When astronomers turn their telescopes to young nearby stars, they see disks of the same type from which the Solar System formed. When the light from the central star is blocked (Figure 7.15), evidence of the planetary disk is observed. The physical processes that led to the formation of the Solar System should be commonplace wherever new stars are being born. Compared with stars, however, planets are small and dim objects. They shine primarily by reflection and therefore are millions to billions of times fainter than their host stars. Thus, they were difficult to find until advances in telescope detector technology in the 1990s enabled astronomers to discover them through indirect methods. In 1995, astronomers announced the first confirmed exoplanet—a planet orbiting around a star other than the Sun. (These are sometimes called “extrasolar planets,” although “exoplanet” is more accepted today.) Today, the number of known exoplanets has grown to the thousands, and new discoveries occur almost daily.
Figure 7.15 An edge-on dust disk around a star is seen extending outward to 60 AU from the young (12-million-year-old) star AU Microscopii. The star itself, whose brilliance would otherwise overpower the dust disk, is hidden behind an opaque mask placed in the telescope’s focal plane. The star’s position is represented by the dot.
The International Astronomical Union (IAU) currently defines an exoplanet as an object that orbits a star other than the Sun and has a mass less than 10–13 Jupiter masses (10–13 MJup). Objects more massive than 10–13 MJup but less massive than 0.08 solar masses (0.08 MSun; about 80 MJup) are brown dwarfs. Objects more massive than 0.08 MSun are defined as stars. Figure 7.16 compares the diameters of typical objects of each class.
Figure 7.16 A comparison of the diameters of the Sun, a low-mass star, a brown dwarf, Jupiter, and Earth.
The Search for Exoplanets
The first planets were discovered indirectly, by observing their gravitational tug on the central star. As technology has improved, other methods have become more productive. Astronomers now have direct images of planets orbiting stars and have taken spectra of some exoplanets to observe the composition of their atmospheres. Almost certainly, between the time we write this and the time you read it, new discoveries will have been made. The field is advancing extremely quickly: more than 100 projects are searching for exoplanets from the ground and from space. We will now look at each discovery method.
The Radial Velocity Method
As a planet orbits a star, the planet’s gravity tugs the star around ever so slightly. If the star has a radial velocity (Chapter 5), an observable Doppler shift may appear in the spectrum of the star. Figure 7.17 illustrates that motion. When the star is moving toward us (negative radial velocity), the light is blueshifted; when the star is moving away from us (positive radial velocity), the light is redshifted. That pattern of radial velocity repeats. After detecting those changes in the radial velocity (Figure 7.18), astronomers can infer the existence of the planet and find the planet’s mass and distance from the star.
Figure 7.17 Doppler shifts observed in the spectrum of a star as it moves around its common center of gravity with its planet. The spectral lines are blueshifted as the star moves toward us and redshifted as it moves away from us.Figure 7.18 Radial velocity data for a star with a planet. A positive velocity is motion away from the observer, whereas a negative velocity is motion toward the observer. The plot repeats as the planet completes another orbit around the star.
The smallest planet that can be found depends on the precision of the observations. For example, in our Solar System, Jupiter’s mass is greater than the mass of all the other planets, asteroids, and comets combined, so Jupiter is the planet with the largest effect on the Sun. Both the Sun and Jupiter orbit a common center of gravity (sometimes called center of mass—the location where the effect of one mass balances the other) that lies just outside the surface of the Sun, as shown in Figure 7.19. Jupiter tugs the Sun around in a circle with a speed of 12 m/s. Alien astronomers would find that the Sun’s radial velocity varies by ±12 m/s, with a period equal to Jupiter’s orbital period of 11.86 years. From that information, the astronomers would rightly conclude that the Sun has at least one planet with a mass comparable to Jupiter’s. Without greater precision, the observers would be unaware of the other, less massive planets. To detect Saturn, they would need to improve the precision of their measurements to 2.7 m/s. Earth would not be detectable unless the aliens could detect motions as small as 0.09 m/s.
Figure 7.19 The Sun and Jupiter orbit around a common center of gravity (+), which lies just outside the Sun’s surface. Spectroscopic measurements made by an extrasolar astronomer would reveal the Sun’s radial velocity varying by ±12 m/s over 11.86 years, which is Jupiter’s orbital period. Jupiter travels around its orbit at a speed of 13,000 m/s.
The precision of radial velocity instruments has been about 0.3 m/s. The technique enabled astronomers to detect giant planets around solar-type stars, but not yet to find planets with masses similar to Earth’s. Finding the signal of the Doppler shift in the noise of the observation requires the star to be quite bright in our sky. A new instrument at the Very Large Telescope is expected to improve precision and be able to detect smaller planets. Working It Out 7.2 explains more about the spectroscopic radial velocity method.
working it out 7.2
Estimating the Size of a Planet’s Orbit
In the spectroscopic radial velocity method, the star is moving about its center of mass, and its spectral lines are Doppler-shifted accordingly. Recall from Figure 7.19 that an alien astronomer looking toward the Solar System would observe a shift in the wavelengths of the Sun’s spectral lines—caused by the presence of Jupiter—of ∼12 m/s.
Figure 7.18 showed the radial velocity data for a star with a planet discovered with that method. How do astronomers use that method to estimate the distance (A) of the planet from the star? Recall from Chapter 4 that Newton generalized Kepler’s law relating the period of an object’s orbit to the orbital semimajor axis:
where A is the semimajor axis of the orbit, P is its period, and M is the combined mass of the two objects. To find A, we rearrange the equation as follows:
According to the graph of radial velocity observations in Figure 7.18, the period of the orbit is 5.7 years. A year has 3.16 × 107 seconds, so
P= 5.7 yr × (3.16 × 107 s/yr) = 1.8 × 108 s
The mass of the star is much greater than the mass of the planet, so the combined masses of the star and the planet can be approximated as the mass of the star, which here is about equal to the mass of the Sun, 2 × 1030 kg. (Stellar masses can be estimated from their spectra.) The gravitational constant G is 6.67 × 10–20 km3/(kg s2). Plugging in the numbers gives
Taking the cube root,
A= 4.8 × 108 km
If we convert that number into astronomical units (where 1 AU = 1.5 × 108 km), the semimajor axis of the orbit of this planet is
The planet is 3.2 times farther from its star than Earth is from the Sun.
The Transit Method
From Earth it is sometimes possible to see the inner planets Mercury and Venus transit, or pass in front of, the Sun. An alien located somewhere in the plane of Earth’s orbit would see Earth pass in front of the Sun and could infer the existence of Earth by detecting the 0.009 percent drop in the Sun’s brightness during the transit. Similarly, for astronomers on Earth to observe a planet passing in front of a star, Earth must lie nearly in the orbital plane of that planet. When an exoplanet passes in front of its parent star, the light from the star diminishes by a tiny amount (Figure 7.20). This is the transit method of detecting exoplanets, in which we observe the dimming of starlight as a planet passes in front of its parent star. Whereas the radial velocity method gives us the mass of the planet and its orbital distance from a star, the transit method provides the radius of a planet. Working It Out 7.3 shows how the radii are estimated.
Figure 7.20 A planet that passes in front of a star blocks some of the light coming from the star’s surface, causing the brightness of the star to decrease slightly. (The decrease in brightness is exaggerated here.)
working it out 7.3
Estimating the Radius of an Exoplanet
The masses of exoplanets can often be estimated using Kepler’s laws and the conservation of angular momentum. When planets are detected with the transit method, astronomers can estimate the radius of an exoplanet. In that method, astronomers look for planets that eclipse their stars and then observe how much the star’s light decreases during that eclipse (see Figure 7.20). When Venus or Mercury transits the Sun, a black circular disk is visible on the face of the circular Sun. During the transit, the amount of light from the transited star is reduced by the area of the circular disk of the planet divided by the area of the circular disk of the star:
Then, to solve for the radius of the planet, astronomers need an estimate of the radius of the star and a measurement of the fractional reduction in light during the transit. The radius of a star is estimated from the surface temperature and the luminosity of the star.
Kepler-11, for example, is a system of at least six planets that transit a star. The radius of the star, Rstar, is estimated to be 1.1 times the radius of the Sun, or 1.1 × (7.0 × 105 km) = 7.7 × 105 km. The light from the Kepler-11 star is observed to decrease by 0.077 percent, or 0.00077 (see Figure 7.20), from planet Kepler-11c. What is the radius of Kepler-11c?
Dividing RKepler-11c by REarth (6,400 km) shows that RKepler-11c= 3.3 REarth.
Current ground-based technology limits the sensitivity of the transit method to about 0.1 percent of a star’s brightness. Telescopes in space improve the sensitivity because smaller dips in brightness can be measured. The small French COROT telescope (27 cm) discovered 32 planets during its 6 years of operation (2007–2013). NASA’s 0.95-meter Kepler telescope has discovered many planets and has found thousands more candidates that are being investigated further. Figure 7.21 shows that if one planet is found with this method, multiple sets of transits can indicate that other planets are orbiting the same star. Several thousand exoplanets have been detected from ground-based and space telescopes by using the transit method.
Figure 7.21 Multiple planets can be detected through multiple transits with different changes in brightness. The arrows point to the changes in the total light as the three planets transit the star.
Other Methods
The gravitational field of an unseen planet can act like a lens, bending the light from a distant star in such a way that it causes the star to brighten temporarily while the planet is passing in front of it. When light is bent by a gravitational field, the effect is called gravitational lensing. In this instance, because the effect is small, it is usually called microlensing. Like the radial velocity method, microlensing provides an estimate of the mass of the planet. To date, about 60 exoplanets have been found with this technique.
Planets also may be detected by astrometry—precisely measuring the position of a star in the sky. If the system is viewed from “above,” the star moves in a mini-orbit as the planet pulls it around. That motion is generally tiny and therefore very difficult to measure. For systems viewed from above the plane of the planet’s orbit, however, none of the prior methods will work because the planet neither passes in front of the star nor causes a shift in its speed along the line of sight. Space missions such as the Gaia observatory, launched in 2013 by the European Space Agency, conduct such observations.
Direct imaging involves taking a picture of the planet directly. The technique is conceptually straightforward but is technically difficult because it involves searching for a relatively faint planet in the overpowering glare of a bright star—a challenge far more difficult than looking for a star in a clear, bright daytime sky. Even when an object is detected by direct imaging, an astronomer must still determine whether the observed object is actually a planet. Suppose we detect a faint object near a bright star. Could it be a more distant star that just happens to be in the line of sight? Future observations could tell whether the object shares the bright star’s motion through space, but it also could be a brown dwarf rather than a true planet. An astronomer would need to make further observations to determine the object’s mass.
Some planets have been discovered through that method with large ground-based telescopes operating in the infrared region of the spectrum, using adaptive optics. Figure 7.22 is an infrared image of Beta Pictoris b. Hubble Space Telescope observations of Fomalhaut, a bright naked-eye star only 25 light-years away, revealed a 3 Jupiter–mass planet in the dusty debris ring about 17 billion km from the central star (Figure 7.23). A related form of direct observation involves separating the spectrum of a planet from the spectrum of its star to obtain information about the planet directly. Large ground-based telescopes have obtained spectra of the atmospheres of some exoplanets and have found, for example, carbon monoxide and water in those atmospheres.
Figure 7.22 The planet Beta Pictoris b is seen orbiting within a dusty debris disk that surrounds the bright naked-eye star Beta Pictoris. The planet’s estimated mass is 8 times that of Jupiter. The star is hidden behind an opaque mask, and the planet appears through a semitransparent mask used to subdue the brightness of the dusty disk.
Credit: ESO/A.-M. Lagrange et al., https://www.eso.org/public/images/eso0842b/. https://creativecommons.org/licenses/by/4.0/.
Figure 7.23 A Hubble Space Telescope image of Fomalhaut b, seen here in its 2012 position in the orbit. Prior positions are indicated by the tic marks on the white line overlaid on the orbit. The parent star, hidden by an obscuring mask, is about a billion times brighter than the planet.
Types of Exoplanets
what if . . .
What if astronomers had not yet discovered any planets around the 5,000 stars that have been observed so far? Would we be able to conclude (with current technology) that there are no planets around the 100 billion stars in the Milky Way? Why or why not?
Searches for exoplanets have been remarkably successful. Between the discovery of the first (in 1995) and this writing, more than 4000 more have been confirmed, and thousands more candidates are under investigation. As the number of observed systems with single and multiple planets increases, astronomers can compare those worlds with Solar System planets, finding more variation than they expected. The field is changing so fast that the most up-to-date information can be found only online.
The first discoveries included many hot Jupiters—Jupiter-sized planets orbiting solar-type stars in circular or highly eccentric orbits that bring them closer to their parent stars than Mercury is to our own Sun. Those planets were among the first to be detected because they are relatively easy targets for the spectroscopic radial velocity method. The large mass of a nearby hot Jupiter tugs the star very hard, creating large radial velocity variations in the star. In addition, those large planets orbiting close to their parent stars are more likely to pass in front of the star periodically and reveal themselves via the transit method. Therefore, those hot Jupiter systems are not necessarily representative of most planetary systems—they are just easier to find than smaller, more distant planets. Scientists call that type of bias a selection effect. Figure 7.24 illustrates the types of exoplanet populations.
Figure 7.24 Different populations of exoplanets. The color of the points indicates the detection method. Jupiter-sized planets can be close to their star and hot, or they can be far from their star and cold. Neptune-sized planets are probably composed of water and ice. Earth-sized planets have higher density, indicating that they are made of rock.
Astronomers were surprised by the hot Jupiters because, according to the theory of planet formation described earlier in the chapter, those giant, volatile-rich planets should not have been able to form so close to their parent stars. From theories based on the Solar System, astronomers expected that Jupiter-type planets should form in the more distant, cooler regions of the protoplanetary disk, where the volatiles that make up much of their composition can survive. Hot Jupiters may form much farther from their parent stars and later migrate inward to a closer orbit. That migration may be caused by an interaction with gas or planetesimals in which orbital angular momentum is transferred from the planet to its surroundings, allowing it to spiral inward.
The radial velocity method yields an estimate of the mass of a planet (see Working It Out 7.2), and the transit method yields an estimate of the size of a planet (see Working It Out 7.3). With both pieces of information, the density (mass divided by the volume) of the planet can be computed to determine whether the planet is mostly gaseous (low density) or mostly rock (high density). Many of the new planets discovered by Kepler are mini-Neptunes (gaseous planets with masses of 2–10 MEarth) or super-Earths (rocky planets more massive than Earth). Figure 7.25 illustrates the formation of those two types.
Figure 7.25 Many exoplanets are between the sizes of Earth and Neptune (∼4RE). “Super-Earths” began with smaller rocky cores and less material, whereas “mini-Neptunes” had larger cores and collected more gas while forming.
Planets with longer orbital periods, and therefore larger orbits, can be discovered only when the observations have gone on long enough to observe more than one complete orbit. Some of the exoplanets have highly elliptical orbits compared with those in the Solar System. Planets have been found with orbits that are highly tilted with respect to the plane of the rotation of their star, and some planets move in orbits whose direction is opposite that of their star’s rotation. Multiple-planet systems have been observed in which the larger mini-Neptunes alternate with smaller super-Earths. The multiple-planet systems that have been found with the transit method reside in flat systems like our own, offering further evidence that the planets formed in a flat protoplanetary disk around a young star. But the current hypothesis to explain the Solar System’s inner, small rocky planets and outer, large gaseous planets may not apply to all other planetary systems.
In addition, some planets, discovered through microlensing, don’t have a star at all. Those planets may have been ejected from their solar systems after they formed and are no longer in gravitationally bound orbits around their stars. Others are “circumbinary”; they orbit two stars that form a binary star system. A scroll through NASA’s exoplanet archive reveals that new planet discoveries are confirmed nearly every week. The frequent new discoveries requiring revisions of existing theories make exoplanets one of the most exciting topics in astronomy today.
We return to exoplanets at several points in this book—namely, when we review the planets in our own Solar System, when we consider the types of stars, and when we discuss the search for Earth-like planets in Chapter 24.
unanswered questions
How Earth-like must a planet be before scientists declare it to be “another Earth”? An editorial in the science journal Nature cautioned that scientists should define “Earth-like” in advance—before multiple discoveries of planets “similar” to Earth are announced and a media frenzy ensues. Must a planet be of similar size and mass, be located in the habitable zone, and have spectroscopic evidence of liquid water before we call it “Earth 2.0”?
CHECK YOUR UNDERSTANDING 7.5
What is the most common method for discovering an Earth-mass planet around a star? (a) Doppler spectroscopy; (b) direct imaging; (c) transit; (d) astrometric
A “failed star” without enough mass to fuse hydrogen in its core. An object whose mass is intermediate between that of the least massive stars and that of supermassive planets.
A method of detecting extrasolar planets by measuring the decrease in light from a star as its orbiting planet passes in front of the star as viewed from Earth.
what if . . .
Answer