VETERAN planet hunter Geoff Marcy has a wide grin on his face. He鈥檚 found
another planet, and one that鈥檚 more significant than any he has spotted so far.
鈥淓very new planetary system reveals some new quirk that we didn鈥檛 expect. We鈥檝e
found planets in small orbits and wacky eccentric orbits.鈥� But no one had found
anything that looked remotely like our own Solar System. Now that has
changed.
Last August, Marcy鈥檚 colleague Debra Fischer announced that a second planet
has been discovered orbiting the star 47 Ursae Majoris. What makes this system
so special is that both planets have almost circular orbits鈥攍ike those of
the planets in our own Solar System. If these planets were orbiting our Sun,
they would be somewhere in the asteroid belt. They are not too dissimilar in
mass to our giant worlds Jupiter and Saturn, giving this planetary system an
uncanny resemblance to our own. Even the parent star is Sun-like.
At last we know there are some planetary systems where a comfortable twin to
Earth might exist. And for astronomers trying to understand how planetary
systems are born, it鈥檚 a welcome respite from a period of feverish work, trying
to revise their theories to fit all those weird planets. The discovery of the
Solar Solar鈥檚 near-twin is now providing some reassurance that their revamped
theories of planet formation are at last on the right lines.
Advertisement
It has all taken some time to achieve, because it starts with one of the most
difficult tasks for an observational astronomer: to track down extrasolar
planets. You can鈥檛 do this just by pointing a telescope. 鈥淩ight now, it鈥檚
impossible even with our most powerful telescopes to see a planet,鈥� explains
Debra Fischer, who works with Marcy and Paul Butler at the Lick Observatory in
California. 鈥淭he planet reflects such a tiny amount of starlight. It鈥檚 like a
little firefly buzzing around in the headlights of an oncoming car.鈥�
But astronomers knew that a star with planets in tow would 鈥渨obble鈥� by a very
small amount, just a few metres per second, as the planets tug the star back and
forth as they orbit. Several teams were on the case. First past the post were
the Swiss astronomers Didier Queloz and Michel Mayor, who in 1995 discovered a
planet half the mass of Jupiter orbiting the star 51 Pegasi. Soon the planets
began to roll in. In just six years, astronomers have turned up 66 systems
containing 74 planets.
But even as the tally of extrasolar planets began to mount, the planet
hunters became increasingly uncomfortable. They had confirmed, to everyone鈥檚
satisfaction, that stars have planets in tow. But these were not like the
obedient worlds of our Solar System, with nice near-circular orbits and gas
giants located a decent distance from their parent star. These were planets from
hell鈥攁nd they were breaking all the rules.
Queloz realised that all was not well right from the start when he discovered
the planet orbiting 51 Pegasi. The star wobbled with a period of just over four
days鈥攊ndicating that the planet circled its star absurdly close in. Even
Mercury, the closest planet to the Sun, takes 88 days to orbit. Queloz鈥檚 world
is so close that it is blowtorched by the star鈥檚 heat. The planet鈥檚 temperature
must be higher than that of a blast furnace, and its atmosphere swollen to
grotesque proportions.
It turned out to be the first of many鈥攁 world similar in mass and
make-up to our Jupiter, but 20 times closer to its parent star than the Earth is
to the Sun. The rogue worlds rapidly acquired the nickname 鈥渉ot Jupiters鈥�.
It certainly wasn鈥檛 what the planet hunters had been expecting. They were
looking for a system much like our own, with massive Jupiter-like planets in
stable, near-circular orbits far from their star. This wasn鈥檛 just a result of
conservative thinking; astronomers simply thought they knew enough about the
basic recipe for making new worlds.
A circular disc of gas and dust surrounds a newly born star. Within the disc,
the matter clumps together to make rocky or icy lumps about a hundred kilometres
across. These, in turn, build up to become planets. The puzzle of the hot
Jupiters was that only out at a much greater distance should you get enough
material to build a giant planet.
But one theorist was unfazed by the hot Jupiters. Since the 1980s, Doug Lin,
of the University of California, Santa Cruz, had been calculating how the
gravity of the massive disc around a star would affect an embryonic planet
growing within. He found that the disc is very good at robbing energy from the
planet. As a result, the planet naturally spirals in towards its parent star.
鈥淚t wasn鈥檛 surprising to me when the observation showed that a planet had formed
and migrated inwards鈥攖hat in fact caused me a great deal of joy,鈥� Lin
recalls. 鈥淲hat surprised me was how it could migrate all the way in and then
stop almost next to the surface of the star.鈥�
Lin suspects the answer lies in the central star. If it is spinning rapidly,
tidal forces can transfer energy to the planet: the star slows down, while the
energy pumped into the planet鈥檚 orbit stops it from shrinking any more. A
similar thing is happening with the Earth-Moon system; the Earth鈥檚 spin is
gradually slowing down as it gives energy to the Moon, which is being pushed
away from us.
Pawel Artymowicz at the University of Stockholm in Sweden has a different
idea. He thinks there鈥檚 something about discs we don鈥檛 understand yet, which
makes them stabilise when two bodies are orbiting with very short periods. It鈥檚
based on the fact that many young stars also live together in very close pairs,
when theory says they too should have spiralled together and coalesced. But
Artymowicz admits that this is one idea among many: 鈥淭his is still a hard-hat
construction area for theories,鈥� he says.
In any case, a runaway giant would have dire consequences for any Earth-like
planet. Gravitational drag from the wayward hot Jupiter would sap orbital energy
from a smaller world and push it into the star. Theory says giant planets should
spiral inwards very quickly, in a matter of a few million years鈥攜et in our
Solar System we seem to have had a lucky escape as Jupiter has stayed out at a
safe distance.
But Lin thinks there may be more to it than luck. He believes our Solar
System could have had previous generations of planets that ended up incinerated
within the Sun. 鈥淲hat we see in our Solar System may be the survivors of
repeated generations of proto-planets that underwent infant mortality,鈥� Lin
explains. 鈥淭he planets we see today may be the last of the Mohicans.鈥�
As if hot Jupiters were not enough, the planet hunters also began to turn up
other bizarre objects: wayward worlds in wildly eccentric orbits. 鈥淲e found a
planet around 70 Virginis that was a real weirdo,鈥� recalls Marcy. 鈥淭he orbit was
so elongated that people thought to themselves, is this a planet or is it some
other kind of beast? But sure enough, it turns out it鈥檚 a planet, and we鈥檝e
found some 30 or 40 more of these planets in elongated orbits. And we now
realise that most extrasolar planets are not in circular orbits, but in these
elliptical ones.鈥�
These planets too pose a deadly threat to worlds like the Earth. Their
gravity tends to sling smaller worlds away into the distant cosmos. Again, most
theorists were caught on the hop. They had assumed the regular disc surrounding
a young star should spawn a planetary system with circular orbits.
The answer, according to Artymowicz, can be found by looking more closely at
the original disc. His computer simulations of the protoplanetary disc reveal
beautiful swirling 鈥渁rms鈥�, like a spiral galaxy. If the disc is massive enough,
the gravity of these arms can divert a young planet into an eccentric path.
Lin instead blames the highly eccentric orbits on other planets orbiting the
star. 鈥淎s in a family, the siblings tend to perturb each other, and sometimes
this interaction becomes so strong it can break up the family.鈥� He has applied
this theory to Upsilon Andromedae, a system that includes a hot Jupiter and two
eccentric planets.
But this answer doesn鈥檛 satisfy Alan Boss from the Carnegie Institution of
Washington. For him, the eccentric solar systems are forcing us to find a new
theory for making the heaviest planets. Boss鈥檚 calculations show that the
gradual assembly of small chunks of rock and ice can鈥檛 build anything much
bigger than Jupiter. Instead, he envisages the primeval disc spontaneously
breaking up into several large lumps of gas, each of which becomes a giant
planet. This theory has the advantage that the unstable lumps of gas won鈥檛
generally be following circular paths around the star.
Amid this welter of new theories explaining why other planetary systems
should be different from ours, Fischer鈥檚 announcement of a Solar System 鈥渢win鈥�
brought people up sharply. The star 47 Ursae Majoris had been on Marcy鈥檚 hit
list from the very beginning, and its bigger planet was one of the earliest to
be discovered. This planet circles its star in just under three years. Fischer
kept on returning to the star, however, concerned with niggling discrepancies
between the prediction and the star鈥檚 actual wobble. Once she鈥檇 taken away the
effect of the first planet, she realised there was a much smaller wobble, with a
period of seven years.
Whereas the first planet was 2.5 times heavier than Jupiter, the second
weighed in at about three-quarters of Jupiter. Intriguingly, their masses are in
the same ratio as the masses of Jupiter and Saturn in our system. And their
distances from the suns are in roughly the same ratio, too, though the 47 Ursae
Majoris planets are less than half the distance out.
So this has proved that our Solar System is not unique, but it begs the
question of why our Solar System and the planets of 47 Ursae Majoris are so
unusual.
Lin puts it down to weight and age. The more massive the planets are, the
more intensely they interact with each other and the more quickly they will
disrupt each other鈥檚 orbit. 鈥淥ur own Solar System is atypical,鈥� he says,
鈥渂ecause we are relatively low in mass.鈥� Given enough time, though, even the
orbits of our lightweight planets will become unstable. Fortunately for us
that鈥檚 many billions of years in the future.
On the theoretical side, then, astronomers have worked out why many planetary
systems are so different from our Solar System. Indeed, they now think that
massive discs around young stars would naturally turn into something very
different鈥攕ystems of hot Jupiters and wayward planets. We live in a
different kind of place because both the original spiral-armed disc and the
subsequent planets are relatively lightweight.
Even the system around 47 Ursae Majoris isn鈥檛 exactly like ours.
Identical twins will be harder to find鈥攁nd will require a longer vigil.
Jupiter takes about 12 years to circle the Sun, and Saturn takes 30. The present
planet search has only been going for a dozen years, so it would only now be
turning up cousins of our giant worlds, those with orbits of less than a dozen
years or so. To find a true twin of our Solar System, we may need another 10
years of data or more.
So it鈥檚 early days yet, but Butler is prepared to speculate on the number of
Solar System lookalikes. 鈥淲e can make a preliminary guess that about 5 per cent
of planetary systems are in circular orbits,鈥� he ventures. Artymowicz tends to
agree: 鈥淚 think it鈥檚 less than 10 per cent.鈥�
These observations need a telescope dedicated to planet searches. Fischer
looks forward to the day when she can monitor stars every night and detect
distant planetary companions. 鈥淲e hope to raise $5 million to purchase a
dedicated 2-metre telescope.鈥� The telescope would be located at the Lick
Observatory on Mount Hamilton, above Silicon Valley.
But the dream of every planet hunter is not to find more Jupiters and
Saturns鈥攊t is to find another Earth. And the only way to do that is from
the cold, clear vantage of space. Queloz is thrilled at the prospect, especially
a mission now being planned by the European Space Agency. 鈥淭here鈥檚 this ultimate
European mission called Darwin,鈥� he explains. 鈥淭he big goal is to see an
Earth-like planet and analyse its light. We want to see if we can detect oxygen
or water鈥攜ou know, the kind of holy elements that are linked to life.鈥�
The American counterpart to Darwin is the Terrestrial Planet Finder鈥攁n
ambitious mission that involves a whole fleet of telescopes sailing together in
space. Like Darwin, TPF will use destructive interference to cancel out light
from the star, making the planets visible. These spacecraft鈥攐r, more
likely, a combined mission鈥攎ay fly within 15 or 20 years. Then we should
learn whether our Solar System is one of a kind or not.
Artymowicz predicts that only a few per cent of systems with a giant planet
will have an Earth-like world: in most cases, the Jupiter will have gone off on
a rampage. But he also thinks there may be many planetary systems which started
off with even less matter than the Solar System. In that case, there鈥檇 be no
disruptive giant planets at all. 鈥淲ho knows,鈥� he muses, 鈥渕aybe 50 per cent of
stars have big chunks of rock鈥攎aybe Earths鈥攃ircling around
迟丑别尘.鈥�