Origins: Fourteen Billion Years of Cosmic Evolution (17 page)

Newton’s law of gravity states that, although the gravity from a planet gets progressively weaker as you travel farther from it, no distance will reduce the force of gravity all the way to zero, and that an object with enormous mass can exert significant gravitational forces even at large distances. The planet Jupiter, with its mighty gravitational field, bats out of harm’s way many comets that would otherwise wreak havoc on the inner solar system. By doing so, Jupiter acts as a gravitational shield for Earth, allowing long (50- to 100-million-year) stretches of relative peace and quiet on Earth. Without Jupiter’s protection, complex life would have a hard time growing interestingly complex, always living at the risk of extinction from a devastating impact.

We have exploited the gravitational fields of planets for nearly every probe sent into space. The
Cassini
probe, for example, sent to Saturn for an encounter late in 2004, was launched from Earth on October 15, 1997, and was gravitationally assisted twice by Venus, once by Earth (on a return flyby), and once by Jupiter. Like a multi-cushion billiard shot, trajectories from one planet to another using gravitational slingshots are common. Otherwise our tiny probes would not have enough speed and energy to reach their destinations.

One of us is now accountable for a piece of the solar system’s interplanetary debris. In November 2000, the main-belt asteroid 1994KA, discovered by David Levy and Caroline Shoemaker, was named “13123 Tyson.” A fun distinction, but there’s no particular reason to get big-headed about it; as already noted, plenty of asteroids have familiar names such as Jody, Harriet, and Thomas. And plenty of other asteroids have names such as Merlin, James Bond, and Santa. Rising through 20,000, the count of asteroids with well-established orbits (the criterion for assigning them names and numbers) may soon challenge our capacity to name them. Whether or not that day arrives, there is curious comfort knowing that one’s own chunk of cosmic debris is not alone, as it litters the space between the planets, joined by a long list of other chunks named for real and fictional people.

When last checked, asteroid 13123 Tyson was not headed toward us, and so cannot be blamed for either ending or starting life on Earth.

CHAPTER 13

Worlds Unnumbered

Planets Beyond the Solar System

         Thro’ worlds unnumbered tho’ the God be known,

         ’Tis ours to trace him only in our own.

         He, who through vast immensity can pierce,

         See worlds on worlds compose one universe,

         Observe how system into system runs

         What other planets circle other suns,

         What varied Being peoples ev’ry star,

         May tell why Heav’n has made us as we are.

—Alexander Pope,
An Essay on Man
(1733)

N
early five centuries ago, Nicolaus Copernicus resurrected a hypothesis that the ancient Greek astronomer Aristarchus had first suggested. Far from occupying the center of the cosmos, said Copernicus, Earth belongs to the family of planets that orbit the Sun.

Even though a majority of humans have yet to accept this fact, believing in their hearts that Earth remains immobile as the heavens turn around her, astronomers have long offered convincing arguments that Copernicus wrote the truth about the nature of our cosmic home. The conclusion that Earth ranks as just one of the Sun’s planets immediately suggests that other planets fundamentally resemble our own, and that they may well possess their own inhabitants, endowed as we are with plans and dreams, work, play, and fantasy.

For many centuries, astronomers who used telescopes to observe hundreds of thousands of individual stars lacked the ability to discern whether or not any of these stars have planets of their own. Their observations did reveal that our Sun ranks as an entirely representative star, whose near twins exist in great numbers throughout our Milky Way galaxy. If the Sun has a planetary family, so too might other stars, with their planets equally capable of giving life to creatures of all possible forms. Expressing this view in a manner that affronted papal authority brought Giordano Bruno to his death at the stake in 1600. Today, a tourist can pick his way through the crowds at the outdoor cafés in Rome’s Campo di Fiori to reach Bruno’s statue at its center, then pause for a moment to reflect on the power of ideas (if not the power of those who hold them) to triumph over those who would suppress them.

As Bruno’s fate helps to illustrate, imagining life on other worlds ranks among the most powerful ideas ever to enter human minds. Were this not so, Bruno would have lived to a riper age, and NASA would find itself shorter of funds. Thus speculation about life on other worlds has focused throughout history, as NASA’s attention still does, on the planets that orbit the Sun. In our search for life beyond Earth, however, a great frost has appeared: none of the other worlds in our solar system seem particularly fit for life.

Although this conclusion hardly does justice to the myriad possible paths by which life might arise and maintain itself, the fact remains that our initial explorations of Mars and Venus, as well as of Jupiter and its large moons, have failed to produce any convincing signs of life. To the contrary, we have found a great deal of evidence for conditions extremely hostile to life as we know it. Much more searching remains to be done, and fortunately (for those who engage themselves mentally in this effort) continues to be underway, especially in the hunt for life on Mars. Nevertheless, the verdict on extraterrestrial life in the solar system shows enough likelihood of proving negative that supple minds now usually look beyond our cosmic neighborhood, to the vast array of possible worlds that orbit stars other than our Sun.

Until 1995, speculation
about planets around other stars could proceed almost entirely unfettered by facts. With the exception of a few pieces of Earth-sized debris in orbit around the remnants of exploded stars, which almost certainly formed after the supernova explosion and barely qualified as planets, astrophysicists had never found a single “exosolar planet,” a world orbiting a star other than the Sun. At the end of that year came the dramatic announcement of the first such discovery; then, a few months later, came four more; and then, with the floodgates open, finding new worlds proceeded ever more swiftly. Today, we know of far more exosolar planets around other stars than of the now familiar worlds that orbit the Sun—a tally that exceeds 100 and is almost certain to keep growing for years to come.

To describe these newfound worlds, and analyze the implications of their existence in the search for extraterrestrial life, we must confront a single hard-to-believe fact: Although astrophysicists assert that they not only know that these planets exist but have also deduced their masses, their distances from their parent stars, the times that the planets take to complete their orbits, and even the shapes of those orbits, no one has ever seen or photographed a single one of these exosolar planets.

How can anybody deduce so much about planets they have never seen? The answer lies in detective work familiar to those who study starlight. By separating that light into its spectrum of colors, and by comparing those spectra among thousands of stars, those who specialize in observing starlight can recognize different types of stars purely by the ratios of the intensities of the different colors that appear in stellar spectra. Once upon a time, these astrophysicists photographed the stars’ spectra, but today they use sensitive devices that register digitally how much starlight of each particular color reaches us on Earth. Though the stars are many trillions of miles from us, their fundamental natures have become an open book. Astrophysicists can now easily determine—purely by measuring the spectrum of the colors of starlight—which stars most closely resemble the Sun, which are somewhat hotter and more luminous, and which are cooler and intrinsically fainter than our star.

But they can also do more. Having grown familiar with the distribution of colors in the spectra of various types of stars, astrophysicists can quickly identify a familiar pattern in the star’s spectrum, which typically shows the partial or total absence of light at particular colors. They often recognize such a pattern, but find that all the colors that form it have been slightly shifted toward either the red or the violet end of the spectrum, so that all the familiar guideposts are now either somewhat redder or somewhat more violet than the norm.

Scientists characterize these colors by their wavelengths, which measure the separation between successive wave crests in the vibrating light waves. Because they correspond with the colors that our eyes and brains perceive, specifying exact wavelengths simply names colors more precisely than we do in normal speech. When astrophysicists spot a familiar pattern in the intensity of light measured for thousands of different colors, but find that all the wavelengths in the pattern are (for example) 1 percent longer than usual, they conclude that the star’s colors have changed as the result of the Doppler effect, which describes what happens when we observe an object either approaching us or receding from us. If, for example, an object moves toward us, or we move toward it, we find that all the wavelengths of the light that we detect are
shorter
than those we measure from an identical object at rest with respect to ourselves. If the object recedes from us, or we recede from it, we find all the wavelengths to be
longer
than those from an object at rest. The deviation from the at-rest situation depends on the relative velocity between the light source and those who observe it. For speeds much less than the speed of light (186,000 miles per second), the fractional change in all the wavelengths of light, called the Doppler shift, equals the ratio of the speed of approach or recession to the speed of light.

During the 1990s, two teams of astronomers, one in the United States and one centered in Switzerland, devoted themselves to increasing the precision with which they could measure the Doppler shifts of starlight. They did so not simply because scientists always prefer to make more accurate measurements, but because they had a straightforward goal: to detect the existence of
planets
by studying the light from
stars
.

Why this roundabout approach to the detection of exosolar planets? Because for now this method offers the only effective way to discover them. If our solar system offers any guide to the distances at which planets orbit stars, we must conclude that these distances amount to only a tiny fraction of the distances between stars. The Sun’s closest neighbor stars are about half a million times farther from us than the distance between the Sun and its innermost planet, Mercury. Even Pluto’s distance from the Sun is less than one five-thousandth of the distance to Alpha Centauri, our closest star system. These astronomically minuscule separations between the stars and their planets, combined with the faintness with which a planet reflects light from its star, make it nearly impossible for us to actually see any planets beyond the solar system. Imagine, for example, an astrophysicist on a planet around one of the Alpha Centauri stars who turns her telescope toward the Sun and attempts to spot Jupiter, the Sun’s largest planet. The Sun-Jupiter distance amounts to only one fifty-thousandth of the distance to the Sun, and Jupiter shines with just one billionth of the Sun’s intensity. Astrophysicists like to compare this to the problem of seeing a firefly next to a searchlight’s glare. We may do it some day, but for now the quest to observe exosolar planets directly lies beyond our technological capabilities.

The Doppler effect offers another approach. If we study the star closely, we can carefully measure any changes that appear in the Doppler shift of the light from that star. These changes must arise from changes in the speed with which the star is either approaching us or receding from us. If the changes prove to be cyclical—that is, if their amounts rise to a maximum, fall to a minimum, rise to the same maximum again, and repeat this cycle over the same intervals of time—then the entirely reasonable conclusion follows that the star must be moving in an orbit that takes it around and around some point in space.

What could make a star dance like that? Only the gravitational force from another object, so far as we know. No doubt that planets, by definition, have masses much less than the mass of a star, so they exert only modest amounts of gravitational force. When they pull on a nearby star that possesses far more mass than they do, they produce only small changes in the star’s velocity. Jupiter, for example, changes the Sun’s velocity by about 40 feet per second, slightly more than the speed of a world-class sprinter. As Jupiter performs its twelve-year orbit around the Sun, an observer located along the plane of this orbit would measure Doppler shifts in the Sun’s light. These Doppler shifts would demonstrate that at a particular time, the Sun’s velocity with respect to the observer would rise 40 feet per second above its average value. Six years later, the same observer would find that the Sun’s velocity is 40 feet per second less than average. During the interim, this relative velocity would shift smoothly between its two extreme values. After a few decades of observing this repetitive cycle, the observer would justifiably conclude that the Sun has a planet moving in a twelve-year orbit that causes the Sun to perform its own orbit, producing the velocity changes that arise naturally from this motion. The size of the Sun’s orbit, in comparison to the size of Jupiter’s, exactly equals the
inverse
of the ratio of the two objects’ masses. Since the Sun has one thousand times Jupiter’s mass, Jupiter’s orbit around their mutual center of gravity is one thousand times
larger
than the Sun’s—testimony to the fact that the Sun is a thousand times more difficult to budge than Jupiter.

Of course, the Sun has many planets, each of which simultaneously pulls on the Sun with its own gravitational force. The Sun’s net motion therefore amounts to a superposition of orbital dances, each with a different cyclical period of repetition. Because Jupiter, the Sun’s largest and most massive planet, exerts the greatest amount of gravitational force on the Sun, the dance imposed by Jupiter dominates this complex pattern.

When astrophysicists sought to detect exosolar planets by watching stars dance, they knew that to find a planet roughly similar to Jupiter, orbiting its star at a distance comparable to Jupiter’s distance from the Sun, they would have to measure Doppler shifts with an accuracy sufficient to reveal velocity changes of approximately 40 feet per second. On Earth this sounds like a significant speed (about 27 miles per hour), but in astronomical terms, we are talking about less than one millionth of the speed of light, and about one thousandth of the typical speed with which stars happen to be moving toward us or away from us. Thus to detect the Doppler shift produced by a change in velocity equal to one millionth of the speed of light, astrophysicists must measure changes in wavelength—that is, in star colors—of one part in a million.

These precision measurements
yielded more than the detection of planets. First of all, because the detection scheme lies in finding a cyclical repetition in the changes of a star’s velocity, the length of each of these cycles directly measures the orbital period of the planet responsible for it. If the star dances with a particular cycle of repetition, the planet likewise must be dancing with an identical period of motion, though in a much larger orbit. This orbital period in turn reveals the distance of the planet from the star. Isaac Newton long ago proved that an object orbiting a star will complete each orbit more rapidly when closer to the star, more slowly when farther away: each orbital period corresponds to a particular value of the average distance between the star and the orbiting object. In the solar system, for instance, a one-year orbital period implies a distance equal to the Earth-Sun distance, whereas a twelve-year period implies a distance 5.2 times larger, the size of Jupiter’s orbit. So the research team could announce not only that they had found a planet but also that they knew both the planet’s orbital period and its average distance from its star.