The problem of identifying planets around other stars is one of enormous challenge. Jupiter’s brightness, at optical wavelengths to which our eyes are sensitive, is one-billionth (10−9) that of the Sun. At infrared wavelengths, where cooler objects tend to radiate (as described in Chapter 3), Jupiter is still a mere 10−5 as bright as the Sun. Imagine looking at our solar system from a great distance, many light-years away. Any normal telescope system will see the Sun, the central star, but even Jupiter is lost in the glare caused by light scattered across imperfections in the telescope mirrors and across the telescope structure.
To overcome these problems, astronomers pursue several types of approaches to detect planets around other stars.
Indirect techniquesrely on observing the effect of the planet on the motion or appearance of the parent star.Adaptive opticsand interferometric imagingare bothdirecttechniques to overcome telescope glare and the smearing effects of our own atmosphere (through which Earth-bound astronomers must look) to achieve images of planets orbiting another star.
10.4.1 Indirect techniques
Indirect techniques illustrated in Figure 10.10 involve watching the position of the star oscillate in the sky, caused by the gravi- tational effect of a planetary companion. Motion back and forth can be seen using precision position measurement determina- tions referred to asastrometry. Star wobble toward or away from the observer can be identified by looking at one or more spectral lines from the star that reveal theradial velocity through the Doppler effect described in Chapter 2. As the star moves toward the observer, the line isblue shifted; as the star moves away, the line isred shifted. The mass of the planetary companion can be determined from the magnitude of these effects.
Another indirect technique works if the orbit of the planet around the star is close to the line of sight to Earth. Then, as the planet passes in front of (“transits”) its parent star, it blocks some of the light and causes a partial eclipsing of the starlight. The technique has been used from the ground, but the low probability that any given planetary system will be suitably aligned for transits makes a space-based system capable of viewing a large part of the sky attractive. Two such missions, Corot from the French and European Space Agencies, and Kepler from the US Space Agency NASA, are currently accumulating statistics from transits of the occurrence of Earth- and near-Earth-sized planets
collide with planets incorporated into
planets or asteroids
short-period comets
Sun
collide/eject
encounters with Neptune
ejected outward
perturbed inward perturbed inward
perturbed outward
Oort Cloud
solar nebula
planetesimals interstellar
medium
Kuiper Belt objects
long-period comets
collisions produce meteorites
Figure 10.8Possible history of small bodies in the solar system, from the original molecular cloud of the interstellar medium in which the solar system was born, through the solar nebula phase to the accumulation of planetesimals into larger bodies. Some planetesimals became part of the growing planets and asteroids; later collisions among asteroids produced fragments, some of which reach Earth as meteorites. As giant planets formed and their gravity increased, orbits of remnant planetesimals were increasingly perturbed; some planetesimals were ejected to the Oort Cloud, others inward to collide with the terrestrial planets. The Oort Cloud became the source of the long-period comets. Remnant planetesimals just beyond Neptune constitute the Kuiper Belt, and some of these have survived to the present day. Others, perturbed principally by Neptune’s gravity, were either ejected outward or shunted inward to form Centaur objects. These either collide with the giant planets or have their orbits further altered to become short-period comets. Based on a scheme by Cruikshank (1997).
around other stars. It appears that such planets are common – perhaps 10% of stars like the Sun have such planets.
Microlensing detects planets through an entirely different effect. As seen from Earth, if a star passes in front of a more distant, background star, the light from the background star is temporarily enhanced by the bending of light rays around the nearer star, in accordance with the general theory of relativ- ity, which predicts that gravitational fields bend light rays. If a planet is present in the right position around the nearer star, it produces a further brightening, which is distinguishable from that of its parent star. Though such microlensing events are rare, a modest-size telescope in space can automatically scan many hundreds of thousands of stars to catch those rare signatures of the focusing of light by a passing interloper and its planet.
Indirect techniques have yielded a plethora of important results. Since 1995, over 500 planets, from the mass of Jupiter down to just a few times the mass of the Earth, have been discov- ered around other stars by these techniques. Transiting planets
with radii down to that of Neptune have been found by the CorotandKeplerspacecraft (Figure 10.11). For those planets seen both in transit and via the Doppler shift, both radius and mass are known so that we also know their density. It has been possible to infer that giant planets like Jupiter do exist around other stars, but they have a variety of different properties includ- ing composition, weather, etc. The atmospheric properties of some of these bodies have been discovered by spectra taken during transits of the planets behind or in front of their parent stars. And around the smallest “M-dwarf” stars, where detection is easier, planets only a few times the mass of the Earth have been seen.
In 1992, radial velocity techniques were employed in a very different fashion to detect planets around a pulsar. A pulsar is the ultradense neutron star core of an exploded star, one that has finished the chain of fusion reactions described in Chapter 3.
Most such neutron stars have very strong magnetic fields, which result in charged particles streaming along the magnetic poles
Figure 10.9Photograph, using an electron microscope, of an interplanetary dust particle, roughly 10 microns across. The dark holes in the background (used to help mount the particle) are 0.4 microns across. Image courtesy of Professor Don Brownlee, Washington University.
of the star, creating a beacon that can be detected at radio wavelengths. Using the Arecibo radio telescope in Puerto Rico to measure the Doppler shift to progressively shorter (bluer) and then longer (redder) wavelengths of radio energy, National Radio Astronomy Observatories (NRAO) astronomer D. Frail and team were able to infer the presence of at least two and, from 1994 observations, possibly as many as four, planets orbiting the pulsar PSR1257+12. These planets range in mass from several times that of Earth to a mass as small as that of Earth’s Moon.
How planets could have survived the pulsar-creating explo- sion of the original star is a mystery. One idea holds that
observed shift in position of star against the background of more distant stars
barycentric orbit of star due to planet Earth (observer)
Doppler shift of star’s light toward the red Doppler shift of
star’s light toward the blue
barycentric orbit of star due to planet
unseen planet
astrometry
radial velocity
Figure 10.10Examples of indirect techniques for detecting planets.
Shown in both sketches is a star and its companion planet. The planet forces the star to be nonstationary, that is to orbit the common center of mass of the system. The observer on Earth is symbolized by the telescope, which is extremely far from the star and its companion planet. In the radial velocity technique, the distant observer on Earth sees the component of the star’s motion (actually its velocity,V) directly toward or away from the Earth via Doppler shift. In the astrometric method, the star’s slight shift side to side in the sky, due to its orbital motion, is detected on Earth. In both cases, the planet itself is lost in the glare of the parent star, and is detected only by its gravitational influence on the star. Adapted from a drawing by NASA.
Figure 10.11Planets found in the early part of the Kepler space mission using the transit technique. Sizes of the planets are shown compared to Jupiter and the Earth; REis the radius of the Earth. From a color figure courtesy NASA-Ames Research Center, Wendy Stenzel.
Figure 10.12A disk of dust is seen around the star Fomalhaut at optical wavelengths, using the coronagraph onboard the Hubble Space Telescope to dim the light of the parent star. The inset is a composite image showing the location of a planet orbiting the star, seen in 2004 and 2006 relative to Fomalhaut. By looking at two succeeding years the motion of the planet can be detected, and indeed it seems to be moving in an orbit nested within the dust belt. From Kalaset al.(2008). See color version in plates section.
the planets did not exist prior to the supernova explosion but were instead created from debris of the explosion in a pro- cess mimicking planet formation around very young stars. The presence of these planets suggests that such formation pro- cesses can occur in many different kinds of environments around stars.
10.4.2 Direct techniques
There are two distinct approaches to suppressing the light from a star to a sufficient extent to see a planet orbiting around it: coro- nagraphy and interferometry. An internal coronagraph blocks the starlight using optical elements within a telescope, while an external-occulter coronagraph blocks the starlight with a separate large starshade positioned in front of the telescope, usually many tens of thousands of kilometers away. The chief advantage of internal coronagraphs is their simplicity in point- ing and centering the coronagraph on the central star. How- ever, there is a practical limit to the size of the telescope that can be used, because it must be launched into space. Coron- agraphic observations have already yielded images of widely separated planets and, in one case, of an intervening disk (Figure 10.12).
The appeal of external coronagraphs, which have been studied for many years, is their potential to circumvent many of the light suppression problems faced by internal coronagraphs by instead blocking the stellar light with a free-flying starshade. The main drawback of the external-occulter approach lies in its operational
complexity relative to a single spacecraft – two vehicles must perform properly for this technique to work and source targeting requires aligning the two spacecraft.
At long infrared wavelengths where the planet–star contrast shrinks by several orders of magnitude, coronagraphs would become huge and unwieldy. Instead, interferometery is the favored approach. An infrared interferometer consists in its sim- plest form of two telescopes joined on a structure, or mounted on separate satellites that maintain a controlled distance by pre- cision formation flying. The starlight is suppressed by arranging the light beams coming into the two telescopes to be combined so that at the center of the image, they destructively interfere with each other and cancel the light of the glaring star. Light that is off center, such as from a planet displaced from the star, is not destructively cancelled and so has a much higher contrast than were the system observed with a single telescope alone.