IT’S not often that someone discovers an oddity new to science while testing some home-made gadget in a car park. But it can happen, as astronomers Tim Brown and David Charbonneau know. In 1999, while tuning a telescope whose optics Brown had cobbled together in his garage, the pair made the first sighting of a shadowy planet crossing the face of a bright star beyond the solar system.
The planet, known obscurely as HD 209458b, was not a new discovery, and nor was it the first planet found beyond the solar system – far from it. HD 209458b’s claim to fame is that it was the first planet seen to “transit” its star – in other words dimming the starlight a little as it passed across the front. By measuring the dimming, astronomers had their first chance to measure the exact mass and size of an alien planet, and even have a stab at its chemical make-up.
“Every now and then, nature tries to do you a favour,” says Brown of the National Center for Atmospheric Research in Boulder, Colorado. “Putting a planet between us and a star is a big favour. It makes a lot of things possible that you just wouldn’t have imagined.” Late in 2002, a second transiting planet popped up. Now dozens of projects are racing to spot the next one.
Advertisement
But professional astronomers are not the only ones searching for alien worlds. They have been joined by an army of dedicated amateurs who are well versed in discovering celestial objects such as comets and supernovae (see graphic). Although much of the follow-up work in probing a transiting planet’s atmosphere requires giant instruments like the Hubble Space Telescope, finding a transiting planet is possible with a mere pipsqueak, a telescope you could hold in the palm of your hand. “You can make progress with telescopes small enough to put in your car, or telescopes large enough to drive your car into. The whole range is involved,” says Brown.
Since the first discovery of a planet around another sun-like star in 1995, astronomers have discovered more than 100 of these extrasolar planets. None of them reflects enough light to be detected directly. Instead, they betray their presence in what are called radial velocity surveys. These pick up tiny periodic shifts in the wavelengths of a star’s light, which reveal that the star is wobbling due to the gravitational pull of planets circling it.
The period of the wobble tells astronomers how long a planet takes to orbit its star, while the wobble’s size is linked to the mass of the planet. However, deducing the exact mass isn’t possible without knowing the inclination of the planet’s orbit – how tilted the orbital plane is relative to the line of sight.
But if astronomers can glimpse a planet passing directly in front of its star, they then know for certain that the planet’s orbit is edge-on to the observer. This allows them to pin down its mass from the observed wobble of the star. What’s more, the amount of dimming reveals the planet’s diameter and, in principle, makes it possible to work out the chemical make-up of its atmosphere by analysing the starlight shining through it and looking for the characteristic spectral “fingerprints” of different elements.
When Brown and Charbonneau, who is based at the Harvard-Smithsonian Center for Astrophysics in Massachusetts, spotted a transiting planet for the first time, they were testing a system called STARE in the car park behind Brown’s lab in Boulder. The STARE project uses a telescope fitted with a scientific-grade digital camera and software to monitor some 24,000 stars in a small patch of sky, looking for starlight to dim when planets pass in front. What sets the STARE telescope apart from many other research instruments is that its mirror is a mere 10 centimetres across, and it cost just $50,000.
More impressive still, the transit measurements of HD 209458b have indeed allowed astronomers to indisputably compute the size and mass of an extrasolar planet for the first time. Surprisingly, they found that the planet is much bigger than Jupiter yet also much lighter, which is proving tricky to explain (see “Puffed-up in Pegasus”). Astronomers have also reported an energetic wind of hydrogen streaming out from the planet, dragging carbon and oxygen reluctantly along.
Big is not always best
Then in late 2002, another team announced a new transiting planet – the first extrasolar world to be discovered solely from a transit rather than in a radial velocity survey. The observations came from a project called OGLE, which been running at Las Campanas Observatory in Chile since 1992, and has surveyed thousands of sun-like stars in the constellation Sagittarius. The OGLE observations pinpointed 59 that seemed to have small, dim bodies passing in front of them. Follow-up observations by a team led by Dimitar Sasselov of the Harvard-Smithsonian Center for Astrophysics confirmed that one of these dimming events was a genuine planet. Now known as OGLE-TR-56b, the planet is roughly the size of Jupiter and is extremely distant, lying 5000 light years away towards the centre of the Milky Way.
Like HD 209458b, the OGLE planet is a little odd. It has the smallest orbit of any known extrasolar planet – its orbital radius is just 6 per cent of Mercury’s, and it races round its star once every 29 hours (żěè¶ĚĘÓƵ, 18 January 2003, p 20). Because all the previous 100 or so extrasolar planets have “years” lasting three days or more, astronomers had thought there must be a no-go zone surrounding a star, inside which a planet would either plummet into the star or be flung back out by the star’s magnetic field. If that exclusion zone really does exist, astronomers wonder why OGLE-TR-56b breaks the rules.
The OGLE planet is unlikely to give up any more secrets soon. The system is too distant and faint for today’s telescopes to glean any useful information about the planet’s atmosphere. Instead, astronomers are pinning their hopes on finding more transits around nearby, brighter stars – shadows of fat and thin planets, hot and cold planets, and ones with different orbits and chemical make-ups. “Hopefully then we can come up with the answers to all our questions,” says Tim Castellano, an astronomer at NASA Ames Research Center in California.
With that in mind, around two dozen transit projects have sprung up around the world. Many of them are copying the STARE team’s approach and opting for cheap-and-cheerful small telescopes combined with commercially available digital cameras and image-processing software. “I think it’s fascinating, this is amateur-scale stuff,” says Brown. “It’s the opposite of cosmology.”
The rationale is simple enough: if you want to find transiting planets keen to give up the secrets of their alien atmospheres, narrow your search to bright, nearby stars. Then, once you have found a transiting planet, you can enlist a heavyweight telescope to check it out and study the spectra of atoms in the planet’s atmosphere.
Cherry-picking the bright stars means that small telescopes are ideal for the hunt, monitoring thousands of stars all at once across a wide field of view. Telescopes with 5 or 10-centimetre mirrors are perfectly capable of picking up transits of Jupiter-sized planets around brighter stars. The transit should dim the light of the star by about 1 to 2 per cent, typically for a few hours.
The 10-centimetre STARE telescope, for instance, is now continuing its search for planetary transits from an observatory in Tenerife in the Canary Islands, a more suitable site than a car park. It has been joined by two similar telescopes: Sleuth at the Palomar Observatory in California, and PSST at the Lowell Observatory in Arizona.
Another project is Vulcan, which uses a 10-centimetre telescope at the Lick Observatory in California, which cost $500,000. Vulcan gathered data on about 100 nights last year, and astronomers are now trawling the results for possible transits. A sister project, Vulcan South, was installed in January at the South Pole, where the cold atmosphere minimises the air turbulence that makes stars twinkle and hampers observations. Vulcan South will switch on again this month as winter brings permanent darkness to the pole, allowing the telescope to watch the sky 24 hours a day.
Similar ventures include a British project called SuperWASP, which will start looking for transits with four 11-centimetre camera lenses installed on La Palma in the Canary Islands.
The smallest and cheapest project of all is the Kilodegree Extremely Little Telescope, or KELT for short. The irony of the name won’t be lost on astronomers currently designing a monster 30-metre telescope initially dubbed CELT, the California Extremely Large Telescope. KELT consists of a camera lens just 5 centimetres across with an off-the-shelf digital camera, with a total price tag of only $26,000.
“It’s nice to work with something hands-on that could more or less have been built in your garage,” says Andrew Gould of Ohio State University in Columbus, who leads the KELT team. “Some small telescopes are being discarded by research institutions, but there’s still a huge amount of interesting work that tiny telescopes can do.” Gould says the team hopes to begin observing with KELT at a site in Arizona this spring.
All these projects face a common enemy, however: false alarms. Stars usually orbit around each other in binary systems. If their orbit is oriented so that they very slightly eclipse one another along the line of sight, their combined light can periodically dip by 1 or 2 per cent and mimic a planetary transit. Even some triple star systems can conjure up the same effect. In fact, calculations by Brown suggest that telescopes will register 12 times as many false alarms as genuine planets.
However, careful follow-up work should weed out the false alarms. Because the stars in a multiple system are never precisely the same colour, the amount of periodic dimming will be different for the different wavelengths each star emits, whereas a transiting planet simply dims the light at all wavelengths.
The fact that transits are detectable with small telescopes means amateur astronomers around the world can observe them too. When astronomy enthusiast Arto Oksanen from Muurame in Finland heard about HD 209458b in 1999, he immediately twigged that his amateur group’s observatory should be able to see the transit too. “It was exciting news,” he says. “I thought this should be an easy task.”
Sure enough his group recorded a transit of HD 209458b in September 2000 using a telescope and camera worth $25,000 – a little less than KELT’s price tag. Professional astronomers confirmed their observations. Inspired by Oksanen’s achievement, Castellano and Greg Laughlin, an astronomer at the University of California in Santa Cruz, have set up a project called Transit Search to encourage other amateur groups to join the hunt.
Although amateurs do not have equipment and analysis software sophisticated enough to find planets that have not already shown up in radial velocity surveys, they could still be the first to discover if any of those known planets actually transit their stars. “It is just a matter of time,” says Oksanen. The Transit Search website directs enthusiasts to the stars known to have planets, and advises when a transit should occur, if indeed one is going to at all. At least a dozen amateur groups have already seen the transits of HD 209458b, according to Castellano.
And in some respects, amateurs even have the upper hand. Whereas a transit might occur in daytime for one professional observatory, having amateurs scattered all around the globe almost guarantees someone will be in the right place at the right time during the night. “Plus, amateur astronomers are not subject to the whims of the committees who allocate telescope time,” says Castellano. “They can look at one star for a whole month if they want to.”
What’s more, amateurs might have more patience for hunting planets that cross their star once a year or less. Such a transit would be rare: a planet in an orbit that resembles the Earth-sun system has only a 0.5 per cent chance of being correctly aligned for transits to come into view. Even worse, they would only happen once a year, and it would be hard to pin down their timing. “It really becomes somewhat daunting,” says Castellano. “I think that’s where the professional astronomers would tend to lose interest.”
Yet these long-period planets will also hold precious information about cool extrasolar planets akin to Jupiter or Saturn. “Their properties would not be as bizarre as those of the hot gas giants found so far, so you could make more sense of their chemistry,” says Brown. “Maybe you’d only get the chance to see a cool planet on one night every couple of years, but on that one night, a whole lot of telescopes would be pointing that way.” With Hubble or its successors, it should even be possible to spot large alien moons circling long-period planets.
For the moment, amateurs and professionals alike can only hope to glimpse transits of gas giant planets. But that could change soon. Next year should see the launch of a spacecraft called COROT, led by the French National Space Agency, which could spot the shadows of planets much smaller than gas giants, although still a few times as big as Earth. And in 2007, NASA plans to launch Kepler, a space telescope capable of finding hundreds of transiting planets roughly the size of Earth, even though they would dim their stars by only 0.01 per cent.
With hundreds of shadowy transits in the bag, astronomers will start to understand how diverse planetary systems can be. “We have no handle on it right now, but what everyone wants to know is this: is our solar system typical or is it weird?” says Brown. “Whether we’re oddballs or pretty normal, we should know in 10 years’ time.”
Puffed-Up in pegasus
THE first planet ever caught creeping across the face of a nearby star has proved just what a gold mine of information a transit can be. The planet, called HD 209458b, lies 150 light years away in the constellation Pegasus, and transits its sun-like star every 3.5 days.
Following reports that the planet’s gravitational pull seemed to be making the star wobble, veteran planet hunters Tim Brown and David Charbonneau first spotted the transits in 1999. And Greg Henry of Tennessee State University in Nashville and his colleagues noticed the transit independently around the same time.
Measurements suggest the planet’s mass is only 69 per cent of Jupiter’s, yet its radius is 35 per cent bigger. Why it is so bloated is unclear. The planet is closer to its star than Mercury is to the sun, so it bakes in scorching heat which could cause it to expand. But the star’s heat alone can’t account for the planet’s huge girth.
One possible explanation is that stormy weather and fierce winds carry heat into the depths of the gaseous planet, making it expand. Alternatively, the large size of the planet might be an illusion, because astronomers are confusing the surface of the planet for a layer of cloud. “These clouds would be made of weird stuff by terrestrial standards,” says Brown.
The transits have also allowed astronomers to detect elements in the atmosphere of an alien planet for the first time. In 2001, Brown and Charbonneau used the Hubble telescope to analyse the starlight that was shining through the atmosphere, and they discovered the spectral fingerprint of sodium.
Then last year, a team led by Alfred Vidal-Madjar of the Institute of Astrophysics in Paris reported that hydrogen is streaming out from the planet’s upper atmosphere like a comet’s tail (Nature, vol 422, p 143). This year, Vidal-Madjar’s team found signs of oxygen atoms and carbon ions pouring out as well. These last two elements are too heavy to evaporate, so Vidal-Madjar thinks the hydrogen wind is dragging them outwards like dust in a whirlwind. Vidal-Madjar estimates that the planet is losing up to a million tonnes of material into space every second.
Vidal-Madjar has christened HD 209458b Osiris, after the Egyptian god of the underworld. But it remains to be seen whether this title becomes official.
- Join the hunt for extrasolar planets at