The
first widely accepted detection of extrasolar planets was made by Wolszczan
(1994). Earth-mass and even smaller planets orbiting a pulsar were detected by
measuring the periodic variation in the pulse arrival time. The planets
detected are orbiting a pulsar, a "dead" star, rather than a dwarf
(main-sequence) star. What is heartening about the detection is that the
planets were probably formed after the supernova that resulted in the pulsar. Thereby
demonstrating that planet formation is probably a common rather than a rare
phenomena. The pulsar planets are indicated by the open diamonds in the figure
below.
Doppler
spectroscopy is used to detect the periodic velocity shift of the
stellar spectrum caused by an orbiting giant planet. (This method is also
referred to as the radial velocity method.) From ground-based observatories,
spectroscopists can measure Doppler shifts greater than 3 m/sec due to the
reflex motion of the star This corresponds to a minimum detectable mass of 33Me
/ sini for a planet at 1 AU from a one solar-mass (1 Mo)
star, where i is the inclination of the orbital pole to the
line-of-sight (LOS). This method can be used for main-sequence stars of
spectral types mid-F through M. Stars hotter and more massive than mid F rotate
faster, pulsate, are generally more active and have less spectral structure,
thus making to more difficult to measure their Doppler shift. The minimum
detectable planet mass increases as the square root of the planet's orbital
size, as shown in the figure below (the red ascending line).
The
planets already detected with this method are indicated by the solid diamonds
in the figure below. Note the mass is given in Earth masses, Me.
Astrometry
is used to look for the periodic wobble that a planet induces in the
position of its parent star. The minimum detectable planet mass gets smaller in
inverse proportion to the planet's distance from the star. For a space-based
astrometric instrument, such as the planned Space Interferometry Mission (SIM), that could
measure an angle as small as 2 micro-arcsec, a minimum planet of mass of 6.6Me
could be detected in a 1 year orbit around a 1 Mo
star that is 10 pc from the Earth (gray descending line for stars out to 10 pc)
and a 0.4 MJ planet in a 4 year
orbit (dark-blue descending line for stars out to 500 pc).
From
the ground, the Keck telescope is being equipped to measure angles as small as
20 micro-arc seconds, leading to a minimum detectable mass in a 1 AU orbit of
66Me for a solar-mass star
at 10 pc.
The
limitations to this method are the distance to the star and variations in the
position of the photometric center due to star spots. There are only 33
non-binary solar-like (F, G and K) main-sequence stars within 10 pc of the
Earth. The furthest planet from its star that can be detected is limited by the
time needed to observe at least one orbital period. This limit is indicated by
the dashed light-blue vertical line chosen to be at 10 years in the figure
below. There are no planet detections that have been confirmed using this
method.
4. Transit
Photometry
Photometry
measures the periodic dimming of the star caused by a planet passing in
front of the star along the line of sight from the observer. Stellar
variability on the time scale of a transit limits the detectable size to about
half that of Earth for a 1 AU orbit about a 1 Mo
star or Mars size planets in Mercury-like orbits with four years of observing.
Mercury-size planets can even be detected in the habitable-zone of K and M
stars. Planets with orbital periods greater than two years are not readily
detectable, since their chance of being properly aligned along the line of
sight to the star becomes very small.
In
the figure below, the white region represents the full range of planet masses
and orbits that the Kepler Mission can detect. Giant outer planets that
produce a transit signal of 1% ( 120 times that of an Earth, i.e., a SNR
>1000) but have orbital periods greater than 2 years can be followed up with
Doppler spectroscopy or ground-based photometry (green horizontal line in the
figure below).
Giant
planets in inner orbits can also be detectable independent of the orbit
alignment, based on the periodic modulation of their reflected light. For the
10% of these that have transits, the transit depth can be combined with the
mass found from Doppler data to determine the density of the planet as has been
done for the case of HD209458b and see if these inner giants are
"inflated".
Doppler
spectroscopy and astrometry (SIM) measurements can be used to search for any
giant planets that might also be in the systems discovered using photometry.
Since the orbital inclination must be close to 90° (sin i=1.) to cause
transits, there is very little uncertainty in the mass of any giant planet
detected.
Detection Limits for Planets Around Solar-Like Stars
The
limiting sensitivities for a solar-like star are shown for:
ˇ
Photometry with Kepler (the white region above and to the left
of the light-blue lines), COROT (above and to the left of the lavender line);
and ground-based photometry (above the solid green line)
ˇ
Doppler spectroscopy at 3 m/s (above and to the left of the red line);
Planets detected using this method: the first 49 are shown as filled diamonds.
Astrometry with SIM at 2ľas (above and to the right of the gray and dark-blue
lines)
Solar system objects: Mercury, Venus , Earth, Mars, Jupiter,
Saturn, Uranus and Neptune are shown as solid blue dots 
.
The
limits to the maximum orbits are related to the length of time needed to
observe one or more complete orbits to see the periodic phenomena repeat
its signature or by the lifetime of a space mission.
Photometry
is the only practical method for finding Earth-size planets in the continuously
habitable zone. This unique search space is shaded
green in the figure.
