Introduction to In Search of Other Worlds
Astronomers in the United States reported in March 2000 that they had discovered the smallest planets yet found outside the solar system. Although the researchers did not see the planets directly, they inferred their existence by measuring variations in the light from the planets' parent stars. The finding of relatively small bodies was important because it showed that the methods for locating extrasolar (beyond the solar system) planets were becoming more sensitive. The discovery, by astronomers Geoffrey Marcy of the University of California at Berkeley and San Francisco State University and Paul Butler of the Carnegie Institution of Washington, D.C., was the latest in a string of findings of such planets. It capped a five-year period in which the number of known planetary systems grew from 1—our own—to more than 30.
Astronomers in 2000 expected to find planets around even more stars. The discoveries up to then had confirmed the longstanding belief of astronomers that planets are common in the universe. The planets they had found, however, had very surprising properties, forcing astronomers to rethink their ideas about how solar systems form. And perhaps the most important question of all remained unanswered: Are there other planets similar to Earth that are capable of supporting life?
The idea that planets might orbit other stars has a long history. Some ancient astronomers believed that the sun and the planets all orbit around the Earth, and that the “fixed stars,” as they called them, were merely points of light dotting a celestial sphere. In 1543, the Polish astronomer Nicolaus Copernicus published his revolutionary theory that the sun is the center of the solar system and that the Earth is one of the planets that circle the sun. The Italian philosopher and monk Giordano Bruno took Copernicus's idea further. Bruno claimed that the universe is infinite, that the stars are other suns around which other planets orbit, and that life is widespread throughout the universe. For these and other beliefs unacceptable to the Roman Catholic Church, he was burned at the stake in Rome in 1600.
Despite Bruno's fate, many scientists after Copernicus speculated that planets around other suns may be common. Modern astronomers found support for that idea in their observations of disks of gas and dust surrounding young stars. The material in a disk, according to current theories of how our own solar system formed, gradually comes together to form planets.
Planets Orbiting A Pulsar
The first planets discovered outside our solar system were two objects orbiting a body called a neutron star. The discovery was reported in 1992 by astronomers Alexander Wolszczan at Cornell University's Arecibo Radio Telescope in Puerto Rico and Dale A. Frail at the U.S. National Radio Astronomy Observatory in Socorro, New Mexico. A neutron star is a small, incredibly dense object made of neutrons (electrically neutral particles in an atomic nucleus). This object is left behind when a massive star explodes and its core collapses due to the force of gravity. Some neutron stars, called pulsars, emit a beam of radio waves. As the star spins, the beam sweeps through space. If the beam from such an object passes across the Earth, astronomers can detect it as flashes, or pulses, of energy in the radio portion of the electromagnetic spectrum. The intervals between flashes are very regular. When Wolszczan and Frail measured the frequency of flashes from a pulsar called PSR1257+12, however, they found that the time between pulses varied. The astronomers attributed the changes to the gravitational pull of two planets orbiting the pulsar. The planets alternately tugged the pulsar away from Earth, increasing the time it took the radio beam to reach Earth, then toward Earth, decreasing the time.
That planets could be circling a pulsar was hard to believe, because the massive explosion of the original star would have vaporized any objects in its vicinity or flung them out into space. Wolszczan and Frail theorized that the planets formed after the explosion, possibly from the debris of a companion star that was destroyed in the initial blast. Because of the powerful radiation continually emitted by the pulsar, these planets are not likely to harbor life. Such objects were not at all what astronomers had in mind when they began searching for planets around other suns.
Finding Extrasolar Planets
It was, of course, normal stars in which astronomers were most interested. Detecting a planet orbiting a regular star is difficult, however, because the star's bright glare overwhelms the meager reflected light of any planets it may have. Throughout the 1990's, almost all extrasolar planets were discovered by indirect means, using one or more of three techniques. In most cases, astronomers indirectly found planets by measuring slight variations in a star's motion caused by the gravitational effect of planets. Another indirect method called the transit technique involved observing the slight lessening in a star's light caused when a planet passes in front of it. A third technique made use of a phenomenon predicted by the renowned German-born physicist Albert Einstein in which a star and its planets can act as a magnifying glass to increase the brightness of a more distant star.
Detecting a planet based on gravitational pull is the most successful indirect method because the pull that a planet exerts on its star is relatively easy to detect. Planets affect the position and motion of their stars because both the planet and the star revolve around a common center of mass (gravitational balance point). If the star and the planet were of equal mass, the center of mass would be midway between the two, and both would orbit at equal distances from that point, just as two people of equal weight create a balance point in the center of a seesaw. But because a star's mass is much larger than that of even a large planet, the center of mass is closer to the star. This situation is like having two people of different weights sitting on a seesaw, in which case the balance point would be closer to the heavier person. As a planet and its sun orbit their common center of mass, the planet's gravity tugs at the star, causing the star to wobble or shift slightly in its orbit. The heavier the planet, the larger the star's resulting shift.
In our solar system, Jupiter is the most massive planet, but still so much smaller than the sun that the center of mass lies just above the sun's surface. Alien astronomers trying to determine whether the sun has planets might not be able to see Jupiter or the other planets. But they might find evidence of Jupiter's existence by detecting the small motion of the sun about the balance point just outside its own radius.
Using the Doppler Shift
The orbital motion of a star can be measured with the aid of a phenomenon known as the Doppler shift, named for an Austrian physicist, Christian Doppler, who discovered it in 1842. The Doppler effect is a shift in the length of waves from a source when the source is moving toward or away from an observer. Although Doppler noted the phenomenon in sound waves, it also applies to light.
If a star (or other luminous object) moves toward Earth, its light waves are compressed and shifted toward the shorter, bluer wavelengths of the spectrum. For a star moving away from Earth, the light waves lengthen and shift to the longer, redder portion of the spectrum. These shifts can be detected with an instrument called a spectrograph, which spreads out a star's light into a spectrum, or rainbow. The spectrum of light from a star is interrupted by dark narrow bands called spectral absorption lines. These lines are formed when atoms and ions (electrically charged atoms) in the star's outer layers of gas absorb energy of particular wavelengths in the light being emitted by the star. Atoms on Earth have the same properties, so we can measure motion by the shift from terrestrial wavelengths. As the star moves toward or away from Earth, the spectral lines are shifted slightly in wavelength toward the blue or red end of the spectrum. The amount of shift in the absorption lines is proportional to the star's speed. The faster the star is moving relative to the Earth, the more the lines are shifted.
If the spectral lines in the light from a star shift regularly between longer and shorter wavelengths, the star must be continually moving away from Earth and then back again. This effect is what astronomers expect to see if a star is orbited by an unseen companion that is pulling it to and fro as the two bodies circle their common center of mass. If the unseen companion object is a massive body such as another star, these wavelength shifts can be quite pronounced. However, if the object causing the star's motion is a planet, then the shifts in the star's spectrum are very small. Normally astronomers analyze the spectra of stars that are moving at speeds of tens of kilometers (1 kilometer is equal to 0.62 mile) per second relative to the Earth. The tug on a star by the gravitational attraction of orbiting planets alters its regular speed by only a few meters per second. Recently, astronomers have developed improved spectrographs that can detect these low speeds.
Doppler Studies Detect Jupiter-sized Planets
Still, there is a limit to the sensitivity of Doppler spectroscopy. The technique is most reliable if a star is being orbited by a planet that is both massive and close to the star. Such planets exert a stronger gravitational pull than smaller or more distant planets, and they have shorter orbital periods. This combination of factors creates higher orbital speeds and larger spectral shifts occurring more frequently, making such planets easier to detect.
These characteristics are just what Swiss astronomers Michel Mayor and Didier Queloz at the Geneva Observatory noted in 1995 when they found the first planet orbiting a normal, sunlike star, 51 Pegasi. Mayor and Queloz had been searching for evidence of brown dwarfs, dim celestial bodies that are more massive than planets but less massive than stars, but instead they found a planet.
From the size of 51 Pegasi's spectral shifts, the astronomers inferred that the planet circling it is as massive as Jupiter. And from the frequency of the shifts in the star's spectrum, they learned that the planet orbits 51 Peg in just 4.23 days, compared with 11.86 years for Jupiter. Based on its mass and orbital period, Mayor and Queloz estimated that the planet lies about 0.05 of an astronomical unit (AU) from its sun. An AU is the distance of the Earth from the sun, about 150 million kilometers (93 million miles). In comparison, Jupiter lies more than 5 AU from our sun and Mercury, 0.39 AU.
Mayor and Queloz's finding was confirmed by Marcy and Butler. Those researchers soon found similar evidence for planets orbiting the stars 70 Virginis and 47 Ursae Majoris. By early 2000, astronomers had found more than 30 extrasolar planets by the Doppler method.
A remarkable discovery was reported in April 1999 by Marcy, But-ler, and their colleague Debra Fischer as well as by another team of astronomers led by Robert Noyes of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Massachusetts. The researchers announced that they had independently discovered three planets circling the star Upsilon Andromedae, located about 44 light-years from Earth. (A light-year is the distance light travels in one year, about 9.5 trillion kilometers [5.9 trillion miles]). This discovery marked the first time that astronomers had found a system of planets around a normal star.
The Transit Technique and Gravitational Microlensing
The transit technique was used to observe a small and repeated dimming in the light from a star named HD 209458 caused when a planet orbiting the star passed in front of, or transited, the star. This discovery was reported in November 1999 by teams of astronomers led by David Charbonneau of the Harvard-Smithsonian Center for Astrophysics and Gregory Henry of Tennessee State University in Nashville. Both groups had been looking for transits of HD 209458 because the Doppler-shift method had already shown the existence of a planet orbiting the star. The confirmation was important because it proved that the transit method could be used to discover extrasolar planets. The effective use of the transit technique was also significant because with this technique, astronomers can learn the size of the planet.
Observations of planetary transits are difficult, however, because the amount of dimming is small—about 1.5 percent of the star's normal brightness in the case of HD 209458. Also, many stars vary in brightness for other reasons. For astronomers to conclude that a star's dimming is caused by a planet, they must observe the star's light growing dimmer and brighter at regular intervals.
The magnifying-glass technique, called gravitational microlensing, enabled researchers to infer the presence of a small planet orbiting a star in the direction of the center of the Milky Way Galaxy. This finding was reported in January 1999 by astronomers David Bennett and Sun Hong Rhie at Notre Dame University in Indiana. Gravitational microlensing is based on the general theory of relativity proposed in 1916 by Einstein. According to Einstein, the gravitational field of a massive object such as a star bends the space around it, much as a bowling ball resting on a rubber mat causes a depression in the mat. Just as a marble rolling across the mat would curve toward the depression in the mat if it passed close to the bowling ball, so a beam of light is bent by the curved space near a star. If one star lies precisely in the line of sight between the Earth and a more distant star, the gravity of the nearer star can act as a celestial magnifying glass, bending and focusing the light from the more distant star. Then the distant star appears brighter to observers on Earth because more of its light reaches Earth.
Opportunities to catch two stars in precise alignment are rare. In addition, the brightening of the distant star is temporary, lasting only as long as it takes for the nearer star to cross in front of the one farther away. Bennett and Rhie surveyed millions of stars throughout the late 1990's before they found an instance of microlensing in which they noted a pattern of brightening that might indicate the presence of a planet. The astronomers used a ground-based telescope to make repeated images of the same patch of sky. Then they used a computer to compare images, looking for any stars that appeared brighter than normal for a period of time. A star that had planets would display a large peak in brightness as well as smaller peaks caused by the additional gravitational lensing effects of the planets.
What Kinds of Planets Are These?
All of the indirect methods for detecting planets also provide some information on the properties of the planets. The Doppler shift method, as noted in the case of 51 Peg, yields rough estimates on planetary mass, very precise measures of orbital period, and the distance of the planet from its star. The transit method allows astronomers to gauge the size of a planet by measuring the fraction of the star's light blocked by the planet. In addition, by observing the duration of the dimming as the planet passes in front of the star, astronomers can calculate the size of the planet's orbit. The gravitational microlensing method provides some information, though not very precise, about the masses of the star and planet. The masses are inferred from the length of time it takes for the background star to brighten and then dim.
What astronomers learned about the newly discovered extrasolar planets is that most of them are very different from the planets in our own solar system. From the amount of wobble in their stars' spectral lines that the planets cause, astronomers inferred that the masses of all of the extrasolar planets found by 2000 are similar to those of gas giants—planets like Jupiter, Saturn, Uranus, and Neptune. Even the two smaller planets found by Marcy and Butler in March 2000 had masses similar to Saturn's. The gas giants in our solar system have a fluid gas-and-liquid structure with no solid surface. Even though astronomers have no information yet on the internal structure of the newly found planets, none of those planets seems similar to the solar system's terrestrial (rocky) planets—Mercury, Venus, Earth, and Mars—which have hard, rocky surfaces and relatively high densities. The only planet that astronomers were able to measure with a greater degree of accuracy—the one orbiting HD 209458—was found to have a diameter 1.4 times that of Jupiter and a mass corresponding to 62 percent of Jupiter's mass. Those figures indicate that the planet's density is lower than that of any of the gas giants in our solar system.
Unusual Orbits
Another way in which extrasolar planets differ from those in our solar system is that most of the extrasolar planets found by 2000 orbit very close to their parent stars. About a half-dozen of them orbit closer to their stars than 1/10 the Earth-sun distance, and nearly all of them are within 2 AU. In part, this finding is a result of the search method by which most extrasolar planets were found—the Doppler method is best suited for locating massive planets that orbit close to their parent stars. However, the discovery of such closely located planets also caused astronomers to question whether our solar system is really the standard by which others should be judged. Astronomers had long believed that gas-giant planets should not be able to form close to their suns. These planets consist of lightweight, easily vaporized gases that, according to current theories of planet formation, should not be able to condense into planets much closer to their stars than Jupiter is to the sun. So why are these gas giants being found where they are?
One theory is that the planets formed far from their stars, but then drag forces caused by interplanetary gas and dust gradually slowed their orbital speeds. As they slowed, the planets spiraled inward to their current locations. Many astronomers consider this explanation plausible because, according to accepted theories of star and planet formation, a lot of gas and dust linger even after planets have formed. Our own solar system—in which Jupiter and the other gas giants have remained in the outer solar system—may be the exception.
Astronomers also wondered, if a planet spirals in toward its sun, what prevents it from going all the way and being destroyed? By 2000 they had not found a satisfactory answer to that question. Some suggested that there may be less interplanetary gas and dust in the inner region of a solar system, so the drag effect ceases at a certain distance from the star. Another possibility is that as a planet gets closer to its star, it gains orbital energy from complex gravitational interactions with the star and settles into a stable orbit.
Another surprising characteristic of the extrasolar planets is their highly elongated orbits. The planets in our solar system have nearly circular orbits. An orbit's shape is measured by its eccentricity, by which a value of 0 describes a circle and a value near 1 corresponds to a very elongated orbit. The most eccentric planetary orbits in our solar system are those of Mercury and Pluto, both about 0.2, while all the others are smaller than 0.1. Some of the extrasolar planets have orbital eccentricities as high as 0.71. In such an orbit the planet is nearly six times farther from its star at its greatest distance than at its closest.
Unlikely Places For Life?
The eccentricity of a planet's orbit is related to the planet's ability to support life. A highly eccentric orbit brings a planet close to its star and then takes it far away from it. Therefore, conditions on the planet's surface vary so much throughout its “year” that life would probably have difficulty forming and surviving. Liquid water, for example, which presumably is essential for the development of life, could not exist much of the time on the surface of such a planet: It would be constantly cycling from frozen to liquid to gas as the planet orbited far away and then close to its sun. In contrast, the Earth, with its nearly circular orbit, has a stable climate. By 2000, astronomers were questioning whether circular orbits are the exception in the universe, not the rule. Nonetheless, most astronomers were not discouraged about finding extraterrestrial life. Their hope has always been based on discovering Earthlike planets, not gas giants, a search that had barely begun.
By early 2000, astronomers around the world were conducting more than 30 searches for extrasolar planets, using ground-based telescopes and indirect methods of detection. They expected to find hundreds of additional gas-giant planets. But in order to discover Earthlike planets, astronomers needed space-based observatories, above the distorting influence of Earth's atmosphere.
Continuing the Search
In 2000, astronomers were planning more than a dozen such missions, centered on advances in three of the basic observational methods. One advance was improved sensitivity to transits. Extreme sensitivity is essential for the transit technique because terrestrial planets are so small in comparison with their star. Thus, the dimming of the star's light when such planets transit it is barely perceptible. Another advance involves greater precision in detecting Doppler shifts in a star's spectrum caused by the gravitational pull of planets. Terrestrial planets have such small masses that they create only minute changes in the motion of their parent stars and thus only tiny shifts in their spectra. And finally, the most intriguing possibility was the development of better methods to observe planets directly.
To improve the transit method, a small NASA spacecraft called Kepler was scheduled to be launched into Earth orbit in 2003. Kepler's mission was to monitor some 100,000 stars from its position above Earth's atmosphere. There, Kepler would avoid the blurring effect that the atmosphere causes in Earth-based telescopes, the interruptions caused by seasonal weather, and the cycle of days and nights to make highly accurate brightness measurements.
Astronomers were also hoping to measure the actual shifts in a star's position instead of just the spectral shifts because of its motion. To do so, they were developing space-based interferometers. An interferometer is an instrument that combines light from two or more telescopes, creating the resolving power of a much larger telescope. With this high resolution, astronomers can measure the positions of stars with unprecedented accuracy. Interferometry has long been widely used with radio telescopes on Earth. However, since stars are not strong radio-wave emitters, the search for tiny stellar wobbles caused by planets required interferometry using visible light, which is much more difficult. NASA's Space Interferometry Mission (SIM), which was to be launched around 2006, and the European Space Agency's Darwin mission, planned for launch after 2008, were to be the first to test this new technology.
In theory, interferometry can also be used to obtain direct images of planets. An instrument can be set up in such a way that unwanted light from a star is cancelled out, or “nulled,” allowing light from a planet to be seen. Astronomers led by Roger Angel at the University of Arizona in Tucson reported in 1998 on an experiment that they had conducted using a nulling interferometer. The astronomers bounced light waves from two mirrors in the Multiple Mirror Telescope with such precise positioning that the peaks from one light wave arrived at a detector at the same time that the troughs of the other light wave arrived, causing the waves to cancel each other out.
Astronomers were also hopeful that computer analysis of light from distant stars could provide a way of detecting planets more directly. In December 1999, astronomer Andrew Cameron and his colleagues at the University of St. Andrews in Scotland isolated a portion of the spectrum from the star Tau Bootis, located about 50 light-years from Earth. They used computer analysis to separate Tau Bootis's light from the light of what they said was a planet orbiting the star. As of early 2000, however, other astronomers had not been able to duplicate their efforts, so the finding remained unconfirmed.
As the year 2000 began, many astronomers were hopeful that within a decade, Earthlike planets would be found oribiting other suns. The new technologies for planet hunting may not tell astronomers whether the human race is alone in the universe, but astronomers hoped at least to answer the question of whether Earth is unique.