How Many Rogue Planets Are There in the Milky Way?

Life as we know it exists on a cozy planet in a stable orbit around a sun shining brightly in its sky. But a new study hints that the most common life in the universe might exist deep inside eternal-night worlds far from any star, adrift in the icy dark of interstellar space.

Free-floating planet
The Milky Way likely hosts billions, and possibly trillions, of unbound planets, some of which may have atmospheres thick enough to support bacterial life. Loose planets may even outnumber stars in the galaxy, but a more precise count awaits future telescopes such as WFIRST and LSST.
Researchers at the Kavli Institute for Particle Astrophysics and Cosmology (KIPAC) at Stanford University estimate that "nomad" planets could outnumber stars by as many as 100,000 to 1. Ejected from their home stellar system, the wandering planets now free-float through the Milky Way. Earlier estimates were more like a handful to 1, though previous studies have only counted unbound planets more massive than Jupiter.

To estimate the number of unbound planets as small as Pluto that could be roaming the galaxy, Louis Strigari (KIPAC), lead author of the study, began with a basic rule of nature: where a few big objects are found, there are many more small, just like a few boulders may be surrounded by thousands of pebbles. Strigari and colleagues calculated the number of unbound planets by extrapolating from the small number detected so far by direct imaging and by gravitational microlensing.

Direct imaging has severe limits because planets are so faint. Microlensing offers more promise. It looks for the characteristic brightening and fading of a background star when an object, even one as wimpy as Pluto, passes nearly in front of it and bends its light slightly by gravity. So far, 24 planet-mass objects have been detected by microlensing — 14 bound to their parent stars, 10 apparently not. Microlensing offers hope for detection of loose objects large and small even if they are completely dark, and even at great distances across the galaxy.

Media outlets have jumped on the vast number of loose planets predicted by this study. But the authors themselves acknowledge the large uncertainty in their result. The statistics remain so weak that the group’s lower limit is still one loose planet for each star in the Milky Way.

"There is no evidence for a large population of unbound planets, but then again there is nothing to rule it out," says Scott Gaudi (The Ohio State University), expert on gravitational microlensing and extrasolar planets, who was not involved with the study.

Exoplanet specialists think that some 80% of planetary systems go through an early period of gravitational chaos that flings some of their worlds into interstellar space. The survivors are left in a mess: in highly elongated orbits that are often not even in the same plane, as exoplanet hunters are discovering. Our solar system was one of the minority that avoided such a violent episode.

Still, skeptics have a hard time imagining how tens of thousands of objects larger than Pluto could originate for each star — regardless of whether they stay with it or get flung off. Our solar system has only 17 known objects that qualify — eight planets, seven large moons, and the two largest Kuiper Belt objects.

Better statistics for unbound planets may come from the Wide-Field Infrared Survey Telescope (WFIRST) and the Large Synoptic Survey Telescope (LSST), two gigantic sky-survey projects that may begin within the next decade or so.

"It's not implausible that a few nomads, most not much different than the size of Pluto, might lie within one light-year of the sun," says Strigari.

If more planets wander interstellar space than orbit stars, what might that mean for life in the universe? As it turns out, not all who wander are lost. Even with no sun in the sky, life could still eke out an existence under a thick atmosphere or ice crust at the depth where the temperature is pleasant, maintained for billions of years by the heat flowing from the planet's interior. Microbial life, at least, might thrive under these conditions.

All of this raises a question: If a planet leaves the system in which it was born, is it still a planet? The literature seems to give a qualified "yes," using terms like free-floating or rogue planets, though Strigari and colleagues prefer the term "nomad planet."

Monica Young has joined Sky & Telescope as an editorial intern. She is a recently minted astronomy PhD (from Boston University) with an extensive background in X-ray astronomy, education, and science writing.

27 thoughts on “How Many Rogue Planets Are There in the Milky Way?

  1. Paul Kinzer

    Decades ago, when I was young and had dreams of writing science fiction, I imagined that there might be dark, lonely worlds between the stars. I came up with an idea that these worlds might have evolved huge creatures that communicated, whale-like, via sound waves deep under the frozen surfaces, and that intrepid explorers on some of them would have found a way to break through to the surface. And, who knows how, they found others like themselves throughout the galaxy, and finally came into contact with us when an earth ship lands on one of their worlds.

    This was long before ‘Star Trek IV: The Voyage Home’, and, though I don’t remember the specific germ of the idea, I recall it came from something I read in an essay by Isaac Asimov.

  2. Henrik

    Very interesting but I have one small problem with this research as reported. It seems the authors assume that the origin of "nomad planets" must be a solar system. Not so. Have they forgotten their own pebble analogy when it comes to the big stellar nurseries? For every star born, there must be hundreds if not thousands of knots with less than the required mass to form even a brown dwarf.

  3. Mike Wrathell

    Dr. Young:

    Congratulations on your recent Ph. D.

    I must take exception with you calling Pluto "wimpy." If you want to be taken seriously as a scientist and writer, piling on Pluto by following in the footsteps of wannabe-killer of Pluto Mr. Mike Brown is not the way to do it.

    Do you ever want to speak to the planetary scientists working on NASA’s New Horizons mission to Pluto. What do you think they think about that little potshot you tossed Pluto’s way?

    Pluto has been maligned enough by Mr. Brown, a man who refused to acknowledge that Mr. Chad Trujillo and Mr. David Rabinowitz co-discovered Eris with him. His credibility as a scientist has suffered as a result of his narcissistic behavior, and I would hate to see you, too, fall into that trap.

    Please show Pluto the respect it deserves.


    Mike Wrathell, Esq. & Wikipedia-listed Artist

  4. Monica YoungMonica Young

    Henrik, you raise a good question. Stars and brown dwarfs do indeed form from the direct collapse of a gas cloud. However, the gas cloud (or region thereof) has to be above a critical mass to collapse. So while stars and objects as large as several Jupiter masses can form directly from the cloud, smaller objects have to form in the gas disk around a star.

    Mike, I do apologize for any disrespect to Pluto, none was meant! I meant only to emphasize the incredible power of gravitational microlensing. Can you imagine trying to image a Pluto-sized object floating alone in interstellar space? My point was only that gravitational microlensing, especially when done with a telescope as powerful as WFIRST, could find very small objects from very far away.

  5. Mike Wrathell

    Dear Monica,

    Apology accepted on behalf of Pluto. LOL! Yes, Pluto is only 2/3rd the size of the Moon, but it does have four known moons, hydrostatic equilibrium, a core, and perhaps an ocean or two! Pretty good for a guy only 7 miles or so wider in diameter than Eris according to Bruno Sicardy’s calculations! Alan Stern, Principal Investigator of New Horizons, and an IAU member who believes Pluto as a dwarf planet (he coined the term) is also a planet said that "barring a mishap," New Horizons will measure Pluto’s diameter, too. Hopefully, the data will show that Pluto is indeed larger than Eris, and put a rest to a lot of the negativity that Mr. Brown has been tossing around ever since the IAU refused to give Xena/Eris the respect he felt it deserved. I support the replanetization of Pluto and Ceres, and full planethood for all dwarf planets in our solar system, by the way. Yes, they are hard to see, but Earth is probably hard to see from Pluto, too. I sure hope the Plutonians do not think we are wimps! I think they might not be too happy with the IAU, though! But, there are a lot of proud Pluto huggers around! It is even in the Urban Dictionary now! Hope to see you in Honolulu at the IAU’s General Assembly in August 2015! I will be the guy with the "Replanetize Pluto!" sign! Truly, Mike.

  6. Martian-BachelorMartian Bachelor

    If Oort Clouds routinely form with stars and planetary systems, shouldn’t there be an even *larger* population of nomad extra-solar comets, which would be identifiable because of their hyperbolic orbits?

    My understanding is it’s not really clear if even one or a few such comets has ever been observed. For those known with perihelia inside the Earth’s orbit, Comet McNaught (C/2008 J4), e = 1.0272, has the largest eccentricity, which isn’t very much larger than 1, and may therefore be a result of gravitational effects within the solar system.

    So, has anybody run similar sorts of calculations for comets as they’re doing here for nomad planets? IOW, should we be worried yet that we’re not seeing enough free-range comets in our part of the galaxy as would be expected?

  7. BMayfield

    So, the incredible srinkage of our vantage continues. If true, this estimate means that in the set of all planets, we live on one of the mere 0.001% that orbit stars. Would this potential addition resolve the missing mass problem in astrophysics? And, what’s next? Is inter-galactic space packed with planets???

  8. Stan Kerns

    And, were this true, just how many planets would each and every star have to have lost if the orphans outnumber the stars by a 100,000 to 1? People really need to step back and look at their logic.

  9. Lawrence Geary

    I once took an astronomy class with W.L.W.Sargent, and he told us the story of an astronomer, a physicist and a mathematician traveling on a train through Scotland. They got off the train in the middle of a field, and in the distance saw a lone, black sheep. "My God!" exclaimed the astronomer. "ALL the sheep in Scotland are BLACK!" "You can’t draw that conclusion," said the physicist. "All you can say is that there is at least one black sheep in Scotland." "Nonsense!" replied the mathematician." All you can conclude is that there exists in Scotland at least one field with P > 0 of containing a sheep that is black on at least one side."

    The point being that astronomers tend to make wild extrapolations from meager data sets. I think this is an example of exactly that.

  10. Grant Martin

    If these nomad planets have been ejected from their birth stellar systems, they should be moving at fairly high velocities relative to the general population of stars. So, if they indeed outnumber stars by 100,000 to one, it would seem that at least a few would have passed through the Solar System’s Oort Cloud, thus sending a hailstorm of comets into the Inner Solar System. Might this be the trigger for the impacts that have caused mass extinctions?

  11. Rod

    "Microbial life, at least, might thrive under these conditions." I have a short comment about faith statements like this in science reports. Before we speculate about microbial life existing on unbound exoplanets, where did microbial life on Earth come from? Spontaneous generation of life from non-living matter or Creation?

  12. Charles Isbell

    Dr. Monica Young: Welcome to S&T’s staff of good writers!
    I enjoyed your article very much and look forward to many more in the not-too-distant future.
    Regardless what you write on/about, there will always be several who don’t agree with you (don’t ask me how I know!)
    Write well and carry a big stick!

  13. BMayfield

    Nice comments, y’all. I love exoplanets, and I’ve even used the more pebbles than boulders analogy myself to reason that planets must be abundant, but this estimate for unbound planets does seem way overblown. It really complicates my estimate the total number of universal planets challenge that I’ve posted earlier. It also reopens the debate about planetary nomenclature. Just when we’re getting used to Pluto as a "dwarf" planet we now find that such objects are NON-whimpy enough to cause detectible gravitational microlense effects? Dr. Young, just how massive does a body need to be to cause such effects? Maybe that should now be added to the list of planetary criteria, replacing the now even more ridiculous "cleared its region of space," since that would seem to be completely meaningless for nomads. So, Mike, congratulations! You’ve opened my eyes! YES, let’s stop calling all the little planets "dwarves!" And, I rant on, stop calling the sun a yellow "dwarf!" It’s yellow jounalism, I say, to malign an average, middle of the main sequence star just because its smaller than bloated, over-the-hill behemoths. Yes, Down with Dwarfism! Replanetize Pluto! and Dedwarf the Sun!

  14. BMayfield

    And, joking aside, to Rod and Lawerence et al I add, Biologists can also fall prey to the trap of wild exrapolation. Taught to believe the orthodox scientific dogma that life evolves without the aid of a creator, they make wild assertions about how common it must be for elements to spontaneously generate life. They reason, ‘there’s life on earth, there’s water on earth, so great numbers of water bearing worlds must have life.’ Now we’re to think that a zillion ice-planet-zeros have life? Reality check, please.

  15. Scott Dommin

    We’ll need a new word to describe these objects. We can’t call them "planets" (unbound or not) because according to the IAU’s definition they are something else. They don’t orbit a star or dominate their neighborhood.

  16. Rich Littleton

    Not a well thought out article.

    "There is no evidence for a large population of unbound planets, but then again there is nothing to rule it out,"

    Actually, there is MUCH to rule it out. Statistical improbility (of near zero) that the 100,000 to 1 ratio is anywhere near accurrate.

    To say that there are 100,000 unbound planets to 1 star is to say that EACH STAR represents 100,000 unbound plantes. Not bloody likely. Absolutely no stellar planetary system shows a number of planets above 20. If most discovered planetary systems are far less than 100,000, that would mean that, to reach the AVERAGE number of 100,000 to 1,there would have to be a large percentage of planets with even more than 100,000 unbound planets, in order to statistically reach the 100,000 planets-to-star ratio. Thus, this proposal for the 100,000 to 1 ratio would mean that many stars would have 200,000 to 1,000,000 unbound planets.

    And remember, the basis of the article is that there are 100,000 UNBOUND planets (per star), so the large numbe would have to apply to ESCAPED planets; all planets still inside stellar planetary systems would not be counted toward the ratio of unbound planets to stars.

    Again, as they say in French, "Not bloody likely"!!!!

  17. Bruce Mayfield

    Ouch, Rich. Our newly minted Dr. Monica Young is a fine writer and an excellent addition to S&T’s staff. Look at all the buzz her frist story is generating (even if you don’t count the double postings). I liked your statistical logic, however. It prompted me to print out and reread the paper that our dear Monica is merely pointing out to us. After doing my homework I come back with two points (1) At the right distances, an object with the mass of Pluto can cause a star to brighten by an easily detectible 10%. (2) All of us critics are largely overlooking the vast contribution of brown dwarfs and everything that forms with them. Think about it, if red dwarfs out-number all other stars, then brown dwarfs could dwarf (pardon the pun) the population of all stars combined! The 10^5/star figure includes everything below the minimum star forming mass. So, maybe this projection isn’t as off base as it may at first seem.

  18. Monica YoungMonica Young

    Glad to see so much feedback and discussion on my first article! I’d like to answer (to the best of my ability) a few questions that were raised, and I’ll do them in separate posts since these comments don’t allow for carriage returns. First, @Martian Bachelor, you raise a good question regarding "nomad" comets. Some astronomers have tried to address this calculation, but none have been able to do it with any degree of certainty. The problem here is that comets will be even smaller than Pluto and therefore even more difficult to detect unless they enter the inner solar system and being to evaporate to form a tail. To my knowledge, no comet has ever been detected entering the inner solar system on a completely unbound orbit. But keep in mind that the inner solar system isn’t very big, and lots of comets could be passing by, say, Jupiter, without us knowing about it.

  19. Monica YoungMonica Young

    @Bruce, you asked whether nomad planets could be the "missing mass" that astronomers have been looking for. The short answer is no simply because even if 100,000 unbound planets exist for every star (and that, again, is an upper limit), that wouldn’t be enough "dark" mass. Moreover, a significant amount of dark matter has to be non-baryonic (i.e., not made of the stuff that makes up planets) if astronomers want to explain the large-scale structure of the universe.

  20. Monica YoungMonica Young

    @Rich, I applaud your skepticism, which is warranted in the face of large uncertainties. However, I take issue with this statement: "Absolutely no stellar planetary system shows a number of planets above 20." The only planetary system we can explore with any degree of accuracy is the solar system, which is not (it seems) representative of planetary systems elsewhere. Exoplanet search techniques are only now starting to find Earth-sized objects. Even gravitational microlensing can’t yet detect Pluto-sized objects – that will require the more powerful WFIRST and LSST surveys, as mentioned in the article. Considering only the solar system, I pointed out in the article that we only know of 17 objects that are Pluto-sized or greater. At the same time, our knowledge of the Kuiper Belt and the much less understood Oort Cloud is constantly expanding. Some things to keep in mind as the discussion continues!

  21. Bruce Mayfield

    Thanks sincerely for the above answers, Dr. Young. I’m a bit puzzled, however, by part of your resonse to Rich; "Even gravitational microlensing can’t yet detect Pluto sized objects." In "Nomads of the Galaxy" (the paper causing all the fuss) Strigari et al. indicate in section 3.2 that for typical lens and source distances of around 5 and 8 kpc respectively and a lens mass of Pluto’s size the peak amplification would be 1.1, or 10%. Couldn’t even a well equipped amateur recording, say a globular cluster for a long enough period have a real chance of finding such an event? That’s just the kind of scoop we love reading about in S&T.

  22. Rod

    These are some interesting posts in this report. I raise a question about the 11 planetary systems Kepler-23 through Kepler-33 that hit the news back in January from NASA Kepler. Many of their average mean densities in g cm^-3 seem out of this world! I have not seen this discussed. You can see the radius and mass posted here, The transiting exoplanets show a growing population of bodies with very large mean densities, i.e. much greater than 5 g cm^-3 as Venus, Earth, or Mars. Just what type of exoplanets are we encountering now out there?

  23. Monica Young

    @Bruce: I think I spoke (wrote) too quickly, so let me rephrase. I think it’s possible that an amateur telescope could pick up such an event, but also extremely unlikely. To find significant numbers of events would probably require a dedicated survey, like the WFIRST survey of the inner galaxy that Strigari et al. suggest.

  24. Bruce Mayfield

    Aha! I say, like a country lawyer who just got the star, bona fide expert witness to make a key admission. Please permit me to outline the case; Strigari and his fellow defendants are being tried in the court of public opinion. The alleged crime is making the outrageous claim that there might be 10^5/star "nomad" whatyoumacallites in the galaxy. These free-range, feral planets lie in the 10^-8 to 10^-2 solar mass range. A 10^-8 (Pluto sized) object can produce a dectable 10% amplification in a background star. I refer now to exibit A, the paper "Nomads of the Galaxy," section 3.1. There those who can follow the math will see that the amplification rate is directly proportional to the mass of the lensing object. Therefor a 10^-7 solar mass object could cause a 100% brightening, a 10^-6 1,000%, and so on. Why then, I ask the court at large, haven’t we seen these events already?

  25. Laurel-KornfeldLaurel Kornfeld

    You wrote: "Our solar system has only 17 known objects that qualify — eight planets, seven large moons, and the two largest Kuiper Belt objects."

    I object to the characterizing of our solar system having only eight planets, which represents only one view in an ongoing debate, not fact. Pluto and Eris are NOT just Kuiper Belt Objects. They are both Kuiper Belt Objects and small planets. According to the equally legitimate geophysical planet definition, dwarf planets are simply a subclass of planets. With five dwarf planets recognized by the IAU, three newly discovered last year, and several that are probably large enough to be in hydrostatic equilibrium, or solar system has a minimum of 16 planets and probably many more. Please do not accept and repeat the controversial IAU line as fact when it is not.

    The Sun happens to be located in an area where there aren’t many nearby stars. Areas with higher concentrations of stars are more likely to have rogue planets that result from interactions between two or more solar systems. One could guess there are probably many rogue planets closer to the galactic center.

All comments must follow the Sky & Telescope Terms of Use and will be moderated prior to posting. Please be civil in your comments. Sky & Telescope reserves the right to use the comments we receive, in whole or in part, and to use the commenter’s username, in any medium. See also the Terms of Use and Privacy Policy.


This site uses Akismet to reduce spam. Learn how your comment data is processed.