You are here: Department / Mad Star Home / Teaching Resources / Extrasolar Planets and the Search for Life / The Search for Extrasolar Planets

The Search for Extrasolar Planets

Extrasolar planets, as the name implies, are planets which orbit not our own Sun, but other stars in the Galaxy. In this lecture, I discuss five techniques which can be used to find extrasolar planets, and review the properties of the planets found to date. More information can be found in Universe, Sections 7.9.

  1. Planets, Brown Dwarfs and Low-Mass Stars
  2. The Significance of Extrasolar Planets
  3. Techniques for finding Extrasolar Planets
  4. Pulsar Timing
  5. Doppler Spectroscopy
  6. Astrometry
  7. Transit Photometry
  8. Microlensing
  9. Planets found To Date

Planets, Brown Dwarfs and Low-Mass Stars

Low-mass celestial objects may be divided into three classes:

  • Low-Mass Stars are significantly less massive than our own Sun, yet still produce their own energy through nuclear fusion reactions in their cores. These reactions will only occur in objects more massive than 0.08 solar; anything below this mass is a brown dwarf or a planet.
  • Brown Dwarfs form from the contraction of nebula, much like a star; however, their mass is too small (below 0.08 solar) for nuclear fusion to commence, so they can be regarded as 'failed stars'.
  • Planets form from a protoplanetary disk around a star (or, conceivably, a brown dwarf), as discussed in Lecture 1.

Clearly, the only distinction between brown dwarfs planets is in their formation processes.

The Significance of Extrasolar Planets

The search for extra-solar planets is considered important for four main reasons:

  1. To test our current understanding of the formation of (extra)solar systems.
  2. To further our insight into the formation of individual planets.
  3. To assist in the search for extraterrestrial life.
  4. To provide insight into complementary studies of brown dwarfs and low-mass stars.

Techniques for finding Extrasolar Planets

Extrasolar planets are incredibly difficult to detect. This is because they shine not by their own light, but by light reflected by the star which they orbit. As a consequence, they are much dimmer than their parent star (in the case of Jupiter, for instance, by a factor of 100 billion), and any attempts to detect them by their own light are doomed to failure.

Therefore, indirect methods must be used to find extrasolar planets. There are five principal techniques which are currently used:

  1. Pulsar timing
  2. Doppler spectroscopy
  3. Astrometry
  4. Transit photometry
  5. Microlensing

All of these techniques rely on the fact that a planet exerts a small influence on its parent star as it travels around its orbit. By observing changes in the parent star, the existence of the planet can be deduced. Since the changes become larger as the planet becomes more massive, it is always easier to detect Jovian planets than to detect terrestrial ones.

Pulsar Timing

A pulsar is a rapidly-spinning neutron star with a strong magnetic field (see Universe, Chapter 23). Radiation produced by the neutron star is focused into two oppositely-directed beams by the magnetic field. As the star rotates, the beam is swept across the sky; if the beam intercepts the Earth once per rotation, then brief but regular pulses of radiation are seen, much like a lighthouse.

When a planet is introduced, the mutual gravitational pull between it and the pulsar means that they both orbit about their common centre of mass. For two equal-mass objects, the centre of mass lies exactly halfway between them; in other situations, the centre of mass lies closer to the more-massive object. In the case of a pulsar and a planet, the centre of mass will lie very close to the pulsar, since it is much heavier than the planet. Therefore, during one orbit the pulsar will move a much lesser distance than the planet.

A star and planet orbiting around their common centre of mass

A star and planet orbiting around their common centre of mass

However, even thought the pulsar's 'wobble' is small, it has an effect on the timing of the pulses emitted by it. When the pulsar is moving away from the Earth, the time between each pulse becomes slightly longer; conversely, when the pulsar is moving toward the Earth, the time between pulses becomes slightly shorter.

By measuring these periodic changes in pulse timing, it is possible not only to deduce the existence of a planet orbiting around the pulsar, but also to estimate the semi-major axis of the planet's orbit, and place a lower limit on the mass of the planet.

To date, only one planet has been detected using this method, by Wolszczan in 1994. Detections of this sort are not of much interest in the search for extrasolar planets, since pulsars are created in supernovas, huge stellar explosions which would doubtless wipe out any life on nearby planets.

Doppler Spectroscopy

The wobble caused by an orbiting planet can also influence the light coming from a star. As the star is moving toward the Earth, the light becomes blue-shifted by the Doppler effect, much like the siren of an approaching ambulance appears higher pitched. Half an orbital revolution later, as the star is then moving away from the Earth, the light becomes red-shifted.

By making precise measurements of the frequency of absorption lines in the star's spectrum, it is possible to see this alternate blue- and red-shift effect, and infer the presence of a planet. As with the pulsar timing technique, the Doppler spectroscopy approach allows one to estimate the semi-major axis of the planet's orbit, and place a lower limit on the mass of the planet.

The principles of the Doppler spectroscopy approach

The principles of the Doppler spectroscopy approach

Beginning with the detection of a planet around the star 51 Pegasi, by Meyor & Queloz in 1995, the Doppler spectroscopy technique has been the most successful so far in finding extrasolar planets. However, it must always be used with caution: some types of star show Doppler effects in their spectra, even though there is no planet present. These effects are due to seismic activity in the star, which causes the star's surface to move around periodically. The resulting blue- and red-shifts in the star's absorption lines might be mistakenly attributed to the wobble caused by an orbiting planet.


A third technique which relies on the wobble caused by an orbiting planet is that of astrometry. This technique focuses on measuring how the position in the sky of a star varies as the star wobbles. If the extrasolar system is seen face-on, this variation will take the form of a circular motion; likewise, if the system is seen edge-on, the variation will appear as a back-and-forth motion on the sky.

The principles of the astrometry approach

The principles of the astrometry approach

The advantage of the astrometry technique is that, unlike pulsar timing or Doppler spectroscopy, it is possible to 'estimate the planet's mass directly', rather than merely being able to place a lower limit on it. However, astrometry has problems of its own. Rather than being able to measure the actual diameter of the star's wobble, it is only possible to measure the angular diameter. Due to the perspective effect, this angular diameter diminishes as one considers stars further and further from the Earth, even if the actual diameter remains the same.

As a consequence, the astrometry technique only works for stars close to the Earth, and even then requires extremely accurate equipment to measure the very small changes in a star's position. To date, only one extrasolar system has been found using this technique: Lalande 21185, discovered by Gatewood, and consisting of two planets orbiting a star (although, of course, there may be many more undetected planets hiding in the system!).

Transit Photometry

The technique of transit photometry relies on the fact that, if an extrasolar system is seen edge-on from Earth, then once every orbit the planet will cross the face of the star and cause a partial eclipse. For the duration of this transit, the star will appear less bright than it would otherwise, and detections of such periodic dimming can be used to deduce the presence of a planet.

The degree of dimming depends on the relative surface areas of the star and planet; therefore, even if the planet is Jupiter-sized, typically only a 1% change in brightness can be expected. Fortunately, however, this is well within the grasp of modern photometers, used measure the apparent brightness of celestial objects. In fact, with space-based telescopes equipped with a photometer, it should be possible to detect eclipses caused by Earth-sized planets; this is the aim of the upcoming Kepler mission planned by NASA.

The principles of the transit photometry approach

The principles of the transit photometry approach

The orbital period of a planet discovered using transit photometry can be found by measuring the time delay between one eclipse and the next. Likewise, the orbital speed can be estimated from the duration of an individual eclipse. Knowing already that the extrasolar system is being viewed edge-on, it is possible to estimate the mass of the planet from these two pieces of information. From the time taken to go from brightness to dimness (stage 2 in the figure above), it is also possible to estimate the radius and the density of the planet, which are useful to know if we are interested in the composition of the planet.

The principal problem with the transit photometry technique is that it only works for those extrasolar systems which are viewed edge on. Since this configuration is rather unlikely, only a few planets have been discovered using this technique.


Microlensing is by far the most fancy of the five techniques discussed here. It relies on an effect predicted by Einstein's General Theory of Relativity: that light rays can be bent by a sufficiently-strong gravitational field. In the case of microlensing, light rays from a distant star are bent by the gravitational field of a nearby star, which happens to be situated in the line of sight. This bending means that more light from the distant star reaches the Earth than would otherwise be the case; the nearby star is behaving like a lens, whose focusing action makes the distant star appear brighter than it would otherwise be.

In almost all cases, the positioning of the lensing star on the line-of-sight to the lensed star is only temporary, since the two stars are moving relative to one another. This means that observations of microlensing (made using sensitive photometers) reveal a steady brightening of the lensed star, as the lensing star moves into position, followed by a subsequent dimming of the lensed star as the lens moves away again.

When a planet is orbiting the lensing star, its own gravitational field can contribute to the bending of light rays, and it behaves like a defect in the lens. This defect will produce a narrow spike in the brightness of the lensed star, which can be used to infer the presence of the planet.

[[Image:lensing.png|50%|center|frame|The simulated magnification of a star's brightness, caused by a lensing star with an orbiting planet"]

Unfortunately, microlensing events are very infrequent, and the microlensing technique has to date not managed to find any extrasolar planets. However, if a detection is made, it will be possible to establish the mass of the planet responsible for the defect, by measuring the height and width of the spike in the brightness of the lensed star.

Planets found To Date

So far, over 100 extrasolar planets have been found. Almost all of these have been discovered using the Doppler spectroscopy technique; therefore, only lower limits for their masses are currently established. The histogram below shows the distribution of these lower-mass limits, over the mass range 0-15 Jupiter masses (1 Jupiter mass is approximately 0.001 solar masses).

The distribution of the lower-mass limits of all currently-known planets

The distribution of the lower-mass limits of all currently-known planets

Clearly, the majority are concentrated around 1 Jupiter mass (although keep in mind that this histogram gives lower limits!). A few much-heavier planets have been found, up to about 70 Jupiter masses; none have been found higher than this limit, since at 80 Jupiter Masses (=0.08 Solar masses), nuclear fusion begins, and the extrasolar system becomes a binary-star system. No planets have been found yet with masses similar to the Earth.

A puzzling aspect of the planets detected so far is that most appear very close (about 1 AU) to the star they are orbiting. Given the fact that these planets are Jovian-like gas giants, it is difficult to reconcile their existence with the Condensation Model for the formation of the Solar System (see [link:diploma-1|Lecture 1]), which predicts no Jovian planets interior to about 4 Au. The solution to this conundrum is that the extrasolar planets may have formed at a distance from their parent star, but then, through interactions with the remaining protoplanetary disk, spiraled closer in. It is not yet clear how often this planetary migration scenario will occur, since the Doppler spectroscopy technique (which accounts for the majority of planets found) is most sensitive to planets close to the star, and is therefore automatically biased toward finding planets which have already undergone migration.

Updated 2009-10-13 12:54:17