The planet hunters

News SG

Exoplanets
The planet hunters

May 31st 2007
From The Economist print edition
The search for alien life is yielding weird new worlds at a remarkable rate

EARLY science is a lot like stamp collecting. In any new field of endeavour, the first priority is to gather lots of specimens. In planetary science the result is a menagerie of exotic new worlds, some 236 of which were this week confirmed as exoplanetsthat is, planets outside the solar system. Over the past few days members of the American Astronomical Society meeting in Honolulu have also been tussling over whether the first Earth-like planet, announced at the end of April, has actually been found.

The rate of progress is extraordinary. The first exoplanet orbiting a normal star was discovered a mere 12 years ago. Before that, the answer to the rhetorical question are we alone?, might quite possibly have been yes. Exoplanets were the stuff of science fiction. Now the hunt is well and truly on for places that are capable of sustaining life, and in some ways the speculations of the sci-fi writers have been far outstripped by reality.

The main rivals in the exoplanet race are a team led by Geoff Marcy of the University of California, Berkeley, and another led by Michel Mayor of the University of Geneva. Overall, the Americans have detected more planets, clocking them up at an impressive rate of three a month for the past year. But the Swiss have had some notable triumphs. Indeed, it was they who started the whole thing off 12 years ago. It was also they who announced the Earth-like planetmeaning, in this context, that they think it is both rocky and capable of harbouring liquid water.

If their interpretation is confirmed, it would be a real coup. For, as the meeting heard, bagging exoplanets is a game with a natural bias that tends to yield exactly the opposite of Earth-like planets.
Spatial prejudice

The reasons for this bias are varied, but they have a common underlying cause. Compared with the stars they orbit, exoplanets are faint and difficult to spot directly. Hence astronomers mostly infer their existence by looking for their effects.

The first type of bias is that the stars which reveal their planets most readily are smaller than the sun. The commonest way of finding a planet is to look for a star that wobbles. The most probable cause of such wobbles is that the star in question has a planet or two tugging at it. Such playground altercations are most readily noticed when a star is small. Though the sun is not a particularly large star, the wobble-detecting technique works best with stars that are even smaller.

The second source of bias is the opposite of the small-star effect. Large planets create more wobble than small ones do. By large, planet hunters mean not just bigger than the Earth, but bigger than Jupiter, which is 318 times more massive than Earth. That is very big indeed.

The third cause of bias is that easily detected planets tend to be close to their stars. This is partly because the closer they are, the more wobble they induce, and partly because, to be certain that they are looking at a new exoplanet, planet hunters want to see two complete orbits. A planet's orbital period depends on its distance from its star. Jupiter, for example, takes 12 years to orbit the sun and would thus require the best part of a quarter of a century to be detected to an astronomer's satisfaction.

There is also a fourth cause of bias, although it favours stars that are like the sun, at least in age (the sun is about 4.5 billion years old). Like a teenager, an adolescent starone that is a mere billion years or so oldis volatile. It spits fire that prevents astronomers from making observations of what is happening around it. Researchers therefore concentrate their efforts on seeing the offspring of calmer, more sedate stars that are that little bit older.

This explains why red dwarfs are popular terrain for planet hunters. They are smaller than the sun (typically weighing between a tenth and a half of its mass) and, because they burn at lower temperatures, they last a long time and are often very old. Unfortunately for those seeking little green men, their coolness means that their habitable zones (where the temperature is right for liquid water to exist) are much nearer to the star than the Earth is to the sun. And that, in turn, means planets in this region are likely to be tidally locked, as the moon is to the Earth, and show the same face to the star at all times. This makes it hard for a planet to develop an atmosphere.

The upshot is that almost all of the worlds found so far are hot, mammoth orbs that rapidly revolve around stars smaller and darker than the sun. They are thus quite unlike the planets of the solar systemand only one of those, namely Earth, is known to be hospitable to life.
Collector's items

Planetary taxonomy is straining to accommodate the new worlds. The distinction between different types of planet has already been broadened so that the solar system is now thought to contain three, rather than two, types. The innermostMercury, Venus, Earth and Marsare rocky. Then come the gas giants, Jupiter and Saturn. Uranus and Neptune are also gas giants, in that they have no visible solid surface, but they now tend to be classified as a subgroup of ice giants that contain water in its solid form. Small, icy objects such as Pluto are no longer regarded as planets, but in any case, they would be undetectable around another star. Most of the exoplanets detected to date are gas giants. Many are far more massive than Jupiter (see chart).

Then there is a question of whether particularly massive planets are really stars that failed to ignite. One school of thought says that if an exoplanet has a mass more than 13 times that of Jupiter, it is not a real planet, but rather a brown dwarf. That is because it is capable, in theory, of sustaining the nuclear fusion which would transform it into a star. If so, then some of the most massive exoplanets found to date might better be classified as failed stars.

One of the most interesting taxonomic discoveries so far was announced two weeks ago by Michael Gillon, a member of the Swiss team who works at Liège University in Belgium. Dr Gillon was looking at an exoplanet circling a red dwarf called Gliese 436. This had been found by the American team using the stellar-wobble method. However, Dr Gillon suspected that, viewed from Earth, this planet's orbit might cross the disc of its parent star, a phenomenon known as transiting. Transits of exoplanets are rare. Even among those that tightly circle their parent star, only 10% would be expected to transit in the Earth's line of sight. For those that orbit their stars at the distance of the Earth from the sun, that figure falls to less than 1%. Dr Gillon, however, was fortunate. The planet in question is closer to its star than Mercury is to the sunand it fell within the lucky 10%.

By measuring the resultant dip in Gliese 436's light, Dr Gillon was able to calculate the exoplanet's diameter. Knowing both its size and its mass, he was then able to calculate its density, and thus make a plausible guess about its composition. The orb in question would appear to be made, in part, of ice. That would classify it as an ice giant somewhat similar to Neptune. It probably has a rocky core surrounded by water and this water mayphysicists are unsure about the behaviour of water in such extreme conditionsbe in the form of hot ice. The pressure of being at the centre of a planet would, in other words, have compensated for the temperature caused by being so close to a star.

It is the rocky planets, though, that excite most public interest. The one that the Swiss team thinks it has discovered is in orbit around a star called Gliese 581, some 20 light years from Earth. Unfortunately, astronomers do not yet know if its orbit transits Gliese 581's disc, as viewed from Earth. This means it may be impossible to discover its composition in the way that Dr Gillon did with the planet of Gliese 436. The team think it is rocky because it weighs only five times as much as Earthtoo little for a gas giant. The exoplanet may also lie in the habitable zone of the red dwarf that it orbits.

Detecting more such rocky exoplanets is the aim of a European spacecraft called COROT that was launched in December 2006. It is searching for exoplanets that transit their parent stars, as seen from 900km above Earth. The mission's designers believe that the camera aboard COROT is sensitive enough to detect not merely transiting gas giants, but also rocky planets that are the size of Earth. The spacecraft is also probing the interiors of stars by studying starquakes on their surfacesin much the same way that geologists learn about the Earth's interior by studying earthquakes. Astroseismology, as this science is known, may allow astronomers to get a better idea of the ages of stars that are found to host exoplanets. Given that exoplanets are thought to be formed soon after the birth of a star (see article), knowing the age of the parent would date the offspring, too.

America's space agency, NASA, also has plans to launch a dedicated rocky-planet hunter. The Kepler mission is similar to COROT, but will be more sensitive. Moreover, it will be placed in an orbit better suited to locating its quarry. The European mission is hampered by its low orbit: the Earth, the moon and the sun periodically obscure its view. Kepler will not suffer that indignity, because it will be placed in orbit around the sun, trailing behind the Earth. If all goes well, the spacecraft should be launched early in 2009.

Whether the rocky exoplanets bagged by these missions are able to support life will depend on whether they have atmospheres. Detecting such ethereal breaths has only just come within the grasp of mankind. Earlier this year two space-based telescopes, Hubble and Spitzer, were both used to probe exoplanet atmospheres. Hubble saw how the upper atmosphere of a scorched gas giant closely orbiting its parent star was bleeding into space. Spitzer, meanwhile, found a surprising lack of water in the atmospheres of two other gas giants.

To examine an alien planet's atmosphere, the planet concerned must transit its parent star in the line of sight from Earth. Starlight contains the signatures of the chemical elements that compose a star. When an exoplanet passes in front of that star, its atmosphere adds further chemical signatures. By subtracting the signals from the star from those of the combined system, it is possible to identify which gases are present in the planet's atmosphere.

Without an atmosphere, all water on Earth would be frozen. Whether there could be life on the Swiss team's rocky planet depends very much on whether it has an atmosphere beneath which water could exist in its liquid form. Some calculations suggest this is unlikely, and that any water on the planet would be steam.
A rocky future

Two further missions are planned to search for signs of life in the new haul of rocky planets that is expected to be found in the coming months and years. America's Terrestrial Planet Finder and Europe's Darwin are still on the drawing board, but astronomers would like to use them to identify biomarkers such as oxygen in the atmospheres of exoplanets.

To inform this work, Lisa Kaltenegger of Harvard University and her colleagues have been assessing what the Earth's atmosphere would have looked like over the past 4.5 billion years. The idea is that planet hunters will be able to spot planets that have the potential to develop life, even if life does not yet exist there.

The researchers assume that 60% of the Earth's surface was shrouded by cloud at any given time in its history. Throughout that history, about 70% of the surface has been covered by oceans and 30% by land. Initially, the land was mostly rock and ice. This slowly altered to today's picture as vegetation emerged. Some 30% of the land is now covered by grass and another 30% by trees. Naked rock accounts for 18%, snow covers 15% and sand another 7%.

By tracing the way they think life developed on Earth against this background, Dr Kaltenegger and her colleagues have come up with a model of how the composition of the planet's atmosphere ebbed and flowed with the rise of various classes of organism. They reckon it has always contained a lot of nitrogen, but that early carbon dioxide was mopped up by the formation of limestone while methane levels rose as a result of the exhalations of a group of bugsthen common, now rarecalled methanogens. Eventually, photosynthesis arrived in the form of cyanobacteria. Their exhalations, oxygen, did for both methanogens and methane and created an atmosphere similar to the one that exists now.

The main upshot is that although oxygen is unquestionably a biomarker (it is so reactive that it would vanish from any atmosphere unless constantly replenished) methane may be one, too. Which is awkward, as methane is found in atmospheres, such as that of Saturn's moon Titan, where life is not thought to be present.

Another biomarker might be a planet's colour. Chlorophyll, the chemical that controls photosynthesis, is green for a reason. It is tuned to the spectrum of the sun in order to absorb sunlight in the most efficient way possible. The reflected green is the light it rejects. That calculation could be made for any star, and the best colour for a chlorophyll equivalent worked out. Perhaps if the exoplanet under investigation were orbiting a dwarf star, the little green menor, at least the vegetation they eatmight turn out to be orange.