes (Plate 2.2). The sizes, shapes, and compositions of asteroids are revealed by infrared observations and by bouncing strong radar signals off these objects (Plate 2.3).
THE LIFE HISTORY OF STARS
In his masterwork the Principia (1687), Newton wrote that “those who consider the sun one of the fixed stars” may estimate the distance from the earth to a star by comparing its apparent brightness with that of the sun—in the manner that the distance to a candle may be judged by comparing its brightness with that of an identical candle nearby. Newton then calculated that the closest stars are about a million times farther away than the sun, in good agreement with later measurements. The sun is the closest star, and its careful study has important implications for our knowledge of all stars.
The sun is a great ball of hot gas, about a million miles in diameter. According to modern theory, its central density is about 100 times that of water, and the temperature is about 15 million degrees celsius. Such high temperatures are needed to smash subatomic particles together violently enough to fuse them and release nuclear energy. The liberated energy does two things. It maintains the heat within the sun, providing sufficient pressure to resist the inward pull of gravity. The liberated energy also turns into radiation, which slowly makes its way to the solar surface, finally creating the light we see. Some of the sun 's energy goes into churning up its surface and producing extremely energetic particles, magnetic fields, solar flares, and a tenuous atmosphere of higher temperature called the corona.
Many mysteries remain. How do the magnetic fields get their energy? What is the role of magnetic fields in the curious 11-year cycle of activity
of the sun? Astronomers from around the world are proposing to build the Large Earth-based Solar Telescope (LEST), a 2.4-m solar telescope in the Canary Islands. The hallmark of LEST is unprecedented angular resolution, obtained by the use of special optics that will remove the distorting effect of the atmosphere. The principal aim of LEST is to observe the solar surface with sufficient acuity to reconstruct the three-dimensional structure of solar magnetic fields.
In conjunction with LEST, orbiting satellites equipped with visible, ultraviolet, and x-ray telescopes will study the radiation from the sun. Other x-ray satellites, such as the Advanced X-ray Astrophysics Facility (AXAF) already under construction, will monitor the coronal emission from other stars to help understand our own sun.
Astronomers have little direct data from the interior of the sun or any other star. The light we see comes from its surface. However, the surface motions contain clues about conditions far below. A major new field of study called helioseismology measures these vibrations of the solar interior. Just as the intensity and intervals of terrestrial earthquakes tell us about conditions deep within the earth, so also do the vibrations of the sun's surface inform us about the density, temperature, and rate of rotation in the deep interior (Plate 2.4).
A better knowledge of the interior of the sun might resolve another problem that has worried astronomers for years. The number of subatomic particles called neutrinos emitted by the sun and detected on the earth is much smaller than predicted. Neutrinos are produced in nuclear reactions at the center of the sun, and their rate of production has been calculated from our theories of nuclear physics and presumed knowledge of the conditions of temperature and density in the sun. Since neutrinos interact extremely weakly with other matter, almost every neutrino produced at the sun's center should escape. For the last 20 years, American physicists have counted neutrinos emitted from the sun and found fewer than one-third of the predicted number. The discrepancy is serious. Either the interior conditions of the sun are not what we think, or the neutrino has some property that allows it to change form and avoid detection once emitted. The former explanation, if correct, could alter our theories of the structure of stars. The latter would have significant implications for our understanding of subatomic physics. Planned for the 1990s are several new, more sensitive experiments to monitor neutrinos emitted from the sun.
The Formation of Stars
It takes two things to form a star—matter, and a mechanism to compress the matter to high density. Matter is plentiful in space in the form of diffuse hydrogen gas along with traces of other elements and small particles of dust. A dense clump within a gas cloud can pull itself together by gravity, becoming even denser. When the inward pull of gravity is sufficiently strong to overcome
the pressure that tends to blow the clump apart, then the cloud will contract and fall toward its center. In a rotating cloud, a disk of gas and dust the size of a solar system may form in orbit around the nascent star. Matter continues to accrete onto the still-contracting central object, now called a protostar. As the cloud and protostar contract, gravitational energy is released as heat, and the protostar glows brightly at infrared wavelengths. When the temperature at the center of the protostar rises to around 10 million degrees celsius, enough to ignite nuclear reactions, a star is born. Depending on the circumstances of their birth, stars have masses ranging from about 0.1 to 100 times the mass of our sun. Smaller masses never get hot enough at their centers to ignite nuclear reactions; larger masses blow themselves apart at formation by the outward force of their own radiation.
This theory of the formation of stars was given support in the 1980s, when the IRAS detected tens of thousands of stars in the process of formation. More specifically, the satellite detected embryonic stars enshrouded in dense cores of gas clouds, during the early phase of collapse before the nuclear reactions had begun (Plate 2.5). Our understanding of star formation was given another boost in the 1980s when telescopes operating at millimeter wavelengths made an unexpected discovery: streams of gas flowing outward in opposite directions as jets from the vicinity of embryonic stars. Theorists argue that these gaseous streams may be aligned by a planet-forming disk around the young star. These jets are observed at millimeter wavelengths and occasionally break out of the surrounding cloud to become visible at optical wavelengths (Plate 2.6).
Because the earliest stages of star formation occur within very dense clouds of gas, impenetrable to visible light, clues must be sought in the radio waves and infrared radiation that can escape the thick clouds (Plate 2.7). During the 1990s, star formation will be a major focus of study with many of the telescopes proposed in Chapter 1, including SIRTF, SOFIA, the MMA, and the infraredoptimized 8-m telescope on Mauna Kea. Each of these telescopes will make unique contributions to these studies. SIRTF will make sensitive spectroscopic measurements at wavelengths inaccessible from the ground. SOFIA will observe submillimeter wavelength radiation from a large variety of molecules and atoms to characterize the conditions of high density and temperature in protoplanetary disks. The MMA and the infrared-optimized 8-m telescope will make high-spatial-resolution studies of the disks and the outflowing jets that will clarify the dynamics of these regions (Plate 2.8).
The systematic, thorough investigation of star formation throughout the galaxy is important, because star formation and the power of hot, young stars is an important energy-generation mechanism in many galaxies. IRAS found that some galaxies emit copious amounts of infrared energy, perhaps due to bursts of star formation thousands of times more intense than anything seen in our own galaxy. We can study the basic physical processes of star formation up close in our own galaxy and apply that knowledge to more distant systems.
The Life and Death of Stars
Once a star has formed, it spends most of its active lifetime burning its initial nuclear fuel, hydrogen, by a process of nuclear fusion that combines four hydrogen atoms together to make a helium atom and energy. Eventually, the star's hydrogen supply is spent, and, if the star is massive enough, the star's helium supply begins to fuse together to make carbon atoms, which then go on to make heavier and heavier atoms until all the nuclear fuel is converted to the element iron, which is incapable of further energy-releasing reactions. Once there exist no further resources of heat and pressure to counterbalance the inward pull of gravity, the star must collapse.
Our own sun has already lived about 5 billion years and will live another 5 billion years, quietly burning its hydrogen, before it swells into a red giant star. Then, in the relatively brief period of about 100 million years, it will exhaust the rest of its nuclear fuel and collapse. More massive stars spend themselves more quickly and less massive stars more slowly. For example, a star of 10 times the mass of our sun burns up its core of hydrogen gas and becomes a red giant in only about 30 million years.
A burned-out star may end its life in several ways. If the mass remaining after the red giant phase does not exceed several times that of our sun, it becomes a dense, dim white dwarf star or, after a violent stellar explosion called a supernova, an even denser cold star called a neutron star. A white dwarf is about 100 times smaller than a younger star of the same mass. A neutron star is yet another 1,000 times smaller than a white dwarf and is composed almost entirely of neutrons, uncharged subatomic particles, packed together side by side. A typical neutron star has an incredible density: the mass of Manhattan Island squeezed into a cherry. Furthermore, that star can spin very rapidly, between 1 and 1,000 revolutions per second; it can anchor magnetic fields that are trillions of times stronger than the earth's, and it can produce periodic pulses of intense radio waves. White dwarfs and neutron stars support themselves against further collapse by the resistance of their subatomic particles to being squeezed more closely together. They can remain in such balance almost forever. Yet these massive spheres that were once shining stars have no source of energy, other than their energy of rotation, and so eventually grow dim and cold.
The first white dwarf was identified in 1914 and the first neutron star in 1967. Astonishing as it may seem, astrophysicists had predicted the characteristics of neutron stars before their discovery. Swiss-born astronomer Fritz Zwicky and German-born astronomer Walter Baade, working together in California, correctly forecasted the existence and properties of neutron stars as early as 1933, only two years after the discovery of the neutron itself in terrestrial laboratories. Such accurate predictions testify not only to the power of theoreti-
cal calculations but also to the validity of the assumption that the same physical laws found on the earth apply to distant parts of the universe.
White dwarfs are the corpses of stars whose initial mass was less than about eight times the mass of our sun. More massive stars that have exhausted their nuclear fuel face a different end. For such stars, no amount of resistive pressure can stave off the overwhelming crush of gravity. These stars will explode in a brilliant supernova, which for a brief time can shine with the power of 100 billion stars. The core remaining after the explosion will become a neutron star, or, if the core is massive enough, an object called a black hole, whose gravity is so intense that not even a ray of light can escape it. The intense bursts of radiation given off by magnetized, rapidly rotating neutron stars are often detected by radio telescopes as pulsars. Stellar black holes are inferred to exist from the x-ray emission seen from some binary stars.
In early 1987, astronomers were handed a rare opportunity to test their theories of the evolution, collapse, and explosion of stars. A star exploded nearby, without warning, offering an unprecedented view of a supernova. By carefully monitoring the light from Supernova 1987A, as it is called, and by identifying in older photographs the star that blew up, astronomers have learned a great deal about the origin and nature of supernovae. The event confirmed the general theoretical outlines. The infrared and gamma rays from the radioactive decay of cobalt, and the amount of nickel and other elements ejected by the explosion, could all be understood. Also detected from Supernova 1987A were neutrinos, whose properties are not completely known. According to previous theories, neutrinos should be manufactured in great numbers during the formation of a neutron star. The neutrinos detected on the earth from Supernova 1987A not only confirmed the predicted temperatures and densities inside a supernova, but they also allowed physicists to learn more about the neutrino.
Supernovae play a vital role in the life cycle of stars. The debris from stellar explosions spreads out into space and adds new ingredients to the gas between stars from which new stars form. Thus supernovae are beginnings as well as ends. Theoretical calculations suggest that essentially all of the chemical elements except hydrogen and helium, the two lightest elements, were manufactured by nuclear fusion inside stars. The vast majority of the 100 chemical elements, including oxygen and carbon and other elements that earthly life depends on, were synthesized in stars and blown into space. Some of this seeding occurs during the red giant phase, as a star sheds its surface layers, and some of it occurs in the wind of particles that flow from the hot stellar atmosphere. The rest happens in supernova explosions. Later generations of stars, such as our sun, are born from the gas enriched by these new elements. The gas between stars connects the generations, receiving from the old stars and giving to the new.
Supernova explosions are thought to be the underlying cause for the acceleration of protons and heavier atoms to extremely high energies. These particles, called cosmic rays, move at nearly the speed of light, gyrating in the magnetic field of the galaxy, to bring us news about distant parts of the galaxy, news that we can currently decipher only imprecisely. Some cosmic rays enter the earth's atmosphere and can be detected by telescopes that look for the light given off when cosmic rays collide with stationary molecules in the atmosphere. Yet an innovative telescope called the Fly's Eye (Plate 2.9), which consists of hundreds of photosensitive tubes that scan the skies over Utah, has found tracks of cosmic rays of such high energy that even supernova explosions may have been inadequate to accelerate them. If these cosmic rays come from outside the galaxy itself, nobody can explain why they are there. An enhanced Fly's Eye telescope planned for the 1990s may solve this question.
The causes of stellar explosions, and the enrichment by supernovae of interstellar gas with life-giving heavy elements, will be the focus of extensive study with many of the telescopes to be built in the 1990s. Measurements of cosmic rays can determine abundances of the elements in the galaxy directly. A satellite called the Advanced Composition Explorer will measure the abundances in cosmic rays of all the elements up to zirconium, an element 90 times heavier than hydrogen. Scrutiny of the x-rays and gamma rays emitted by supernova remnants will help identify the various kinds and proportions of atoms dispersed in supernovae (Plate 2.10). Such a task will be on the dockets of the Gamma Ray Observatory; an Explorer satellite equipped with a gamma-ray spectrometer; AXAF; and two ultraviolet instruments currently under construction, the Far Ultraviolet Spectroscopy Explorer (FUSE) and the Extreme Ultraviolet Explorer. Chemical elements manufactured and spewed out into space by supernovae can also be identified by their infrared emissions. The proposed SOFIA telescope, quickly deployable because it is aboard an aircraft, will have the flexibility to study supernova debris on short notice. The high sensitivity of the proposed SIRTF mission will permit infrared measurements of supernovae out to 30 million light-years from the earth.
THE LIFE HISTORY OF GALAXIES
The Milky Way as a Galaxy
The Milky Way, the faint opalescent band of light that sweeps across the night sky, is our galaxy of 100 billion stars, all orbiting in a flattened disk. The central portions of the Milky Way, obscured at visual wavelengths by intervening interstellar dust, are revealed by infrared observations from ground-based telescopes and by orbiting satellites. Clear views of the stars and gas clouds that form the disk of the Milky Way have been obtained by the IRAS and the Cosmic Background Explorer (COBE) satellite and are shown on the
cover of this report and as Plate 2.11. The universe is filled with more distant galaxies, each containing billions of stars and gas in orbit about its center. It takes our sun about 250 million years to complete one orbit about the center of the Milky Way. Galaxies come in a variety of shapes. Some are nearly spherical, while others, like the Milky Way, are flattened disks with a bulge in the middle. Individual galaxies are separated from one another like atolls in a Pacific archipelago; clusters of galaxies are like individual island chains scattered across the vast ocean of nearly empty space.
The Evolution of Galaxies
It was only in the 1920s that astronomers realized that many of the fuzzy patches revealed by their telescopes were indeed distant assemblages of stars, that is, galaxies. For many years after their discovery, galaxies were assumed to be fixed and unchanging, but by the 1970s, astronomers realized that galaxies should, in fact must, evolve. Stars alter the chemical composition of interstellar gas in a one-way process that builds heavier and heavier atoms. Thus the chemical composition, color, and luminosity of a galaxy should all change in time. In addition, galaxies can evolve dynamically as giant galaxies cannibalize their smaller companions.
Finding direct evidence for the evolution of single galaxies is not easy. But, fortunately, light travels at a finite speed. Since the distances in space are large, we can use this effect to observe evolution directly. When we take a picture today of the Andromeda Galaxy, 2 million light-years away, we see that galaxy as it was 2 million years ago. When we look at a galaxy in the Virgo cluster of galaxies, 50 million light-years away, we see light that was emitted 50 million years ago. Looking deeper into space is looking further back in time. Telescopes are time machines. With larger telescopes, we can see more distant galaxies at earlier stages of evolution.
Unfortunately, the light from distant galaxies is faint. To detect such feeble light, astronomers need large telescopes equipped with sensitive detectors. With new telescopes and more sensitive electronic cameras, astronomers have begun to study galaxies at much greater distances and thus at much earlier stages of evolution. For example, some observational evidence suggests a systematic color change of galaxies with age, as theoretically predicted.
For the 1990s, astronomers are building several large, visible-light and infrared telescopes with diameters ranging from more than 300 in. (8 m) to nearly 400 in. (10 m), far larger than the 200-in. telescope at Palomar Mountain, California, which was completed in 1949. In the 1990s astronomers hope to see more distant galaxies, at a much earlier stage of evolution than any galaxies previously seen. What types of stars inhabit young galaxies? While individual stars are born and die, how does the bulk population of stars in a galaxy age in time? Does the shape of a galaxy change in time, or is it determined completely
when a galaxy first forms? How does the total luminosity of a galaxy change in time? Like stars, galaxies are found in congregations. These congregations are called groups and clusters of galaxies. How do the neighboring galaxies in groups and clusters affect each other?
Astronomers believe that many galaxies went through an extremely energetic early phase of evolution in which almost all of their energy was produced in their centers. This belief is based on the discovery of quasars in the 1960s. On photographic plates quasars resemble stars, yet are as distant as and far brighter than entire galaxies. Evidently, an enormous amount of energy is produced in a tiny volume of space, perhaps as small as our solar system. Importantly, most quasars have been found far away. Since distance translates into time in astronomy, we can infer that most quasars lived and died in the distant past. Quasars are the dinosaurs of the cosmos. Astronomers theorize that quasars constituted the central regions of some galaxies at a very early stage of their evolution.
The new generation of large, visible-light and infrared telescopes and the already launched Hubble Space Telescope may be able to detect the weak light of infant galaxies harboring quasars and to advance the study of the connection between quasars and galaxies. New infrared telescopes, such as SIRTF and the ground-based, infrared-optimized 8-m telescope, will also play important roles in quasar research. Finally, new radio telescopes with extremely high angular resolution, particularly the Very Long Baseline Array, should be able to make radio-wave images of quasars themselves.
Great dust clouds apparently surround many quasars, absorbing their visible light and turning it into infrared radiation. In the 1980s, IRAS discovered extremely luminous galaxies emitting 90 percent or more of their energy as infrared radiation and apparently harboring quasars at their centers. Furthermore, many such galaxies appeared to be colliding with other galaxies (Plate 2.12). Could collisions of galaxies give birth to quasars, or refuel them? With its much greater sensitivity, SIRTF should be able to study the nature and evolution of these curious infrared galaxies. If a big fraction of quasars are produced by collisions of galaxies, SIRTF will make it possible for astronomers to find out.
The Power Source of Quasars and Active Galaxies
Quasars and active galaxies emit copious amounts of power across a broad range of wavelengths, from radio waves to x-rays. Where does their great energy come from? It certainly cannot come entirely from stars that radiate predominantly visible light. Furthermore, stars live on nuclear energy, converting matter into energy with an efficiency of less than 0.1 percent. Nuclear energy is not efficient enough to balance the huge energy budget of quasars and active galaxies. Finally, the stars in a galaxy are scattered about, while the energy of active galaxies is produced in a highly concentrated region
smaller than a light-year in diameter. Even if a sufficient number of stars were somehow crammed into such a small volume, the resulting stellar system would be so dense that the stars would quickly collide with each other and coalesce into a single unstable, massive object.
For these reasons, most astronomers believe that quasars and active galaxies can be powered only by gravitational energy released at the center of the system. According to current ideas, a massive black hole, with a mass of a million to a billion times that of our sun, inhabits the middle of an active galaxy or quasar. Surrounding gas and stars fall under the gravitational grip of the central black hole. As gas plunges toward the black hole, it releases gravitational energy, which is then transferred into high-speed particles and radiation. Matter falling toward a black hole can convert 10 percent of its mass into energy before it enters the black hole and is never heard from again.
A key to understanding quasars and active galaxies is the mechanism for feeding gas to the central black hole. Is it constant or intermittent? What triggers it? Possible sources of gas include ambient gas in the central regions of the galaxy, the gravitational shredding of hapless stars that wander too close to the black hole, the disintegration of stars by collisions with each other, and the agitation of one galaxy by a close encounter or merger with another. For the more luminous active galaxies and the less luminous quasars, gas must be fed to the central black hole at the rate of about a sun's worth of mass per year. Isolated black holes, no matter how massive, produce very little energy. Thus an understanding of the environment of the central black hole and how it is fueled may be crucial to understanding why some galaxies are highly energetic and others are not.
How can we test the hypothesis of massive black holes? A massive black hole, even as massive as a billion times the mass of our sun, would have a diameter smaller than our solar system. At the distance of the nearest big galaxy, 2 million light-years from us, such a black hole would have an angular size of only a few billionths of a degree, too small to be seen by any telescope in the near future. However, a massive black hole might reveal itself by the way that it affects the motions and positions of surrounding stars. Trapped by the gravity of the hole, surrounding stars would huddle together closely and would hurtle through space more rapidly than if no black hole were present. Infrared observations by a telescope carried aloft in the Kuiper Airborne Observatory have revealed rapid orbital motions in the center of our galaxy, indicating the possible presence of a black hole with a mass of a few million solar masses. Hints of these effects have also been found in a number of nearby galaxies, and the Hubble Space Telescope will look at a larger sample of more distant galaxies.
Black holes might also be indirectly identified by the high-energy emission of the surrounding gas. It is believed that gas near a black hole orbits it in a flattened disk, similar to a protoplanetary disk, only hotter and much more
massive. Some of the gas in the disk, heated to temperatures between 1 million and 1 billion degrees celsius, would radiate x-rays. Such x-ray emission will be studied by AXAF. Another characteristic feature of such high energies could be the production of electrons and their antiparticles, positrons. Once produced, particles and antiparticles annihilate each other in a burst of gamma rays. The Gamma Ray Observatory, to be launched by NASA in 1991, and other proposed gamma-ray detectors in space, will search for such gamma rays. Ground-based telescopes may have found gamma rays of extremely high energy coming from astrophysical objects: pulsars, x-ray binary stars, and black hole candidates. The gamma rays are detected by light or particles produced in showers when the gamma rays enter the earth's atmosphere. Experimental physicists are working on better ways to detect this radiation, and theorists are studying novel astrophysical mechanisms to explain its existence.
The mysterious “x-ray background” that has puzzled astronomers for 25 years may also yield its secrets to new x-ray satellites. They should be able to indicate definitely what part of radiation comes from hot gas, what part from distant quasars, and what part, if any, from still more exotic objects.
The dramatic and energetic behavior of active galaxies and quasars raises other questions. Some of the observed gaseous “jets” emanating at great speed from these objects are extremely narrow and well collimated. What produces and controls such columns of matter? Recently, some progress has been made on these questions. The jets radiate x-rays, visible light, and radio waves. Images made with large visible-light telescopes, radio telescopes, and the Einstein Observatory x-ray telescope of the 1980s have revealed exquisite details of the structures and blobs in the gaseous jets (Plate 2.13 and Plate 2.14). In particular, radio observations with very high angular resolution show that new concentrations of gas enter into the jet every year or so. New radio interferometers should be able to investigate jets and the central sources at much higher resolution.
Continued theoretical work in the coming decade will also be crucial to understanding energetic jets. In the 1980s, some of the features of jets were reproduced in large computer simulations (Plate 2.15). In such simulations, the scientist programs the computer with the basic laws of physics describing how gas, radiation, and magnetic fields behave, sets up some initial configuration of matter and radiation, and then lets the computer calculate how the system evolves in time. Comparison to observation then guides refinements of the theory and suggests new observations.
The Birth of Galaxies
If we peer out far enough into space, we should see back to the epoch of galaxy formation. What should a young galaxy look like? Astronomers are not sure. Galaxies were probably formed about 10 billion to 20 billion years ago,
perhaps a few hundred million years after the beginning of the universe itself. But little is known about infant galaxies and even less about galaxies being born.
Astronomers generally suppose that the mass in the universe long ago was smoothly spread about, but was bunched up very slightly here and there, like ripples on a pond. The origin of these initial ripples is still unknown. In a place where the mass was bunched up, gravitational forces were slightly stronger. This caused nearby mass to bunch up more, attracting more surrounding gas. The force of gravity then became even stronger, and the process continued until a stronger concentration of mass formed. For sufficiently large concentrated regions, the inward pull of gravity exceeded the outward force of pressure, and the region collapsed into a dense and coherent structure. Just as for individual stars, the collapse of a giant gaseous cloud to form a galaxy might be affected in later stages by the forces of gas pressure, radiation, and rotation as well as the force of gravity. In the last decade, a great deal of theoretical work has been done to understand competing models for the formation of galaxies.
Once galaxy formation is under way, do galaxies continue to grow by accumulating gas in their vicinity, or do they reach their final size rather quickly? What determines the shape of a galaxy? Why are some galaxies nearly spherical, while others appear as flattened disks? In spiral galaxies, like our own, does the central bulge form first and then the disk, or vice versa? What determines the odd features of some galaxies—the rings and warps and bars? Were these features built in at the beginning, or did they form later, as the result of gravitational forces within the galaxy? Or were they fashioned by a close encounter or merger with another galaxy?
In reaching their conclusions, theoretical astronomers have had little help from observations made to date, since no protogalaxy has yet been convincingly identified. The telescopes of the 1990s may change this state of affairs. A number of telescopes, including the ground-based 8- and 10-m telescopes, the orbiting Hubble Space Telescope with its infrared camera, and SIRTF, will all survey the sky for the faint wisps of light emitted by the first generation of stars forming in infant galaxies. Infrared wavelengths will be used, since the effect of the cosmic expansion is to shift the intrinsic blue light of young stars to redder and redder wavelengths.
Finally, important hints about galaxy formation lie in the gas between galaxies, called the intergalactic medium. As in the interstellar medium within individual galaxies, this gas is enriched with the various chemical elements manufactured within stars. And the intergalactic medium is the material out of which galaxies formed. A primary tool for analyzing the intergalactic medium has been the study of radiation from quasars. As this radiation travels from there to here, it passes through the intergalactic medium, and some of it is absorbed. The particular wavelengths absorbed indicate the chemical makeup of the intervening gas. How does the composition of the intergalactic gas
change in time? Can this gas be used to date the epoch when galaxies first formed? Analysis of the intergalactic medium at large distances requires both a high sensitivity to dim light and the ability to disperse the incoming light into its component wavelengths. The needed abilities are beyond current ground-based 4-m telescopes but within reach of the Hubble Space Telescope and the 8- and 10-m telescopes of the coming decade.
Still further back in time the universe consisted of smoothly distributed, hot gas. It emitted radiation that we should be able to see today. In 1965, astronomers did discover a bath of radio waves filling all space. It is believed that this radiation has been traveling freely through space, cooling as it goes, since the universe was only about 300,000 years old. At that time, the enormously hot energy of the cosmic fireball dominated the mass of the universe. Imprinted on this cosmic background radiation should be a record of the distribution of cosmic matter at that time, well before the epoch of galaxy formation. Irregularities in the distribution of matter at that time should be detectable now. They would show up today as variations in the intensity of the radiation detected by our radio telescopes pointed in different directions—and indeed all theories of the formation of galaxies demand the existence of such variations.
So far, to our puzzlement, no variations have been observed. From the measurements of the COBE, an orbiting satellite launched in 1989, and from other experiments, astronomers have recently determined that any variations in the intensity of the cosmic background radiation must be less than several parts in 100,000. Some theories of galaxy formation have been demolished by this fact. Revised theories that require the existence of large amounts of so-far undetected matter predict variations 10 times smaller. In the next decade, detectors now being developed should have the sensitivity required to challenge the new theories by looking for temperature variations of 1 part in 1 million. If no variations are found at these increased sensitivities, then theoretical extragalactic astronomy will be thrown into crisis. Something will be seriously wrong—either with our theories of galaxy formation or with our understanding of the cosmic background radiation. From such confrontations of theory with observations, deeper understanding emerges.
THE LIFE HISTORY OF THE UNIVERSE
The Big Bang Model
As we look further and further into space, will we come to an edge of space or a beginning of time? If the universe had a beginning, how did it begin? Will it have an end? These are questions in cosmology, the branch of astronomy concerned with the structure and evolution of the universe as a whole.
Every culture has had a cosmology. Aristotle's universe had an edge of
space, an outermost sphere upon which were fastened the stars. But the cosmos had no beginning, or end, of time. In On the Heavens, Aristotle wrote that the “primary body of all is eternal, suffering neither growth nor diminution, but is ageless, unalterable and impassive.” The Judeo-Christian world view did away with eternity but maintained the idea of a cosmos without change. According to this tradition, the universe was created from nothing by God and has remained much the same ever since. Copernicus, who in 1543 demoted the earth to a mere planet in orbit about the sun, changed the way we look at many things, but not the Aristotelian belief in a universe both spatially finite and static in time. In 1576, Thomas Digges became the first Copernican to pry the stars off their crystalline spheres and spread them throughout an infinite space. Now, the universe was without limit in space. However, the universe was still viewed as unchanging in time. Newton argued the same view a century later. Individual planets moved in the sky, to be sure, but the universe as a whole remained the same from one eon to the next.
Modern theories of cosmology date back to Albert Einstein's 1917 theory of general relativity and the models of the Russian mathematician and meteorologist Alexander Friedmann, who found a solution of Einstein's theory of gravity that described a universe that began in a state of extremely high density and then expanded in time, thinning out as it did so. Friedmann's model eventually came to be called the “Big Bang” model.
In 1929, the American astronomer Edwin Hubble, after whom the Hubble Space Telescope is named, made what was perhaps the most important observational discovery of modern cosmology: the universe is expanding. The universe is not constant in time; it is evolving and changing. Specifically, Hubble concluded that the other galaxies are moving outward from us in all directions. Hubble discovered that the distance to each galaxy is proportional to its recessional speed. A galaxy twice as far from us as another galaxy is moving outward twice as fast. This quantitative result had been predicted for a homogeneous and uniformly expanding universe.
If the galaxies are moving away from each other, then they were closer together in the past. At earlier times, the universe was denser. Continuing this extrapolation backwards suggests that there must have been a definite moment in the past when all the matter of the universe was compressed together in a state of enormous density. From the rate of expansion, astronomers can estimate when this point in time occurred: about 10 billion to 20 billion years ago. It is called the origin of the universe, or the Big Bang. As mentioned earlier, radioactive dating of rocks and meteorites, begun a couple of decades before Hubble's discovery, suggests that the sun and the earth are between 4 billion and 5 billion years old. Thus with two totally different methods, one using the outward motions of galaxies and the other using the rocks underfoot, scientists have derived roughly similar ages for the universe. This success has been a powerful argument in favor of the Big Bang model. However, it is important
to remember that cosmology, of all the branches of astronomy and indeed of all the sciences, requires the most extreme extrapolations in space and in time.
The Big Bang model, although widely accepted, rests on a rather small number of observational tests. In addition to explaining the observed expansion and age of the universe, the Big Bang model has successfully met two other major tests against observations. Calculations showed that the material of the universe should be, and measurements show that it is, approximately 74 percent hydrogen and 24 percent helium. The model also correctly predicts the abundances of other light elements such as deuterium and lithium.
In addition, the hot infant universe would have produced blackbody radiation that is easily identifiable by its universal spectrum ( Figure 2.2). The most precise measurements of the cosmic background radiation have come from the COBE satellite, which has confirmed that the spectrum of the cosmic background radiation is extraordinarily close to that predicted by the Big Bang model.
According to the Big Bang model, hydrogen, helium, and some of the light elements, as well as the cosmic background radiation, were all created in the universe long ago, when the universe was very different from what it is
today. Imagine a movie of cosmic evolution played backward in time, starting from the present. The universe contracts. The galaxies move more and more closely together. As the universe grows ever denser, stars and galaxies lose their identity, and the matter of the universe begins to resemble a gas. Like any gas being compressed, the cosmic gas becomes hotter and hotter. Eventually, the heat becomes so high that atoms cannot retain their electrons, and they disintegrate into atomic nuclei and freely roaming electrons. At a still earlier stage, closer in time to the Big Bang, the atomic nuclei themselves disintegrate into their constituents, protons and neutrons. At an even earlier time, each proton and neutron disintegrates into elementary particles called quarks. The universe becomes a roiling sea of subatomic particles. Very close to the Big Bang, the subatomic, or quantum, effects of gravity itself become important. At present, we have no theory that can handle quantum gravity, although some first steps have been taken in this direction. The concept of the Big Bang is the ultimate challenge to our understanding of physical laws.
The Large-Scale Structure of the Universe
The assumption of perfect homogeneity that underlies the Big Bang model is obviously not valid in the region we have observed with our telescopes. The universe nearby is not evenly filled with a smooth and featureless fluid. Rather, it is lumpy. Matter clumps into galaxies, and galaxies huddle together in groups and clusters of galaxies, and so on. Given the local lumpiness, the assumption of homogeneity means that space should appear smooth when averaged over a sufficiently large volume, just as a beach appears smooth when looked at from a distance of a few feet or more, even though it appears grainy when looked at from closer range.
Groupings of galaxies, which astronomers call “structures,” are intriguing in their own right. Structures of various sizes abound, and astronomers want to understand the nature of these structures and how they were born. Were structures formed completely by gravitational attraction, or were other forces involved? What shapes do the structures form? How large is the largest structure? Until such questions are answered, it will be hard to decide whether the observed inhomogeneities and structures are simply details in the standard view or hints of a radically different picture.
Much evidence for inhomogeneities has been obtained in recent years, including the discoveries of chains of galaxies, sheets of galaxies, and giant voids with very few galaxies in them. In the late 1980s, American astronomers found evidence for the largest structure so far known, a “wall” of galaxies stretching at least 500 million light-years. The new three-dimensional maps of large samples of galaxies, produced from surveys of galaxy redshifts, have been made possible by advances in technology that allow the redshifts of galaxies to be measured with fast and automated procedures. Under the assumption that the
universe is approximately homogeneous and uniformly expanding, the redshift of a galaxy translates into an approximate distance, thus providing the elusive third dimension.
It is not yet known whether the new cosmic structures found in a few selected regions of space are typical. What seems clear is that structures of some kind have been found at the largest possible scale in each survey of galaxies; that is, a survey that looks over a region of 100 million light-years usually finds some chain or wall or absence of galaxies extending roughly 100 million light-years in size; a survey of a 200-million-light-year region finds structures of 200 million light-years, and so on.
Galaxy surveys are now in progress that extend out to a few billion light-years. In addition, surveys with many more galaxies are being planned. The largest redshift surveys to date include only several thousand galaxies and sample only limited directions in the sky. In the coming decade, astronomers hope to initiate surveys of a million galaxies. Such surveys could be accomplished with moderate-sized visible-light and infrared telescopes. If in the future we find filaments and bubbles and voids with sizes of a few billion light-years, several times larger than those now mapped, then there would be a direct contradiction with the uniformity of matter implied by the cosmic background radiation. The Big Bang model might even be called into question.
On the theoretical side, astronomers are attempting to make sense of the observed positions and motions of galaxies by the use of large computer simulations. Such simulations involve 10,000 to several million particles, each representing a portion of a galaxy or a number of galaxies. The particles are placed at initial positions, given an initial outward velocity corresponding to the expansion of the universe, and then allowed to interact via their mutual gravity. The hypothetical galaxies fly around the computer screen, gravitate toward each other, and form clumps and wisps and voids. The largest simulations require the fastest and biggest computers in the world. By comparing computer simulations to the observed large-scale structure of the universe (Plate 2.16) scientists hope to test their assumptions about the initial conditions and forces at work in the cosmos. The current computer simulations, although 10 times larger than those of a decade ago, still do not have enough particles for a decisive comparison between theory and observation. Within the coming decade, larger computers and new methods for using those computers should give more reliable answers. And the new calculations will have more brain as well as brawn. Additional physics will be taught to the computers, and the resulting simulations will be more realistic and believable.
Whatever the outcome of the computer simulations, the ultimate theory of the distribution of matter in the universe must be consistent with all the observations. Astronomers have become increasingly worried about reconciling the smoothness of the cosmic background radiation with the lumpiness of matter
nearby. Both theoretically and observationally, an understanding of the large-scale structure of the universe is at the top of most lists of the outstanding problems in cosmology.
An obstacle to understanding the distribution of mass in the universe and the motions of galaxies is that at least 90 percent of the detected mass in the universe is not seen. No radiation of any kind—not visible light, nor radio waves, nor infrared, nor ultraviolet, nor x-rays —has yet been detected from this “dark matter.” It is invisible to our instruments. We know that dark matter exists, because we have observed its gravitational effects on the stars and galaxies that we see, but we have little idea what it is.
The conundrum was uncovered in 1933 by Fritz Zwicky, who estimated the mass of a cluster of galaxies in orbit about one another by measuring the amount of gravity needed to hold the cluster together. He discovered that the total mass thus inferred was about 20 times what could be accounted for by the visible stars in the cluster. Zwicky's startling discovery was not appreciated by most astronomers until the 1970s, when it was confirmed by new observations of the orbital motions of stars and gas in individual galaxies. The amount of matter needed to produce these motions is far larger than the observed light-emitting matter. On the scale of both individual galaxies and of clusters of galaxies, most of the mass inferred to exist from the motions of stars and galaxies is not seen. Large ground-based optical and radio telescopes of the 1990s will be the instruments of choice to map the motions of stars and gas at the periphery of galaxies.
What is the nature of dark matter? Does it consist of numerous dark objects, like dim red stars, planets, or black holes, or does it consist of subatomic particles that interact with other matter only through gravity? Dark matter could alter our theories of the formation of galaxies or of subatomic particles. Whatever it is, astronomers have been startled by the realization that the luminous matter they have been stating at and pondering for centuries may make up a mere tenth of the inventory. And it is not just the unknown identity of dark matter that causes concern. Its quantity and arrangement in space are also uncertain, foiling attempts to understand why the luminous mass is arranged as it is.
A better understanding of how dark matter is distributed in space, particularly over the largest possible distances, may be the key to many of the puzzles discussed earlier. A careful reconciliation of the velocities of galaxies with the observed inhomogeneities in luminous matter should reveal the presence of dark matter, which contributes to the velocities through its gravitational effects. Detailed maps of the positions and motions of galaxies carried out on
ground-based optical and radio telescopes may yield better maps of the location of dark matter.
Dark matter will also be mapped in the coming decade by x-ray emission from hot gas. Very hot gas has been detected inside large clusters of galaxies extending 5 million to 10 million light-years out from the center of many clusters. So hot that it should boil away, the gas is evidently held by the gravity of invisible matter. From the precise distribution of the gas, astronomers can work back to infer the gravity confining it and the distribution of dark matter producing that gravity. In the coming decade, the German X-ray Roentgen Satellite (ROSAT) already in orbit, the Japanese x-ray satellite ASTRO-D, and especially NASA's AXAF telescope will make better maps of the distribution and temperature of the hot gas in galaxy clusters.
The identification of dark matter may be a process of elimination. For example, the dark matter could be large planets, with masses between a thousandth and a tenth the mass of our sun. Such objects should have enough heat generated by their slow contraction to emit a low intensity of infrared radiation. A highly sensitive infrared telescope like SIRTF may be able to detect them. In particular, SIRTF will scrutinize the far reaches of our own galaxy, where dark matter may be lurking, and search for a faint excess glow.
There are other ways to probe dark matter. One of the most recent and potentially very important new techniques makes use of the “ gravitational lens” phenomenon. When he published his new theory of gravity, Einstein pointed out that light, like matter, should be affected by gravity. Thus as light from a distant astronomical object, such as a quasar, travels toward the earth, that light should be deflected by any mass lying between here and there. The intervening mass can act as a lens, distorting and splitting the image of the quasar. Even if the intervening mass is totally invisible, its gravitational effects are not. By carefully analyzing the distortions of quasar images, astronomers can reconstruct many of the properties of the intervening gravitational lens, including its distribution in space and total mass. Gravitational lenses were first discovered only a decade ago; about a dozen have been found since that time. In the coming decade, the gravitational-lens phenomenon will be used as a powerful tool to uncover the nature of dark matter.
Alternatively, dark matter could consist of individual, freely roaming sub-atomic particles, rather than aggregates of particles such as planets. The possibilities have stirred the imaginations of particle theorists. Dozens of particles have been proposed, some on the basis of new theories of subatomic physics. None, however, have as yet been seen in the laboratory. If dark matter does indeed consist of these exotic particles, then it may be identified in the laboratory rather than at the telescope. Within the last few years, the first laboratory detectors have been built to search for some of these hypothesized particles. The experiments are extremely difficult, owing to the elusiveness of
the particles, and it is estimated that detectors of the next decade need to be approximately 100 times more sensitive before the particles can be found—if they exist.
The Origin of the Universe
The goal of physics is to explain nature with as simple a theory as possible, and the goal of cosmology is to explain the large-scale structure and evolution of the universe in terms of that theory. Astronomers and physicists today believe that many properties of the present universe probably depend on what happened during the first instants after the Big Bang. One such property, ironically, is the apparent uniformity of the universe on the large scale, as evidenced by the cosmic background radiation. Although such uniformity and homogeneity have been assumed in the Big Bang model, they still must be explained, or at least be made plausible. It seems unlikely to many scientists that the universe would have been created so homogeneous by accident.
In the 1970s an important change occurred in theoretical cosmology. Physicists with expertise in the theory of subatomic particles joined astronomers to work on cosmology. The physicists brought a fresh stock of ideas and a new set of intellectual tools to bear on the question of why the universe has the properties it does, not just what those properties are. Particle physicists in the United States and in the Soviet Union proposed a modification to the Big Bang model called “the inflationary universe” that has caused a major change in cosmological thinking. The essential feature of the inflationary universe model is that, shortly after the Big Bang, the infant universe went through a brief and extremely rapid expansion, after which it returned to the more leisurely rate of expansion of the standard Big Bang model. By the time the universe was a tiny fraction, perhaps 10−32, of a second old, the period of rapid expansion, or inflation, was over. The epoch of rapid expansion could have taken a patch of space so tiny that it had already homogenized and quickly stretched it to a size larger than today's entire observable universe. Thus the inflationary expansion would make the universe appear homogeneous over an extremely vast region, far larger than any region from which we have data.
The inflationary model makes specific predictions about the formation of structures in the universe. In particular, processes in the early universe would have determined the nature of the initial inhomogeneities that later condensed into galaxies and clusters of galaxies. Some of the predictions of this model will be tested by the COBE satellite and by galaxy surveys to be made in the 1990s with new optical telescopes. Whether these and other observations will confirm the inflation model or lead theorists to a different approach altogether is, of course, unknown. But astronomers and physicists will be working together to learn the past history of the observable universe.
The End of the Universe
As the universe expands, its parts pull on one another owing to gravitational attraction, and this slows down the expansion. The competition between the outward motion of expansion and the inward pull of gravity leads to three possibilities for the ultimate fate of the universe. The universe may expand forever, with its outward motion always overwhelming the inward pull of gravity, in the way that a rock thrown upward with sufficient speed will escape the gravity of the earth and keep traveling forever. Such a universe is called an open universe. A second possibility is that the inward force of gravity is sufficiently strong to halt and reverse the expansion, just as a rock thrown upward with insufficient speed will reach a maximum height and then fall back to the earth. A universe of this type, called a closed universe, reaches a maximum size and then starts collapsing, toward a reverse big bang. This universe grows smaller and hotter, and has both a beginning and an end in time. The final possibility, called a flat universe, is analogous to a rock thrown upward with precisely the minimum speed needed to escape from the pull of the earth. A flat universe, like an open universe, keeps expanding forever. Meanwhile, stars and galaxies evolve, heavy elements are synthesized, radioactive decay transmutes elements, and the universe grows colder and colder.
The Big Bang model allows all three possibilities. Which one holds for our universe depends on how the cosmic expansion began, in the same way that the path of the rock depends on the rock's initial speed relative to the strength of the earth's gravity. For the rock, the critical initial speed is 7 miles per second. Rocks thrown upward with less than this speed will fall back to the earth; rocks with greater initial speed will never return. Likewise, the fate of the universe was, according to the Big Bang model, determined by its initial rate of expansion relative to its gravity. Even without knowledge of these initial conditions, however, we can infer the fate of our universe by comparing its current rate of expansion to its current average density. If the density is greater than a critical value, which is determined by the current rate of expansion, then gravity dominates; the universe is closed, fated to collapse at some time in the future. If the density is less than the critical value, the universe is open. If it is precisely equal to the critical value, the universe is flat. The ratio of the actual density to the critical density is called omega. Thus the universe is open, flat, or closed depending on whether omega is less than 1, equal to 1, or larger than 1, respectively.
Omega can be measured. According to the best current measurements, the critical density of the universe, as determined by the rate of expansion, is equivalent to a few atoms of hydrogen in a box a meter on a side, roughly the density achieved by spreading the mass of a postage stamp through a sphere about the size of the earth. The average density of all the matter we can detect by its radiation or by its gravitational effect is about one-tenth this critical value.
This result, as well as other observations, seems to suggest that our universe is of the open variety.
There are uncertainties in these estimates, mostly connected with uncertainties about cosmic distances. If the universe were precisely homogeneous and uniformly expanding, then the rate of expansion of the universe (the Hubble constant) could be determined by measuring the recessional speed and distance of any galaxy, near or far. Conversely, the distance to any galaxy could be determined from its redshift and the application of Hubble's law. The average density of matter could be figured by estimating the mass of a group of galaxies and then dividing by the volume of space that it occupies. However, the universe is not at all homogeneous; structure is apparent on every scale we have studied. Because of local inhomogeneities, the rate of expansion of the universe and the average density of matter need to be measured over as large a region as possible. Accurate determinations of distances to galaxies are needed for both of these measurements.
One possible means of accurately determining distance involves the scattering of the cosmic background radiation by hot gas in clusters of galaxies by a process called the Sunyaev-Zeldovich effect. The hot gas gives a slight energy boost to the radio waves as they pass through it on their way to the earth. As a result of measuring both the change in energy of the radio waves and the x-ray emission from the hot gas, the distance to the cluster of galaxies will be well determined. With the MMA, AXAF, and other instruments, astronomers hope to make these measurements in the coming decade. Such measurements repeated for a large number of galaxy clusters would permit a more accurate determination of the rate of expansion of the universe.
Likewise, studies of the velocities and distances to a large number of galaxies using new 4-m telescopes equipped with spectrographs of novel design are capable of measuring the distances to hundreds of galaxies at a time. These data could pin down the local values of both omega and the Hubble constant. Peculiar velocities of galaxies depend on the extra amount of matter concentrated in a region, over and above the average density of cosmic matter. Measurements of peculiar velocities, together with a knowledge of how much matter there is above the average, lead to an estimate for omega.
Despite these difficulties, cosmologists are almost positive that the value of omega lies between 0.1 and 2. Enough matter has been identified so that omega cannot be less than about 0.1. On the upper end, an omega larger than 2, together with the current rate of expansion, would translate to an age of the universe that is less than the age of the earth as determined by radioactive dating.
The inflationary universe model firmly predicts that omega should be equal to 1, exactly. On this basis, the model can in principle be either ruled out or supported from observational evidence. Since current observations suggest a value of omega closer to 0.1, scientists who believe on theoretical grounds that
the model is right must have faith that an enormous amount of mass is hiding from us, escaping detection, perhaps in a uniform and tenuous gas of particles between galaxies.
To summarize, the light-emitting matter we see accounts for sufficient mass to make omega about 0.01; the unseen but gravitationally detected dark matter accounts for another factor of 10 of mass, increasing omega to about 0.1. Advocates of the inflationary universe model must hypothesize that space contains yet 10 times more mass—not only unseen but undetected and composed of some exotic species of matter. As mentioned earlier, some candidate particles will be searched for in the laboratory using “dark matter” detectors currently under development.
Some of the general features of the inflationary universe model are so appealing that many astronomers and physicists believe that some form of the idea is correct. It is sobering to realize that the highly influential inflationary universe model was unknown only a decade ago. Like Supernova 1987A, the idea exploded. We should expect similar observational and theoretical surprises in the future.
Even if the cosmos is infinite in extent, only a limited volume is visible to us at any moment: we can see only as far as light has traveled since the Big Bang. As we look farther into space, we are seeing light that has traveled longer to reach us. Eventually, at some distance, the light just now reaching our telescopes was emitted at the moment of the Big Bang. That distance marks the edge of the currently observable universe, some 10 billion to 20 billion light-years away. We cannot see farther because there has not been time for light to have traveled from there to here. And we have no way of knowing what lies beyond that edge. Some theorists have recently proposed that extremely distant regions might have different properties from the cosmos we know—different forces, different types of particles, even different dimensionalities of space. In such a universe, it would be impossible for us ever to learn about more than a tiny fraction of the possibilities and realities of nature.
Even in the universe we can see, many fundamental surprises surely await us. It is likely that major properties of the universe are yet unknown. The expansion of the universe was unknown in 1920 and the existence of quasars unsuspected in 1960. Who can imagine what astronomers will find by the year 2000?