ANOTHER COUP FOR KEPLER?

If the results bear fruit, the moon would be the latest in a remarkable string of discoveries for Kepler. Launched in 2009, the space-based observatory has found more than 2,000 alien worlds and about 4,000 candidate planets, and astronomers are not yet done mining its riches. In June, astronomers using Kepler data identified 219 more candidates alien planets, including some that may be habitable like Earth.

Kepler works by detecting when these distant planets pass in front of their home stars from Earth’s point of view. This transit momentarily blocks a fraction of the star’s light, causing a periodic dip in apparent brightness.

Detecting a moon orbiting a planet using this same technique is extremely difficult. Moons are even smaller than their planets, which means that their transits don’t block much starlight. In addition, astronomers must painstakingly tease apart the signals from moon and the planet it’s orbiting.

But these challenges haven’t stopped scientists from trying to find alien moons, some of which might be habitable, à la Pandora from Avatar or the moons of Endor in Star Wars. Since 2012, Columbia University astronomer and study coauthor David Kipping has spearheaded the Hunt for Exomoons with Kepler (HEK), an effort to comb through Kepler data for hints of moons.

The researchers’ new paper, published on the preprint service arXiv, focuses on 284 Kepler planets deemed the likeliest to bear moon systems resembling Jupiter’s. The team then statistically stacked these planets’ transit data, hoping to see smears that moons would leave in the collective signal.

Some of the planets are Jupiter-size but are snuggled close to their stars. Astronomers think these so-called hot Jupiters formed in the chillier outskirts of their star systems but then migrated inward—raising questions about what would happen to their moons.

“They’re looking at planets that are much closer to their suns than Jupiter is to our own,” says Leiden Observatory astronomer Matthew Kenworthy, who wasn’t involved with the study. “So the question is, during this process of migration, do big fat gas giants lose their moons?”

According to the latest data, these Kepler planets aren’t teeming with moons. At most, the researchers say, no more than 108 of the 284 studied worlds could have them. This constraint suggests that many Jupiter-like planets do shed their moons if they migrate.

Hoping for Hubble

But when researchers applied quick-and-dirty moon models to the 284 individual planets, they also uncovered a compelling signal from Kepler-1625b. Additional bumps in the data suggested a smaller, Neptune-size body was orbiting the planet.

Under certain assumptions, there’s at most a 1-in-24,000 chance that these fluctuations are a fluke. While that may be sound convincing, it merely qualifies as evidence in the realm of astrophysics. The October Hubble observations will make or break the case for the moon.

Coauthor Teachey says that if he were a gambling man, he’d be willing to bet a bottle of wine—but not his car—on the moon’s existence. But Teachey, by his own admission, isn’t a scientific gambler, and neither are the other astronomers contacted for this story.

“If it’s true, it’d be awesome,” says Kenworthy. “But right now, and [the study authors] say this very clearly, it’s tantalizing. It’s not a detection.”

MIT planetary scientist Sara Seager, a world authority on planets beyond our solar system, agrees.

“Any time the word ‘candidate’ is in the [study] title, it is just that, a candidate,” she says in an email. “I am definitely looking forward to the Hubble Space Telescope observations in 2017 to see if anything is actually there.”

Nadia Drake contributed reporting.