Exoplanet Detection Methods: How Do We Find New Planets?
What is an exoplanet?
An exoplanet is a planet that orbits a star other than our Sun (there are also free-floating planets that aren't orbiting a host star). This is a relatively new field of research in astronomy and is particularly exciting for its input to the search for extraterrestrial life. As of April 1, 2017, there have been 3607 exoplanets discovered. The definition of a solar system planet, set by the International Astronomical Union (IAU) in 2006, is a body that meets three criteria:
- It's in orbit around the Sun.
- It has sufficient mass to be spherical.
- It has cleared its orbital neighbourhood (i.e. the gravitationally dominant body in its orbit).
There are multiple methods that are used to detect new exoplanets.
Directly imaging exoplanets is extremely challenging because of two effects. There is a very small brightness contrast between the host star and the planet and there is only a small angular separation of the planet from the host. In plain english, the star's light will drown out any light from the planet because of us observing them from a distance much larger than their separation. To enable direct imaging both of these effects need to be minimised.
The low brightness contrast is usually addressed by using a coronagraph. A coronagraph is an instrument which attaches to the telescope to reduce the light from the star and hence increase the brightness contrast of nearby objects. Another device, called a starshade, is proposed which would be sent into space with the telescope and directly block the star light.
The small angular separation is addressed by using adaptive optics. Adaptive optics counteract the distortion of light due to the Earth's atmosphere (atmospheric seeing). This correction is performed by using a mirror whose shape is modified in response to measurements from a bright guide star. Sending the telescope into space is an alternative solution but it is a more expensive solution. Even though these issues can be addressed and make direct imaging possible, direct imaging is still a rare form of detection.
Radial velocity method
Planets orbit around a star because of the gravitational pull of the star. However, the planet also exerts a gravitational pull on the star. This causes both the planet and the star to orbit around a common point, called the barycentre. For low mass planets, such as Earth, this correction is only small and the movement of the star is only a slight wobble (due to the barycentre being within the star). For larger mass stars, such as Jupiter, this effect is more noticeable.
This movement of the star will cause a Doppler shift, along our line of sight, of the stellar light that we observe. From the Doppler shift, the velocity of the star can be determined and hence we can calculate either a lower limit for the planet's mass or the true mass if the inclination is known. This effect is sensitive to the orbital inclination (i). Indeed, a face-on orbit (i = 0°) will produce no signal.
The radial velocity method has proven very successful in detecting planets and is the most effective method for ground-based detection. However, it is unsuitable for variable stars. The method works best for nearby, low mass stars and high mass planets.
Instead of observing the doppler shifts, astronomers can try to directly observe the star's wobble. For a planet detection, a statistically significant and periodic shift in the centre of light of the host star image needs to be detected relative to a fixed reference frame. Ground based astrometry is extremely difficult because of the smearing effects of the Earth's atmosphere. Even space based telescopes need to be extremely precise for astrometry to be a valid method. Indeed this challenge is demonstrated by astrometry being the oldest of the detection methods but so far only detecting one exoplanet.
When a planet passes between us and its host star, it will block out a small amount of the star's light. The period of time while the planet passes in front of the star is called a transit. Astronomers produce a light curve from measuring the star's flux (a measure of brightness) against time. By observing a small dip in the light curve, the presence of an exoplanet is known. Properties of the planet can also be determined from the curve. The size of the transit is related to the planet's size and the duration of the transit is related to the planet's orbital distance from the sun.
The transit method has been the most successful method for finding exoplanets. NASA's Kepler mission has found over 2,000 exoplanets by using the transit method. The effect requires an almost edge-on orbit (i ≈ 90°). Therefore, following up a transit detection with a radial velocity method will give the true mass. As the planetary radius can be calculated from the transit light curve, this allows the planet's density to be determined. This as well details about the atmosphere from light passing through it provides more information about the planets composition than other methods. Precision of transit detection depends on any short term random variability of the star and hence there is a selection bias of transit surveys targeting quiet stars. The transit method also produces a large amount of false positive signals and as such usually requires a follow up from one of the other methods.
Albert Einstein's theory of general relativity formulates gravity as the curving of spacetime. A consequence of this is that the path of light will be bent towards massive objects, such as a star. This means that a star in the foreground can act as a lens and magnify light from a background planet. A ray diagram for this process is shown below.
Lensing produces two images of the planet around the lens star, sometimes joining up to produce a ring (known as an 'Einstein ring'). If the star system is binary the geometry is more complicated and will lead to shapes known as caustics. The lensing of exoplanets takes place in the microlensing regime, this means that the angular separation of the images is too small for optical telescopes to resolve. Only the combined brightness of the images can be observed. As stars are in motion these images will change, the brightness changes and we measure a light curve. The distinct shape of the light curve allows us to recognise a lensing event and hence detect a planet.
Exoplanets have been discovered through microlensing but it depends on lensing events that are rare and random. The lensing effect isn't strongly dependant on the planet's mass and allows low mass planets to be discovered. It can also discover planets with distant orbits form their hosts. However, the lensing event won't be repeated and hence the measurement can't be followed up. The method is unique when compared to the others mentioned, as it doesn't require a host star and therefore could be used to detect free floating planets (FFPs).
1991 - First exoplanet discovered, HD 114762 b. This planet was in orbit around a pulsar (a highly magnetised, rotating, small but dense star).
1995 - First exoplanet discovered through radial velocity method, 51 Peg b. This was the first planet discovered orbiting a main-sequence star, like our sun.
2002 - First exoplanet discovered from a transit, OGLE-TR-56 b.
2004 - First potential free-floating planet discovered, still awaiting confirmation.
2004 - First exoplanet discovered via gravitational lensing, OGLE-2003-BLG-235L b/MOA-2003-BLG-53Lb. This planet was independently discovered by the OGLE and MOA teams.
2010 - First exoplanet discovered from astrometric observations, HD 176051 b.
2017 - Seven Earth-sized exoplanets are discovered in orbit around the star, Trappist-1.
© 2017 Sam Brind