Sometimes I imagine an ancient place where a dim sun hangs unmoving at zenith, and a race of philosophers and poets works out life’s verities under an unchanging sky. Could a place like this, on a terrestrial world orbiting an M-class red dwarf, really exist? A new paper by Jack Lissauer (NASA Ames) casts doubt on the idea. Lissauer argues that planets inside an M dwarf’s habitable zone are probably lacking in water and other volatiles, and are thus unable to produce life as we know it.

The question is important because M dwarfs make up as much as 75 percent of the stars in our part of the galaxy. If we include them as candidates for life, we add a hundred billion or more potential habitats in the Milky Way alone. We’ve known for some time that although the proximity of such a terrestrial M dwarf planet to its star would cause it to be tidally locked — one side in constant light, the other in darkness — habitable regions might still occur on the dayside given a dense enough atmosphere to transport heat globally.

But ponder the water question. Remarkably, the Earth itself is volatile-poor, with oceans and other reservoirs of near-surface water accounting for less than 0.03 percent of our planet’s mass. Lissauer believes that most of Earth’s water probably came from planetesimals that originally condensed beyond 2.5 AU. Our planet needed, in other words, help from farther out in the Solar System, and it can be shown that primitive meteorites from the outer regions of the asteroid belt have over 100 times as much water as those closer in.

But if our terrestrial world relied on such sources, could an M dwarf planet have done the same? Lissauer sees problems with the scenario. Stars are more luminous when they’re forming than when they reach the main sequence. In the case of M dwarfs, the difference in luminosity is significant. The zones that could become habitable around these stars are hotter during the planet formation period than similar zones around Sun-like stars. The so-called ‘snow line,’ which separates regions where rocky planets form from regions of icy planet formation, is proportionately more distant from an M dwarf’s habitable zone than it would be around a star like the Sun.

Another problem: planets take less time to form in the habitable zone of an M dwarf than they would around more massive stars because orbital periods are shorter and planetesimals bang into each other more frequently and with higher impact speeds. If anything, these frequent impacts may cause the young planet to lose gases and water rather than gaining them.

In short, our would-be terrestrial world forms in a volatile-poor environment and seems unable to retain what water it does accumulate. Lissauer’s conclusion is clear, and it makes grim reading for my imaginary M dwarf civilization:

In sum, under nominal circumstances, planets in main sequence habitable zones around M stars are likely to be fully formed and in their ?nal orbits by the time the gaseous circumstellar disk has dissipated or several million years after planetesimal formation, whichever is later. If growth is in situ, dynamical and thermal factors imply that the planets are unlikely to have large volatile inventories, and planetary masses are likely to be small. The large collision speeds of impacting comets, as well as the high activity and luminosities of young M stars, may lead to substantial mass loss from planetary atmospheres, depleting any reservoirs of volatiles that planets within the HZs are able to accrete.

A bleak picture for living worlds indeed. Are there any mitigating factors? Perhaps. A water-rich world could conceivably migrate inwards to the habitable zone of an M dwarf while the gaseous protoplanetary disk was still present, retaining some water even during the star’s active youth. Eccentric orbits can also be found that could allow water worlds to attain stability within the habitable zone.

But on balance, these exceptions look to be few. Writes Lissauer, “…the number of such planets is probably small, and Sun-like stars, despite being considerably less numerous, may well be the hosts of far more habitable planets.”

The paper is Lissauer, “Planets Formed in Habitable Zones of M Dwarf Stars Probably are De?cient in Volatiles,” in press at Astrophysical Journal Letters. Abstract available.