Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://star.arm.ac.uk/~jec/stardust.ps Äàòà èçìåíåíèÿ: Fri May 8 20:14:27 2009 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 21:28:51 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: veil nebula

Searching for Other Planetary Systems
by John E. Chambers
Do planets exist around other stars? If so, where are they? What do they look
like? Are they capable of supporting life? Traditionally, these questions have remained
firmly outside the realm of science, and were left to science fiction writers. Several
times this century, astronomers have claimed to detect extra­solar planets (including
one orbiting Barnard's Star, one of the closest stars to the Sun), but subsequent
searches failed to produce evidence to support any of these claims. Lately, some
scientists even began to despair of finding any planets beyond the solar system. Then,
in the early 1990s two groundbreaking discoveries gave us the first tantalising glimpses
of worlds around other stars.
In 1991, Alexander Wolszczan discovered the first planets to be confirmed around
another star. These new objects---two the size of Earth and one the size of the moon---
are in orbit around a pulsar with the catchy designation PSR 1257 +12. Despite the
importance of this find, it is hard to imagine an environment less hospitable to life than
a planet orbiting a neutron star, with its harsh radiation and feeble warmth, so the
search continued for planetary systems more like our own. Four years later, quite by
chance, Michel Mayor and Didier Queloz found evidence of a Jupiter­sized companion
orbiting 51 Pegasi---a G2 star almost identical to the Sun, and less than 50 light years
distant. The discovery was quickly confirmed by other astronomers, and extra­solar
planets made the transition from science fiction to science fact.
Why do astronomers think planets might be commonplace in the universe? Firstly,
we see the material needed to form planets around many nearby stars. In the 1980s,
observations using the IRAS satellite led to the discovery that more than 100 stars,
including Vega and Fomalhaut, give off more infra­red radiation than expected. These
``infra­red excesses'' are almost certainly produced by disks of dust grains orbiting each
star, and in at least two cases---Beta Pictoris and the star BD +31 ffi 643---these disks
have been imaged directly. Many very young stars, such as the Proplyds (short for
``proto­planetary disks'') in the Orion nebula, are apparently veiled in cloaks of gas
and dust, whereas mature main­sequence stars are not. Where does all this material
go? Probably, much of it forms into planets.
The second reason for thinking extra­solar planets are widespread comes from the
favoured theory for how our own planetary system formed---the planetesimal theory.
In this scenario, the early Sun was surrounded by a disk of gas, dust and ice, much like
the disks we observe around other young stars. Gentle collisions between dust grains
caused them to stick together forming larger conglomerates. This continued until the
1

objects grew to a few kilometres in size. These ``planetesimals'' were large enough
for their gravitational fields to attract more dust grains, and also other conglomerates
smaller than themselves. At the same time, the gas surrounding the Sun kept the
planetesimals' orbits almost circular and coplanar---conditions ideal for further growth,
and also the reason why all the Sun's planets have orbits of this kind.
Eventually most of the planetesimals were used up, and the largest had grown
to several thousand kilometres in size---objects called planetary embryos. Now the
evolution entered a violent and often catastrophic phase, as gravitational tugs between
neighbouring planetary embryos distorted their orbits, causing them to collide with
one another in giant impacts. One of these titanic collisions probably formed the
binary planet we call the Earth and Moon. Another may have stripped off the outer
rocky layers of Mercury, leaving behind a super­dense iron core. Once the dust had
settled, four rocky planets were left orbiting close to the Sun.
Farther from the Sun, where temperatures were cold enough for ice grains to survive
in addition to dust particles, the planetary embryos grew larger. So large, in fact, that
their gravity was strong enough to pull in huge amounts of gas, forming the planets
Jupiter and Saturn, which have enormous atmospheres of hydrogen and helium. A
competing theory holds that these giant planets were created when parts of the Sun's
proto­planetary disk collapsed under their own gravity. Future observations of extra­
solar planets may help to determine which of these scenarios is the correct one. Either
way the conclusion is the same: the disks of gas and dust which we see around many
young stars should eventually form planetary systems.
So, with scientific theory and observation both suggesting that many stars are ac­
companied by planets, how can we actually detect them? The most obvious method---
looking for small, faint objects close to nearby stars---is actually one of the most
difficult. Although a large planet orbiting another star would be bright enough to
detect, the light coming from the planet would be lost in the glare of its brilliant
neighbour, since no telescope in the world is perfect enough to prevent some scattered
light from the star swamping the signal of its much fainter companion.
Because of this, all the search programmes currently in operation exploit ways in
which a planet betrays its presence by modifying the light coming from its parent
star. There are several ways to do this. For example, imagine a star with a planet
whose orbits lies precisely edge­on to us. Each time the planet passes in front of it
the star will dim and then brighten again in a characteristic fashion. Three of these
``transits'' allow us to deduce the planet's orbital period, its size, and its distance from
the star---not bad considering the planet itself remains invisible throughout.
Only a few percent of stars will have planetary systems with the correct orientation
2

to make transits visible, but monitoring the brightness of many stars should detect
several that do. This is the motivation for the French Space Agency's planned COROT
mission. Alternatively, one can look for transits across binary stars in which the
binary's orbit is edge­on to us. Any planets present will probably lie in the same
plane as the binary, making their transits visible. The TEP (``Transits of Extra­
Solar Planets'') network is already looking for these events around the binary star CM
Draconis.
Microlensing is another way to look for planets. When a nearby star passes almost
directly in front of a more distant one, it acts like a lens, magnifying and distorting
the distant star's light for a few weeks or months. If the closer star also has plan­
ets, these can act like miniature lenses, causing brightening that lasts for only a few
hours---brightening that must be due to a planet. The appropriately­named PLANET
programme (``Probing Lensing Anomalies Network'') is now searching for microlensing
events just like this.
At the time of writing, neither microlensing nor the transits technique has led to
the discovery of any planets around other stars. Astrometry, on the other hand, can
claim at least one success. This method looks for the tiny changes in the position of
a star caused by the gravitational tug of a planet in orbit around it. Although widely
adopted by astronomers earlier this century to look for planets, astrometry is actually
very challenging---a planet as big as Jupiter located 30 light years away would cause
its star to shift position by only about one millisecond of arc, equal to the angular
size of a one pence piece 4000 kilometres away. Despite the magnitude of this task,
George Gatewood claims to have detected a planet the size of Jupiter, orbiting the
star Lalande 21185 only 8 light years from Earth.
By far the most successful method for finding extra­solar planets, however, is
Doppler spectroscopy. The spectrum of light from a star consists of all the colours
of the rainbow apart from a series of small gaps where the light is absorbed by gas in
the star's atmosphere before it can start the long journey towards us. These ``spectral
lines'' have a very distinct pattern that can be reproduced by samples of gas on Earth.
However, the lines in a star's spectrum are offset by an amount that depends on the
star's speed relative to us. By measuring this ``Doppler shift'', we can see whether
a star repeatedly moves toward and away from us due to the gravitational pull of a
planet in orbit.
Doppler spectroscopy was the technique used by Mayor and Queloz to discover the
planet around 51 Pegasi, and is also the method chosen by the most prolific planet
hunters of all---Geoff Marcy and Paul Butler. To date, this technique has uncovered
nine planets around stars similar to the Sun, and also a host of brown dwarfs---stars
3

Parent Star Mass Distance from Orbital Period Distance from Sun
(Jupiter=1) star (AU) (days) (light years)
51 Peg 0.5 0.05 4.23 50
upsilon And 0.7 0.057 4.61 54
55 Cnc 0.8 0.11 14.6 44
Lalande 21185 0.9 1.5 5.8 years 8
rho CrB 1.1 0.23 39.6 54
16 Cyg B 1.5 1.7 804 72
47 UMa 2.8 2.1 2.98 years 46
tau Boo 3.9 0.046 3.31 49
70 Vir 6.6 0.43 117 59
HD 114762 10 0.3 84.1 91
too small to shine by burning ordinary hydrogen, but large enough to burn rarer
deuterium.
What do we know about these new planets? Well, they are all very massive---
comparable in size to Jupiter or even larger---so they are probably mostly made of gas
like Jupiter and Saturn. Many of the new planets also lie extremely close to their star---
much closer than Mercury's distance from the Sun (see table). This is curious because
in the solar system the giant planets lie much further out. Astronomers are now having
to modify their theories to explain how these new oddballs might have formed far from
their stars and then moved inwards to their present locations. One idea is that the
gas in these stars' proto­planetary disks robbed the newly­formed planets of energy,
causing them to spiral in towards the star. However, it will probably be some time
before we can piece together precisely what happened.
Do these planets have life? At the moment it is impossible to say. An easier
question to answer is whether the planets are capable of supporting life---are they
``habitable''? This is usually taken to mean that they can have liquid water on their
surfaces. Clearly, if these planets are gas giants, like Jupiter, they will have no solid
surfaces, and so they would not be habitable. Still, it is quite likely that they are
accompanied by smaller, rocky moons like the Galilean satellites of Jupiter. Could
4

these be habitable?
The main requirement for water to be liquid is that, like Goldilocks's porridge, the
temperature is not too hot, and not too cold. For geologically active planets like Earth,
orbiting a star like the Sun, this means they have to be at a distance of between 0.9
and 1.5 AU (1 AU is the average Earth­Sun distance). In this ``habitable zone'', a
planet's climate is stabilised by the greenhouse effect, in which carbon dioxide, and
other gases in the atmosphere, trap some of the Sun's heat close to the ground. A
combination of volcanic eruptions and the weathering of rocks ensures that there is
always just the right amount of carbon dioxide in the atmosphere to stop the oceans
freezing or boiling away.
Sadly, none of the planets discovered so far lies within the critical range of distances
for it, or its moons, to be habitable. Still, it is early days yet. Pulsar planets aside,
each of the new objects is orbiting a different star, so there is plenty of room left in each
system for additional planets that may have escaped detection. These could become
apparent as the sensitivity of the observational programmes improves---already there
is some evidence that a second planet is orbiting the star 55 Cancri. In addition,
astronomers have looked for planets around only a few hundred stars so far---there are
many more stars to be examined.
Improving our existing search programmes will surely produce more planet discov­
eries, but to find out if life exists elsewhere we need a more radical approach. The
European space agency has recently proposed just such a scheme. Called DARWIN,
it will put a set of large infra­red telescopes into space, out near the orbit of Jupiter.
Here, freed from the Earth's atmosphere, the combined sensitivity of these telescopes
will be sufficient to detect Earth­sized planets around nearby stars. Not only that, the
telescopes will be able to identify gases in the planets' atmospheres. Finding a planet
with an oxygen­rich atmosphere---a state of affairs currently unique to Earth---would
be a strong indication that life is at work on the planet's surface.
If it receives the go­ahead, DARWIN will be launched in 2009. Not long afterwards,
we could become the first generation in history to know that we are not alone in the
cosmos.
5