Special Report: Alien Earth

    With improved techniques, growing data sets and a new space mission, 2007 is the first year in which we might discover another planet like our own. Katharine Sanderson reports.

    When marvelling at the stars on a clear night, it's hard to imagine that there are up to 400 billion of them in our Galaxy alone. Even harder to comprehend is how many planets may be orbiting these stars — a number that could run into trillions. Surely somewhere among them there must be a comfortable home for alien life, even if it's not advanced enough to be gazing back at us?

    This is the question that exoplanet hunters are trying to answer. So far, they have spotted 209 planets beyond our Solar System. These tend to be gas giants in searingly hot orbits close to their parent stars — unlikely to be habitable. But researchers are edging closer to finding the one type of planet that we know can support life — a carbon copy of our own Earth. Thanks to improved techniques, mounting data and a new space mission, many believe that 2007 could be the year we find the first truly Earth-like planet. At the very least, we should have a much better idea of how common alien Earths may be.

    The main obstacle for planet hunters is that planets outside the Solar System are obscured by the light from their stars, so our telescopes can't see them directly. Most researchers make use of the fact that when a planet orbits a star, its gravitational pull causes the star to wobble slightly. As the star wobbles, its speed as seen from Earth (its radial velocity, or RV) changes, and this shows as a change in the wavelength of the star's light. This can be used to estimate a lower limit for the planet's mass.

    Unfortunately, RV tends to detect big planets that are close to their stars. Heavier planets cause more obvious wobbles. And close planets have shorter orbits, allowing researchers to observe several wobbles over a relatively short time.

    So far, the method has found 197 exoplanets, the smallest of which is at least 7.4 times the mass of Earth. But improvements in accuracy are allowing researchers to spot ever smaller planets, and as more observations are made, it's possible to detect planets that are farther from their stars. Geoffrey Marcy of the University of California, Berkeley, is responsible for an impressive 121 of the planets found using RV, and says his team is set to announce several exciting discoveries in 2007. He adds that he expects RV to reveal several rocky planets this year, and that within the next few years we may know whether Earth-like planets are common or rare in the Universe.

    Transit tricks

    Another limitation of RV is that, on its own, it doesn't reveal what a planet is made of. But coupling its mass with an estimate of the planet's radius — which can be worked out as it passes in front of (transits) its star — gives the density, which indicates whether a planet is a diffuse gas giant or a smaller rock. The main problem with this is that to see a transiting planet you need to look at the correct angle at the correct time, and with most exoplanets found we haven't been so lucky. Transits are most common for large planets close to their stars, which again skews our knowledge away from planets like Earth.

    In 2007, however, a European mission called COROT aims to improve the odds. Launched on 27 December, it will look specifically for transiting planets, which will then be followed up with RV measurements, either from Earth or from NASA's Spitzer Space Telescope.

    After a couple of months of calibration, COROT will make observations for 150 days, then turn 180° and observe for another 150 days. To obtain reliable data, scientists need to watch three or four transits, so the craft's short observing period limits the search to planets close to their stars. This means COROT is unlikely to find a true Earth analogue. But it will tell us a lot about how common large, rocky planets are. “It's a necessary proof,” says Malcolm Fridlund, project scientist on COROT. Fridlund is proud that Europe is leading the field rather than NASA. “We're taking the first step for once,” he says.

    Not alone? The search for Earth-like planets beyond our Solar System is hotting up. Credit: NASA

    “If rocky planets are common enough, COROT could find planets that are habitable,” says David Charbonneau, an astronomer at the Harvard-Smithsonian Center for Astrophysics in Cambridge, Massachusetts.

    “COROT will find oodles of short-period planets, both hot Jupiters and hot super-Earths.”

    The first results from COROT should be in and analysed by mid-2007, with ground-based follow-up experiments done by the end of the year. “COROT will find oodles of short-period planets, both hot Jupiters and hot super-Earths,” says Alan Boss, a theoretician at the Carnegie Institution in Washington DC. But some see COROT simply as a stats-gathering prelude to NASA's more ambitious Kepler mission, which is due to launch late in 2008 and will make observations over four years. “Kepler holds even greater promise to detect rocky planets around Sun-like stars,” says Marcy.

    For 2007, the best chance of spotting an Earth-like planet may come from another ground-based technique: gravitational microlensing. When two stars are closely aligned along our line of sight, the front star acts as a lens and magnifies light from the star behind, sometimes enough for an orbiting planet to be seen directly. Whereas RV is most sensitive to planets close to their stars, microlensing works best for distantly orbiting planets. Four planets have so far been detected in this way, including an icy 'super-Earth' (J.-P. Beaulieu et al. Nature 439, 437–440; 2006).

    A star undergoes a microlensing event as seen from Earth only once every 100,000 years, so spotting them is “pretty hairy” says Andrew Gould of Ohio State University in Columbus. Gould coordinates a team of astronomers around the globe (the MicroFUN group) who monitor changes in the brightness of around 50 million stars. They post a list of possible microlensing sites online, and Gould alerts astronomers in the appropriate locations where and when to look.

    The technique has bagged the lowest-mass planet so far, at 5.5 Earth masses. Gould predicts a one in ten chance of spotting an Earth-size planet in 2007 using microlensing, and reckons that in five years' time, he may be able to spot ten per year.

    The COROT spacecraft will spend most of this year looking for rocky planets that might host life. Credit: D.DUCROS/CNES

    If an alien Earth is found, the big question of whether it could support life will require information about its atmosphere. And researchers are set to make big progress here too this year. It's possible to scan a distant planet's atmosphere by taking the spectrum of the light emitted by its star when the planet is completely hidden from view, and subtracting it from the combined spectrum when the planet is in front. The technique was used to investigate the spectra of Pluto and Charon in the 1980s, and in 2001, Charbonneau and his team used the Hubble Space Telescope to detect sodium in the atmosphere of a Jupiter-class planet orbiting the star HD 209458.

    Last month, Jeremy Richardson of NASA's Goddard Space Flight Center in Greenbelt, Maryland, and his colleagues reported at the American Geophysical Union meeting in San Francisco that they have now used the Spitzer Space Telescope to make a more detailed study of the infrared spectrum of the same exoplanet. They measured the radiation emitted by the planet at various wavelengths, and are hopeful that they will be able to glean some information about the composition of its atmosphere. The team doesn't want to reveal details ahead of publication early this year. But co-author Sara Seager says that although these first spectral measurements are “pretty crummy”, the work is crucial proof that the technique will “provide the first-ever spectrum of an exoplanet. This is a big breakthrough in exoplanet science.”

    This will be a big year for Spitzer, adds Charbonneau. He predicts that the telescope will study ten planets in detail in 2007, making temperature measurements and detecting gases such as methane and carbon monoxide.

    For the future

    Looking beyond 2007, “Kepler is the big one”, says Charbonneau. Kepler's goal is to find an Earth-like planet. Like COROT, it will seek out transits, but will have more time to spot planets farther out from their stars.

    After that, prospects for the field are less rosy. NASA recently cut funding for its proposed Space Interferometry Mission (SIM), which is now seriously delayed. SIM would make very precise measurements of stars' wobbles in two planes, which would give an actual mass, rather than the lower limits provided by RV.

    The decision to cut funding has not gone down well, with Gould citing “stupidity” as one reason for the cut. Boss says the situation is “depressing” and a backward step for exoplanetary research. He warns that even a slight dip in funding could cause well-trained scientists to leave the field. Two other projects, the European Space Agency's Darwin and NASA's Terrestrial Planet Finder, would take the first images of Earth-like planets. But these are also receiving too little funding to proceed.

    Should planet hunters be allocated scarce funds ahead of other projects? Scott Gaudi, who works with Gould, argues that the search for other worlds is a societal imperative: the question of whether other Earths exist is one that humans have always asked, he says. Despite the funding problems, he remains upbeat: “There is a very good chance that in our lifetime we will answer the age-old question: is there life out there?”

    Related links

    Related links

    Related links in Nature Research

    Planet hunt ready for lift-off

    News in brief

    Found: one Earth-like planet

    Rocky planet found outside Solar System

    Neptune-sized planets spied around distant stars

    Extrasolar planets

    Related external links

    CNES's COROT site

    ESA's COROT site

    The extrasolar planets encylopaedia

    Rights and permissions

    Reprints and Permissions

    About this article

    Cite this article

    Special Report: Alien Earth. Nature 445, 10–11 (2007). https://doi.org/10.1038/445010a

    Download citation

    Further reading


    By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.