WHEN YOU WISH UPON A STAR

Techniques used to seek out extrasolar planets could provide the key to finding alien forms of life. Norman Miller explains

The Earth is invisible. At least, it would be if a twin alien civilisation were looking at our solar system from theirs. Looking outwards, of course, poses similar difficulties for us, and until 1995 astronomers were getting increasingly frustrated as the hunt for planets in other star systems continued to prove fruitless despite surveys that had been going on for over 10 years.

Then in October 1995, the first needle appeared in the cosmic haystack. The Geneva Observatory announced that Michel Mayor and Didier Queloz had discovered a world orbiting 51 Pegasi, 51 light years away. The Swiss had given us the first extrasolar planet (abbreviated by some to exoplanet). Ironically when Mayor and Queloz began monitoring 142 Sun-like stars in 1994 they weren't looking for planets. Their quarry had been objects called "brown dwarfs", which the theorists believe form from collapsing clouds of gas and dust in the same way as stars but never become massive enough to initiate hydrogen fusion and shine.

Their search technique involved looking for the tiny change in the radial velocity of a star which would occur if an unseen companion was present, causing the star itself to orbit around the gravitational centre of mass of the system as a whole. The presence of a planetary companion will show as a variation in the so called Doppler shift of the parent star so tiny that it takes hours of computing time to separate the planet's signal from that of the main star. This technique provides the orbital period of the mystery object and a lower limit for its mass.

When Mayor and Queloz looked at 51 Pegasi they found precisely the right kind of frequency shifts but caused by something half the mass of Jupiter - far too small to be a brown dwarf but fine for a planet. Though this was good enough for the duo to treat themselves to champagne, it was not enough of a clincher in itself to go public. Before accepting cheers from their peers they had to rule out other possible causes for the signal from 51 Pegasi.

Rival possibilities such as pulsations in the star or giant sunspots were all ruled out, leaving one remaining worry - the closeness of the planet to its parent star. At just 0.05 astronomical units - 1AU is the distance from the Sun to the Earth - this meant that a giant planet half the mass of Jupiter was orbiting 51 Pegasi much closer than even Mercury is to our Sun. Compared with Jupiter's 12-year orbit, exoplanet candidate number one whipped round its star in just four days. Could a giant planet really exist so close to its parent star or would it have been turned into an unstable space cinder long ago? To resolve the situation, Mayor put a basic question to several theoreticians. How close could Jupiter be to the Sun before it became unstable? Nobody knew. This was not just a new planet, this was new planetary theory.

The word went out on Mayor's poser and the astronomical grapevine quickly brought it to Adam Burrows at the University of Arizona. Twenty-four hours and many computer simulations later, he had an answer. Anything further than 0. 04AU from the star would be stable. At 0.05AU, the Swiss team's planet could go into the record books.

News of the first exoplanet prompted an incredible range of responses, from an American six-year-old who e-mailed Mayor to ask if he'd visited his new planet yet, to a Roman Catholic priest who went on the record about his concerns for the souls of any of its possible inhabitants.

The discovery also prompted some soul searching from one of Mayor's fellow astronomers, San Francisco State University's Geoff Marcy. A long- time planet-hunter, Marcy and his colleague Paul Butler have spent seven years monitoring 60 stars and had just begun gathering data from another 60. Like most astronomers, however, they had assumed that alien solar systems would look like our own, with planets large enough to detect lying far from the parent star and therefore taking years to complete a single orbit.

Expecting to have to do a lot of stargazing before finding anything, they hadn't even begun to analyse their data. Soul-searching quickly switched to data-searching, and there on their computer hard disk were planets just waiting to be discovered.

Determined to at least get the silver and bronze medals in the exoplanet hunt, they announced two new planets in January 1996 - one orbiting 70 Virginis (78 light-years away), the other orbiting 47 Utsaa Rajoris (44 light-years away), lying around 0.5 and 2.1AU from their respective stars and with masses of 6.6MJ (Jupiter masses) and 2.4MJ respectively. Marcy and Butler then completed their hat-trick in April by pinpointing an 0.8MJ planet 45 light-years away around 55 Cancri, though in the interim several astronomers claimed another Jupiter-mass planet 5AU out in the dusty disc surrounding the nearby star Beta Pectoris - long suspected of harbouring a planetary companion. This rush of discoveries lay as much in advances in detection techniques as the sudden realisation that alien systems might not look quite how astronomers had imagined.

The search for exoplanets relies not on observing them directly but rather by spotting the minute effects they have on the behaviour of their parent star, as observed through the light we see coming from it. The key measurements focus on the star's velocity, position and brightness.

The first triumph with 51 Pegasi was achieved using the radial velocity method, which allows astronomers to detect the presence of more than one planet around a star by working through the oscillations in the signal. This technique has, in fact, detected several planets around distant pulsars.

Pulsars are like cosmic lighthouses, sending out massive beams of radiation with a very precise frequency. Tiny variations in this frequency are evidence of planets. Several planets have now been found around three pulsars but even though these are multi-planet systems - PSR1257+12, for example, has four known planets, two of them on a par with the Earth in terms of mass - none are candidates for life owing to the intense X-rays which their pulsar parents rain down on them.

Another technique involves precise measurement of changes in a star's position in the sky with respect to much more distant stars (so far away they are thought of as "fixed") - the astrometric method. A periodic displacement of the star about a centre of mass again indicates an unseen companion, and astrometry also allows a precise measurement of the mass of the unseen companion to be made.

It was with this technique that the first discovery was made of a normal star (as opposed to a pulsar) with more than one planetary companion - Lalande 21185. This eight-light-year cosmic neighbour of the Sun had long been the subject of exoplanet speculation, along with another near neighbour, Barnard's Star. Early claims of an exoplanet around Barnard's Star were shot down by US astronomer George Gatewood at the University of Pitts- burg, but the whistleblower turned cheerleader this year when Gatewood claimed two Jupiter-mass companions orbiting about 2.5 and 10AU from Lal 21185 - a discovery which, if confirmed, raises hopes that solar systems similar to our own do exist.

A third method of exoplanet hunting centres on photometry. If a planet is orbiting a star with its orbital plane edge on to the line of sight then the planet will periodically pass in front of the star, reducing its light by a tiny amount. Even an Earth-sized planet orbiting at 1AU in an identical twin system to ours might reduce its parent star's brightness sufficiently to be detected.

A related technique is gravitational micro-lensing. When a star passes in front of another star in our line of sight, Einstein predicted that the light from the more distant star would be concentrated by the gravity of the nearer star. The distant star would appear to brighten and then fade as the nearer star passed across the line of sight. Einstein's prediction was vindicated a few years ago, and in theory provides another method of exoplanetary detection.

But though the approach has been used to detect otherwise invisible single and faint binary stars, the effect has not yet come up with an exoplanet.

But what of the exoplanets themselves? Astronomers believe this growing family of stellar companions fall into three categories: brown dwarfs, gas giants and terrestrial planets. The last two are familiar because they are present in our own system, with Jupiter being the premier gas giant and Earth being home. Despite the name, gas giants share the property with terrestrial (rocky) planets of having a solid core, in contrast to brown dwarfs which are gas through and through. Beyond that, however, the criteria for distinguishing between planets, "super-planets" and brown dwarfs is an astronomical battleground.

Jean Schneider, compiler of an Extrasolar Planets Encyclopedia on the internet, puts a cut-off for planets at 13MJ, the mass at which an object can in theory begin fusion reactions. The number of confirmed exoplanets now known around normal (main sequence) stars using this definition stands at eight, with Marcy and Butler announcing the most recent discovery, around 16 Cygni, two months ago.

Before we can ask whether any natives might be friendly, how friendly are the planets themselves? The first discovered, 52 Pegasi B (the B refers to the fact that it is the first planet out from its parent star, which is designated A) wouldn't be a good start, since the tight huddle it has with its parent gives it an estimated surface temperature of 1000C. 70 Virginis B is another matter since Marcy and Butler put its temperature at a very reasonable 80C - hot for a holiday but acceptable if you're a water molecule. Marcy has even speculated on complex molecules floating in its atmosphere, though other astronomers are more sceptical. "There's nothing wrong with dreaming, but let's try to keep cool," one doubter said.

The debate about the nature of individual exoplanets extends to what they imply for theories of planet formation. The accepted credo is that stars form from the collapse of dense clouds of gas and dust while planets are believed to form in the flat circumstellar disc of dust and rubble leftover from the formation of the parent star, with individual planets forming as rocks and dust cluster together and form clumps of material.

Once these clumps are about a kilometre in diameter, their gravitational pull gathers more and more particles, building up into planets - the number and size of planets depending on the amount and distribution of material in the circumstellar disk. This theory has been tested in the lab by Torsten Poppe at the University of Jena in Germany, where a clumping process took place. Further experiments to recreate the earliest stages of planetary formation in the weightlessness of space are planned for a Space Shuttle mission later this year, which should reveal how fast particles clump together and the shape of these planetary seeds.

Studies suggest that around half of young (T Tauri) Sun-type stars show evidence of protoplanetary discs, with the figure rising towards a whopping 80 per cent for young stars in dense clusters. Since the conditions for planetary-system formation appear to be common, there seems every likelihood that solar systems like our own are also common. However, the systems so far observed have thrown the experts into confusion, since the number of massive planets existing so close to their parent stars flies in the face of current theory that such planets can only form a good distance away from their parent.

One suggestion is that these massive planets did form several AU further out but then migrated inwards under the influence of gravitational interaction with other matter in the planetary disk. Astro-theorist Alan Boss, meanwhile, has revived a previously discarded idea from the Fifties which allows massive planets to form in one step via gravitational collapse from a dust cloud rather than the currently favoured two-step process. This is where an icy or rocky core forms first, then gathers a vast gas atmosphere around it over a long period. This latter method requires a healthy distance from the parent star to allow the necessary timescale, but Boss's theory would allow the much closer arrangement observed in several exoplanetary systems.

What of the biggest question of all - the possibility of life? For that, Europe's exoplanet astronomers have turned to Project Darwin, while Nasa has gone for a more prosaically named project, Planet Finder.

Darwin is a new avenue in the ongoing search for life in the universe. Until now, projects such as Seti have scanned the cosmos to detect signals which only an advanced civilisation would produce - like the broadcasts of I Love Lucy that are by now about 45 light-years from Earth and head- ing outwards towards any ET with a TV. Darwin, by contrast, has a less ambitious but more practicable aim of looking for simple forms of life such as algae by searching for the key chemical they produce - oxygen.

The hardware involved in both the US and European projects centres on a special type of space-based telescope - a 100m-wide interferometer - which will be 50 times more powerful than the Hubble Space Telescope. With this, says Darwin's UK coordinator Alan Penny, "we'll not only be able to detect Earth-mass planets around nearby stars, but take spectra of their atmospheres and look for any oxygen. Oxygen can only be produced by life". Don't hold your breath, however - tentative launch dates are set for about 2010.

But while Darwin could provide the firmest evidence that alien life exists, the mere discovery of exoplanets fills in a key element in the equation - more spe-cifically, in Drake's Equation. Devised by American astronomer Frank Drake, this puts a figure on the likely number of advanced civilisations it our galaxy. Its terms involve the rate of star formation, the fraction of stars that form planets, the number of planets hospitable to life, the fraction of planets on which life evolves into intelligent beings, the fraction of those capable of interstellar communication, and the time such a civilisation remains detectable.

The growing list of exoplanets is providing a clearer idea of the fraction of stars with planets. And now we know our solar system is not alone. !