The idea that moons around exoplanets could be host to life is certainly not a new one. Thanks to Star Wars (and Ewoks), the idea of habitable moons is quite deeply embedded into sci fi pop culture — and as with so many things in the wonderful world of astronomy, both serious science and science fiction have given a lot of thought to the idea ever since. Being as we have a distinct shortage of Millennium Falcons to use, however, the difficult part is actually finding these moons.
Or is it? This recent paper by Kipping et al gives some compelling evidence that such exomoons could be found with the Kepler telescope. Kepler’s mission is to discover exoplanets by looking for transits. In other words, it very carefully watches stars for a long time. As a NASA mission specialist once said, “if Kepler were to enter a staring competition, it would win.” If any planets transit in front of their parent star, they’ll block a tiny fraction of the star’s light. Tiny, but enough for eagle-eyed Kepler to notice. Kepler then watches for a repeat of the transit. It can only decisively say that it’s found an exoplanet after three transits. One could be an anomaly. Two shows that something’s really there. Three confirms that it’s a planet with a regular orbit.
Even if the planet found is a gas giant, any moons in tow will exert a tiny gravitational influence on their parent planet. This will cause subtle irregularities in the planet’s orbit as the companion moons tug and pull on it. Kepler is sensitive enough to notice any such tiny irregularities. Then it’s a matter of doing the mathematical gymnastics to determine the moon’s mass and orbital characteristics. If the gas giant is the right distance from the star and the moon is massive enough to hold an atmosphere, there’s just a chance it could be habitable.
The authors refer to two techniques specifically: Transit Time Variation (TTV) and Transit Duration Variation (TDV). How precisely they work should be fairly self-explanatory. The basic principle is actually not dissimilar to the astrometry method of detecting explanets. A planet was discovered just a few months ago, using exactly that technique to watch for wobbling stars, proving the concept. If the concept can work for stars and planets then, in principle, there’s no reason why it shouldn’t work for planets and moons.
Seemingly, Kepler is sensitive enough to detect (under ideal conditions) exomoons as small as 0.2 Earth masses, as far as 650 light years away — at least if it’s looking towards M, K or G stars (that’s anything from a red dwarf up to slightly larger than the Sun). In fact, the smaller the star, the smaller the moons that could potentially be found. Even though the ideal conditions needed are unlikely in reality, it’s still impressively powerful.
These ideal conditions are a set of assumptions made in creating this model. Assumptions not unlike those made concerning Kepler detecting any planets in the first place. Things like stellar variability and non-circular or out-of-plane orbits could throw a spanner in the works of any detections. What’s more they throw in an assumption that only one moon orbits the planet of interest. This seems, at first, to be a little odd. Only one moon? The four gas giants in our solar system each host a small horde of moons. But if a planet was to host only one large moon, then it would approximate to much the same thing.
Having one moon that’s significantly larger than all the others makes Saturn a fairly good example. Alas though, despite being prominently the most massive of Saturn’s cohort, Titan is still too small at a mere 0.02 Earth masses. We wouldn’t see an exo-Titan around an exo-Saturn. It’s been said in the past (by a number of other authors) that for a moon to be habitable, it would need to be at least 0.1 – 0.2 Earth masses. That’s significantly larger than any moon in the Solar System! All the same, Saturn, with it’s low density, is just the kind of planet these techniques would work best on. Having a relatively low mass for its size, Saturn would get pulled around a lot more by its moon than, say, Jupiter.
Sadly, there’s one fact which could potentially spoil this whole idea. It’s not entirely clear if a moon could even form with enough mass to be seen this way. It’s been previously suggested that the largest moon that could form normally around a planet would only be one ten thousandth the mass of the parent. If Kepler does pick up any such exomoons, they’d probably be captured exoplanets. And no one knows exactly how common these are. The bottom line would seem to be that Kepler’s much more likely to spot an Earth-like exoplanet than an Earth-like exomoon. On the other hand, no one knows how common captured planets are because no one’s ever been able to look before. There’s still a fair chance that Kepler might find the first exomoon ever discovered. And that’s pretty special, irrespective of how that moon got there.
David M. Kipping, Stephen J. Fossey, & Giammarco Campanella (2009). On the detectability of habitable exomoons with Kepler-class photometry Monthly Notices of the Royal Astronomical Society (accepted) arXiv: 0907.3909v2