Today, the team running NASA’s Kepler observatory is announcing the largest collection of planets yet spotted orbiting a single star. That star, now called Kepler-11, hosts at least a half-dozen planets, most in orbits closer to their star than Mercury is to ours. The tightly packed system consists of low-density planets, suggesting that they have a substantial gas or liquid composition at distances close enough to the star that the material risks boiling off. The whole thing, from the orbits to the planets themselves, appears to be teetering on the edge of stability.
The Kepler team’s paper starts with a short description of the Kepler hardware, which is pointing a CCD and a telescope down a spiral arm of the Milky Way, watching for periodic dips in the amount of light originating from stars. The raw light curve for Kepler-11 shows a large series of gaps in the star’s light, as several of the planets transited in rapid succession at several points. Single planetary transits are a bit harder to detect, but when the data is zoomed in, a characteristic U-shaped curve for each planet is apparent, which occurs as each planet passes through the line of sight between the star and Earth.
(Ground-based spectroscopy shows that, within measurement errors, Kepler-11 is identical in mass and size to our own Sun.)
One of the challenges faced by the Kepler team is confirming that the periodic dips come from an actual planet, not from another astronomical phenomenon. Because of the tight packing of the planets—their orbital periods range from 10 to 47 days—this task was made much easier; their mutual gravitational interactions cause variations in the precise timing of their transits across the star. By tracking these shifts, researchers were able to confirm that five of the bodies were all orbiting close enough to gravitationally interact, so they must be planets.
The sixth body is quite a bit further out (it orbits the star every 118 days, a period 2.5 times longer than the next closest body), so this method wouldn’t work. Instead, the authors performed a probability analysis, comparing the chances that it was a planet to the chances that the signal matched a number of other options. The probabilities worked out such that they’ve termed its planetary identity “validated,” though detailed followups would obviously be welcome.
The system is largely flat. Five of the planets have an inclination within a single degree of each other, even without measurement errors being considered. The sixth is less than half a degree off.
A question of stability
Can you actually cram this many planets into this small a space? Yes, apparently, but just barely. All but the innermost two planets are far enough apart to avoid destabilizing gravitational interactions. The inner two could reach a stable resonance, provided that they don’t gravitationally interact significantly with the rest of the planets. (Their orbit is right on the edge of where that should happen.)
Models of the system are very sensitive to initial conditions; if the distances are just right, the system is stable, but many of the model runs showed the system going unstable over short time periods. Continued imaging should narrow some of the errors from Kepler’s initial measurements, and may give us a clearer picture of whether the system can persist in this configuration.
What about the planets themselves? The amount of starlight blocked gives us a measure of their size, while the orbital interactions and distance from the star gives us a measure of mass, so it’s possible to estimate densities for all of them. The planets are all larger than the Earth, with radii that range from twice as large as ours to 4.5 times as large. All are quite a bit less dense than the Earth, however, suggesting that they have hydrogen-rich atmospheres, or are primarily composed of liquids. There’s a single planet that might be consistent with an iron-free silicate solid, but the authors can’t imagine how something like that would form, so they find the prospect pretty unlikely. Followup observations should allow us to distinguish between steam-heavy atmospheres and hydrogen-rich ones.
Based on the energy delivered by the nearby star, the authors estimate that the planets will lose portions of their atmospheres to space annually, with about a tenth of the Earth’s mass lost in 10 billion years, which could cause them to lose their entire atmosphere within this span. Our current models of this process are crude, but the authors suggest that it’s safe to assume the system has seen a lot of hydrogen pushed off the planets by their host star.
Because the planets are at the edge of stability, the authors suggest they formed in place, rather than migrated inwards. That, in turn, suggests a “massive protoplanetary disk of solids” near the star, or else that small bodies wandered in, were slowed by gasses, and then gathered those same gasses into an atmosphere. Either case would help us update our ideas on planet formation, so the Kepler team is keen for more data on this system.