(Free Press Release)

By John Timmer. Although our catalog of exoplanets is expanding rapidly, researchers are still looking for one that can unequivocally play host to liquid water on its surface. Over the past several years, attention has focused on the collection of planets orbiting the red dwarf Gliese 581, which is only 20 light years from Earth. The star plays host to at least five planets, some of which have been put forth as candidates for habitability based on the presence of liquid water. A new climate model, however, has now shifted attention towards GJ581d, a super-Earth that was thought to be too cold to support liquid water.
The history of the Gliese system is enough to make anyone cautious about the new announcement. Back in 2007, GJ581c was discovered orbiting right at the inner edge of the habitable zone, and was estimated to have a temperature that could support water. Further refinements, however, indicated that the presence of any greenhouse gasses would quickly boost the temperatures there to the point where the water would boil off.
Last year came the announcement of GJ581g, squarely in the heart of the habitable zone and only a few times heavier than the Earth. Other research groups with observational data on Gliese 581, however, can't seem to detect any indication of an additional planet, leaving GJ581g in scientific limbo until enough additional data comes in.
While waiting for all of that to be sorted out, a research team has gone back and taken another look at GJ581c's ugly step-sister, GJ581d. GJ581d is a heavy super-Earth orbiting at the far edge of the habitable zone, and had been thought to be too cold to support liquid water. And, unlike GJ581c, GJ581d was thought to be incapable of supporting an atmosphere that could produce enough of a greenhouse effect to warm things up.
Although it's at the far edge of Gliese 581's habitable zone, that's actually fairly close to the host star, a dim red dwarf. Close enough that the planet is thought to be tidally locked to the star, meaning its rotation and orbit are synchronized such that only one side of the planet ever faces the star (the tilt of its axis of rotation should also be minimal). This ensures that the planet's poles and far side are extremely cold, which creates a problem similar to the one that occurs on Mars: below a certain temperature, things like carbon dioxide start freezing out of the atmosphere, leading to its ultimate collapse. And with no atmosphere, there's not much chance for a greenhouse effect.
Previous attempts to understand the planet's atmosphere relied on an extremely simplified model. The new work, which has been accepted for publication in Astrophysical Journal Letters, marks a major improvement over those, as the authors have created a three-dimensional, general circulation model (GCM) that can work for a range of planets. Starting with a GCM that was developed for studying Mars, the authors pulled out all the Mars-specific features and replaced them with adjustable parameters. Thus, things like the planet's mass and orbit, as well as the radiation produced by the host star, could be adjusted to match those of the Gliese system. Things like the atmosphere's composition can be changed to try different levels of greenhouse warming, and the surface could be switched from a rocky composition to a planet-wide ocean.
Plugging in the values for GJ581d dictated by astronomical observations, the authors starting testing out atmospheres with different compositions and densities. With a rocky surface and CO2 rich atmosphere at pressures below about 10 bar, the atmosphere was unstable, and would begin to condense out at the poles and on the planet's dark side. But things changed when the atmosphere got thicker. "For denser atmospheres," the authors state, "we found that horizontal heat transport and greenhouse warming became effective enough to remove the threat of collapse and allow surface temperatures above the melting point of water."
Something similar happened on a watery world. At 20 bar and up, the strong greenhouse effect produced by water vapor was sufficient to raise the temperatures enough that the whole planet was above the point where the atmosphere would collapse, and liquid water could persist on GJ581d's sunny side. As long as the atmosphere was dense enough, there was no danger of a global glaciation of the sort that has occurred on Earth. Below that density, local pockets of melting could occur on the day side of the planet, but the dark side would get so cold that even nitrogen would condense out of it, and any water vapor would quickly freeze back out.
What are the chances that GJ581d actually has a dense atmosphere? It's really difficult to judge. Stars like Gliese 581 have an early period in which they emit a lot of extreme UV light and ion fluxes, which could be sufficient to blast any atmosphere off GJ581d. Geological processes might later restore one, but it's difficult to guess what that atmosphere might look like. A hydrogen-/helium-rich atmosphere wouldn't generate a greenhouse effect, and would thus leave the planet cold. Water and CO2 are certainly possibilities, but the quantity involved would have to be sufficient to bring things above the critical 10 bar value.
Fortunately, the authors note that Gliese 581 is close enough that we may be able to observe it directly with a future space-based observatory. For when that happy day arrives, they've used their model to produce some possible emissions spectra that should provide us with an indication of what's going on in the planet's atmosphere. If we ever get to the point where we can image the planet, we should be able to tell whether there's liquid water there. Until then, however, astronomers will no doubt keep looking for an exoplanet where the case for habitability is less ambiguous.
Astrophysical Journal Letters, 2011. DOI: not yet available; the paper is available through the arXiv.