A new result from astronomers who have spent years peering toward the center of the Milky Way has led to a startling conclusion: there may be billions of Jupiter-sized planets wandering the space between the stars, unbound by the gravity of a parent sun. In fact, there may be nearly twice as many of these free floating planets as there are stars themselves in our galaxy, and they may even outnumber planets orbiting stars!
The study, published in Nature, is the result of the Microlensing Observations in Astrophysics (MOA) project. Instead of looking for tell-tale blips of light near stars, or the effect of planets on their parent stars, microlensing looks for the effect of the planet on background stars that are far more distant than the planet itself.
Warped view It's a little weird, and is due to gravity warping space. Imagine me sitting on a flat floor, rolling marbles away from me in all directions. If you're sitting a few meters away, you can only catch the marbles that are aimed at you. But if there's a dip in the floor between us, some of the marbles I roll that might have otherwise passed you will get their path diverted toward you as they curve around the dip. You get more marbles!*
The same thing with light and gravity. A star emits light in all directions, but we only see the small amount of light headed our way. If a massive object like a planet gets between us and the star, the gravity of that planet can warp space, causing light we otherwise wouldn't see to bend toward us. We see more light: the star gets brighter! This is called a gravitational lens. If that massive object is a planet moving in space, then we the starlight get brighter as the planet moves between us and the star, and then fainter as the planet moves on. The way the light changes is predicted by Einstein's equations of relativity, and can be used to find the mass of the planet doing the warping.
OGLEing a MOA So the astronomers with MOA sat down and stared at a patch of sky near the center of the Milky Way. In fact, they looked at an astonishing 50 million stars near the galactic bulge -- stars are densely packed there, maximizing the chance of seeing a rare event. The lensing of starlight by a passing planet only lasts for a couple of days, so they took images every 10 - 50 minutes to make sure they caught as many as possible. The amount of data they amassed is fiercesome.
And even with all that, in a year of observations (from 2006 - 2007) they only caught about a thousand events. At first that sounded like a lot to me, but it's only one lensing event per 50,000 stars! Yikes. Anyway, of those 1000, a bit less than half were solid enough observations to use in the study. And of these, only 10 -- ten -- had that magic characteristic time of about 2 days, indicating the lens was a planet with about the mass of Jupiter. Stars are more massive, and the lensing effect can take weeks from start to finish; only a planet can make such a short event. Being careful, the astronomers took those 10 events and asked the folks using a different survey (OGLE, for Optical Gravitational Lensing Experiment; acronyms using "GL" tend to be somewhat droll) to see if they saw them as well. OGLE caught 7 of the 10 seen by MOA, confirming their results. So what makes the astronomers think these are free-floating planets, and not ones orbiting stars like Earth does? Well, the lensing events themselves show only a single rise and fall of the background starlight. If the planets were orbiting stars, those stars would also act like lenses, and their effect would be seen. They weren't. Now, it's possible that if a planet were on really wide orbit, the parent star would be too far away to have a significant lensing effect. However astronomers can determine statistically how often that should happen, and the likelihood is only about 25%, meaning a significant number of the events must have been caused by planets without stars.
Goose! Eject! Eject! Amazing! But where did these planets come from? Since they're floating free in space, they either formed like stars, directly from the collapse of interstellar gas clouds, or they formed in solar systems like our own and somehow got tossed out. The first case -- that these objects form like stars -- makes a definite prediction on the distribution of masses of the objects (in other words, how many will have a mass 0.1 times Jupiter, how many 0.5 times, and so on). The mass distribution seen doesn't fit the predictions at all, making that unlikely. So that leaves them forming in solar systems like our planets did. But how does a planet get ejected from a star? Actually, this comes about naturally, and in fact may be common.
I've long suspected this was the case; it makes sense. We see lots of massive planets huddling close in to their parent stars, far closer than any reasonable model can predict. Most likely, these planets form farther out in their native solar system and then migrate inwards toward the star as they plow through the material left over from their formation. Any planet between them and their star will be affected; some will shift orbit, dropping toward the star themselves, others will get flung into wide orbits, and others still will be tossed out of the system entirely. It's those last that are so interesting. If the inward-moving planet is, say, five times the mass of Jupiter, it can gravitationally eject a smaller planet, even one as massive as Jupiter. And we do see lots of very massive planets orbiting close in to their stars. This strongly implies that for every "hot super-Jupiter" we see, there is one or more planet that got kicked out of the system, sent out into the galaxy at large á la Space: 1999
Living la vida interstellar The MOA results seem to confirm this idea: the statistics imply that there may be twice as many Jupiter-mass free-floating planets in the galaxy as there are stars! Just to let you know, there are hundreds of billions of stars in the galaxy, so there must be many, many billions of planets floating in the vast, empty regions between stars. Billions. Wow. In fact, these free-floaters may outnumber "regular" planets by a factor of 1.5 or so. There are more of them than there are of us! Mind you, the MOA survey is sensitive to planets with masses about that of Jupiter. They can't see smaller planets, which should in fact be more common. These planets, surprisingly, may not be frozen solid as you might expect. Jupiter and Saturn, for example, give off more energy than they receive from the Sun
. The centers of both planets are still warm from a number of heat sources, including radioactive decay as well as having trapped a considerable amount of the tremendous heat generated when they formed 4.6 billion years ago. Any free-floating planet in the galaxy may be presumed to contain as much heat, keeping them gaseous despite the intense cold of interstellar space. You may be wondering about any potential habitability of these nomads. The planets found are gas giants, not Earth-like at all. But they might have moons orbiting them that could be heated by tides the same way Jupiter's Io and Saturn's Enceladus are. It seems unlikely that any moon could stay orbiting a planet ejected from a solar system -- I would think they'd get stripped from their parent planet in the process -- but nature has surprised us before. Like, say, it's doing now with this whole "wandering planet" thing. I'd love to see some studies of that. Also, while there may be an even greater number of smaller planets out there, these likely would be frozen through and through. Too bad. The view would be cool. So to speak.
Conclusion The MOA study in question is pretty interesting to me scientifically. The results look pretty good, and come from only a year's worth of data; as the astronomers look at more data they're bound to find more of these suckers. I expect to see their statistics get better with time. They seem to have done their work carefully and skeptically; it's a fascinating result and I'm glad they sought out OGLE observations to back them up. Personally, too, this is exciting. Imagine, a galaxy full of roaming planets! It's not like they present a navigational hazard were we to fly starships through space; the galaxy is vast indeed and even a hundred billion planets would be spread pretty thinly. But it sparks my imagination to think of these planets -- dark, cold, lonely -- plying their way through the blackness of interstellar space. If we ever could voyage to one, what would we find? It's thoughts like that which make me glad to be an astronomer, especially one living now. Just when you think the Universe is running low on surprises, it reminds us it's a lot more clever than we are.
Artwork credit: Dan Durda; NASA/JPL-Caltech
^* While I slowly lose mine, of course.