WANTED: Rocky planet outside of our solar system. Must not be too hot or too cold, but just the right temperature to support life.

It sounds like a simple enough wish list, but finding a planet that fulfils all of these criteria has kept astronomers busy for decades. Until recently, it meant finding a planet in the "Goldilocks zone" - orbiting its star at just the right distance to keep surface water liquid rather than being boiled off or frozen solid.

Now, though, it's becoming increasingly clear that the question of what makes a planet habitable is not as simple as finding it in just the right spot. Many other factors, including a planet's mass, atmosphere, composition and the way it orbits its nearest star, can all influence whether it can sustain liquid water, an essential ingredient for life as we know it. As astronomers explore newly discovered planets and create computer simulations of virtual worlds, they are discovering that water, and life, might exist on all manner of weird worlds where conditions are very different from those on Earth. And that means there could be vastly more habitable planets out there than we thought possible. "It's like science fiction, only better," says Raymond Pierrehumbert, a climate scientist at the University of Chicago, who studies planets inside and outside of our solar system.

Distance from the nearest star is, of course, important. In our own solar system, Venus has long served as an example of what can happen if a planet gets too close to its star. Venus is only 28 per cent closer to the sun than Earth is, but its surface is a sweltering 460 °C, hot enough to melt lead, and it chokes under a thick carbon dioxide atmosphere 90 times the density of Earth's.

Put Earth where Venus is and it would probably end up looking rather similar. The extra solar radiation would increase evaporation from the oceans, boosting the amount of water vapour in the atmosphere. As water vapour is a greenhouse gas, this increase would set off a vicious cycle, with higher temperatures triggering more evaporation, until the planet's surface was hot enough to boil away the oceans. At the other extreme, water on a planet that is too far from its star will simply freeze, like on Mars.

However, in 1993 a study by James Kasting of Pennsylvania State University, University Park, demonstrated that even in our own solar system, the habitable zone is not based on distance alone. In a calculation based onthe sun's current brightness,

Kasting worked out that while moving Earth just 5 per cent closer to the sun would doom it to the same fate as Venus, it could move almost 1.7 times its current distance from the sun before it would freeze ( Icarus, vol 101, p 108). This outer limit is interesting because it is beyond the orbit of Mars, whose orbit has a radius about 1.5 times that of Earth.

So if Mars is in our solar system's Goldilocks zone, why isn't it teeming with life? The answer lies in how a planet's mass affects its ability to hold on to a habitable atmosphere. On Earth, the carbon cycle works as a kind of thermostat that keeps the climate liveable. Volcanic activity releases CO₂, which warms the Earth's surface via the greenhouse effect, increasing evaporation and rain. The rain erodes carbon-containing minerals from rocks, washing them into the sea. Eventually, these minerals are pulled deep into the Earth in subduction zones.

This balance between emitting and sequestering CO₂ has helped keep the Earth's climate stable for the past 4 billion years. Mars, though, is only half the size of Earth, so its interior cooled quickly, shutting down the volcanic activity needed to supply CO₂ to the atmosphere. Its weaker gravity also allows its atmosphere to drift away into space. As a result, there is too little CO₂ in the Martian atmosphere to warm its surface enough to sustain liquid water. This has probably been the case for much of the past few billion years.

Goldilocks not required

Mass, however, is not the only factor. In a series of computer simulations published earlier this year, David Spiegel of Princeton University explored whether factors such as a planet's spin axis or speed of rotation could allow a planet outside of the habitable zone to hold onto liquid water long enough to sustain life ( The Astrophysical Journal, vol 681, p 1609). "I've been kind of twisting the knobs so that they're different from Earth, but they all have the same mass as Earth," says Spiegel, who was at Columbia University in New York when he carried out the work.

In some simulations, the team altered the tilt of the planet's spin axis. Earth's axis is tilted 23.5 degrees relative to the plane of its orbit, which is why each hemisphere has longer periods of sunlight during summer and shorter ones during winter. When they gave planets a tilt of 90 degrees, similar to that of the gas giant Uranus in our own solar system, the much larger variations in illumination led to more extreme seasons.

When this large axial tilt was combined with a rate of rotation three times Earth's, the summers became warm enough for ice to temporarily melt around the pole facing the star (see diagram). This meltwater was only sustainable when the planet rotated faster than the Earth, as the centrifugal force created made it harder for air to flow from the poles to the equator. This trapped heat at the illuminated pole.

Spiegel argues that this kind of simulation shows that astronomers should not think of habitability as an all-or-nothing thing. It makes more sense to think in terms of "fractional habitability", he says, as in what fraction of a planet's surface is habitable, for what fraction of the year, or for what fraction of its history. "Even the Earth is not 100 per cent habitable, at least by the standard liquid-water definition," Spiegel points out. "Parts of the planet are frozen part of the time. Parts of the planet are frozen all of the time."

Spiegel also created a desert world which was partially habitable. The planet was 90 per cent land, with just 10 per cent of its surface covered by liquid water. By earlier standards, the only part of this planet that would be considered habitable is a narrow zone around the equator where liquid water can exist all the time. Elsewhere, seasonal extremes would make water alternately boil and freeze at different times of the year, with liquid water present only in the spring and autumn. But Spiegel's team suggests that even these zones should not be ruled out as uninhabitable. They point out that there are microbes on Earth that can reproduce below 0 °C and others that can do so above 100 °C. None are known to be capable of both, as far as the team is aware, but that doesn't mean it is impossible.

What's more, life-giving heat need not necessarily come from the nearest star. This year, a team led by Brian Jackson of the University of Arizona in Tucson explored the extent to which some planets have internal heat sources. Planets orbiting close to a star or with non-circular, eccentric orbits move towards and away from their star in the course of an orbit. As a result, they are stretched and squeezed by variations in the gravitational pull from their star, and this causes enough friction in their interiors to generate heat.

Jackson's team calculated the amount of heat generated by this process of "tidal heating" for virtual rocky planets in a variety of orbits, focusing on planets in close orbits around red dwarf stars. These are the most abundant type of star in the universe, but they do not give out much in the way of heat.

While the amount of tidal heating varies depending on the mass of the star and planet, the team calculated that, given a large enough variation in gravitational pull around the orbit, this additional heat from below could be enough to thaw out frozen planets orbiting a red dwarf, despite the feeble radiation they receive from their host stars. The extra heat could also stimulate volcanic activity, even on planets with a low mass, potentially giving them thicker atmospheres and the stronger greenhouse effect needed to maintain liquid water beyond the Goldilocks zone.

Planetary mysteries

There is such a thing as too much of a good thing, though. Tidal heating is strongest for planets in the closest, most eccentric orbits. Some of these would receive tidal heat at an even greater rate than Jupiter's moon Io, which, because of tidal heating from variations in Jupiter's gravitational pull, erupts vigorously enough to completely remake its surface every 150 years or so. The massive amounts of volcanic activity that result from such intense heating could make it impossible to sustain life on these planets, the team say in a paper to appear in Monthly Notices of the Royal Astronomical Society.

With more than 300 planets already discovered outside our solar system, and many more sure to be found, these kinds of insights will help researchers to reconsider planets once dismissed as inhospitable, and focus on the highest priority targets for follow-up observations.

There is much still to discover, of course. For example, clouds can complicate the picture as they cut the amount of stellar radiation reaching a planet's surface, while also trapping infrared radiation emitted by the planet. Factoring in these different effects will be "absolutely central" to better understanding the climates of planets outside our solar system, says Pierrehumbert.

Despite this broadening of the criteria for potentially habitable planets, not everyone is convinced that these new insights are particularly helpful in the search for worlds that might support life. There is a lot left to figure out, even for Earth, says Jonathan Lunine of the University of Arizona. "I don't think we really understand how or why the Earth has been habitable in its history and what the excursions from habitability really were," he says, "and until we do, it's hard to be anything but sceptical that some of these models are really going to inform the search."

Plus, Lunine says, we are still grappling with puzzles over climates in our own solar system, regardless of the relative wealth of data we have on them. "We still don't really understand whether Mars had a globally habitable environment for any significant amount of geological time, and if it did, how early in its history did that climate come to an end and why," he adds.

There is always the chance that the search for liquid water on the surface may be missing the point. What if exotic forms of life could thrive where there is no liquid water at all - swimming around in lakes of liquid methane on Saturn's frigid moon, Titan, for example? "One should not rule out the notion that a kind of life or organised chemistry could exist in that kind of liquid," says Lunine. "Let's cast the net broadly."

• From issue 2683 of New Scientist magazine, page 36-39. Subscribe and get 4 free issues.
• Browse past issues of New Scientist magazine

If you would like to reuse any content from New Scientist, either in print or online, please contact the syndication department first for permission. New Scientist does not own rights to photos, but there are a variety of licensing options available for use of articles and graphics we own the copyright to.