Most alien planets are nothing like what we've got in the solar system. Scientists are homing in on these mysterious worlds to see what they’re made of.
Of the Kepler spacecraft’s impressive haul (3,538 exoplanet candidates as of November 2013), roughly three-quarters of them scale between Earth and Neptune, which is four times the size of Earth.
In other words, most transiting planets are nothing like the worlds in our own solar system.
With no local representatives, it’s not clear how many of these planets are super-Earths (that is, rocky) or gaseous mini-Neptunes. The only way to find out is to measure their mass, and that’s hard for Kepler to do. The spacecraft spent four years detecting planets as they passed in front of their host stars, measuring the dips in starlight as they did so, but astronomers can measure only the planet’s size from the shape of the blip, not its mass.
So the Kepler team took to the ground, using the Keck Observatory in Hawaii to follow up space-observed transits with good, old-fashioned radial-velocity measurements — the planet-finding technique of choice before Kepler came around. The team looked for each host star’s telltale wobble as the star and its attendant planet orbit a common center of mass.
At the January 2014 meeting of the American Astronomical Society, Geoff Marcy (University of California, Berkeley) announced the results of four years’ worth of these ground-based observations, which followed up on 22 stars with 42 exoplanets. The team measured masses for 16 planets, and rough guesses (in technical terms, 1-2σ measurements) or upper limits for the rest.
Another team led by Yoram Lithwick (Northwestern University) took a different approach. Rather than measuring the star’s wobble, Lithwick and graduate student Sam Hadden measured the planet’s wobble. If two or more planets orbit the same star, the planets will gravitationally tug each other, tweaking the timing of each transit. Using this method, Lithwick and Hadden measured the masses of another 58 super-Earths. Most of these are relatively firm estimates, with only about 10 "rough guesses."
Measuring the masses was hard work, but well worth it because now we know the densities of all those planets and can start to talk about what they're made of. Both Marcy and Lithwick found the same thing: up to somewhere between 1.5 to 2 Earth radii, the bigger the planet, the denser it becomes. That’s about what you would expect for rocky planets.
But theory says that at a certain point, gravitational compression wins out — add more mass and it just compresses further — so rocky planets should never get any bigger than about twice Earth’s size. And that’s exactly what Marcy and Lithwick find: planets bigger than about twice the size of Earth actually get fluffier as they grow. These bigger-but-less-dense planets are probably adding an extended atmosphere of mostly hydrogen and helium around a rocky core.
So is there a simple dividing line between the super-Earths and the mini-Neptunes? That would be convenient for targeting exoplanet searches. And observations so far bear out the theory that no rocky planet can be bigger than two Earths.
But apparently it doesn’t work the other way around: in the same set of presentations, David Kipping (Harvard-Smithsonian Center for Astrophysics) announced the discovery of KOI-314C, the first Earth-mass planet that is surprisingly nothing like Earth. Its radius is 60% larger than Earth’s, so in classic mini-Neptune style, it too probably has a rocky core surrounded by gas.
“If you told an astronomer yesterday that they’d found an Earth-mass planet, they’d think it would surely be rocky,” Kipping says. “[This planet] suggests we can’t just draw a line in the sand.”