≡ Menu

Evening Landscape with Exomoons

I often work out my thoughts on the topics we discuss here while taking long walks. I try to get in five miles a day but more often it’s about three. In any case, these long, reflective walks identify me as the neighborhood eccentric, an identity that is confirmed by the things I write about. What’s interesting about that is that so many people have a genuine interest in the stars and how we might get there. Some of the best questions I’ve ever had have been from people whose interest is casual but persistent, and one good question usually leads to another.

Hence I wasn’t surprised on yesterday’s walk to find myself talking with a neighbor about exomoons and why we study them. After all, we have a Solar System in which moons are commonplace. Isn’t it perfectly obvious that different solar systems would have planets with moons?

The answer is yes, but it also follows that things that seem perfectly obvious still have to be confirmed. But let’s unpack it a bit more than that. We’re familiar with our own system’s configuration, in which moons of astrobiological interest are orbiting gas giants a long way from the inner system. But we know from our exoplanet work that large planets like these can exist in warmer places. Thus the notion of habitable moons around gas giants, or perhaps double planets in the habitable zone, something like a larger version of Pluto and Charon in a comfortable orbit.

Popular films like Avatar keep the exomoon theme in front of the public, whose interest is understandable. After all, could anything be more exotic than a warm gas giant orbited by something a bit like the Earth? From an astrobiological perspective, the thought of Europa or Titan analogs in warm orbits is thrilling, a reminder that life may have gained many footholds in the galaxy. The Hunt for Exomoons with Kepler project is all about figuring out the occurrence rate of large moons so we can learn whether such moons are common.

The other aspect of exomoon detection has to do with increasing our expertise. It wasn’t so long ago that we had yet to detect our first exoplanet. Now we’re delving into planetary atmospheres and working out the orbital dynamics of multi-planet systems. An exomoon detection would be a major proof of concept, demonstrating the growth of our skills. It would also begin to build an exomoon catalog that will help us understand how important exomoons may be to planetary habitability. How big a role does our own moon play in keeping our planet habitable?

There’s also plenty to learn about how planetary systems form in the first place. We now think our Moon formed in a massive collision (the Big Whack) with a Mars-sized object in the early days of our planet’s history. How likely an event is this, and how often does it happen in other Solar Systems? We still have a lot to learn about how the satellite systems around various planets emerge, especially when we consider the wild variety of moons we see in our Solar System. Building the exomoon catalog will help answer these questions.

The Joys of Beta Pictoris b

I hadn’t planned to get into exomoons today, but serendipity struck. After yesterday’s conversation I ran across Phil Plait’s latest essay for Slate. The popular astronomer and science popularizer (author of Death from the Skies! and, of course, Bad Astronomy), now explains that because of an unusual alignment beginning in 2017, we may be able to detect an exomoon, if there is one, around the planet Beta Pictoris b.

We’re dealing with a system far different from our own. Some 60 light years away, the star Beta Pictoris is more massive than the Sun and a mere infant, at 25 million years old, compared to our own star (around 4.5 billion years). This is a solar system in formation. Moreover, it has been under intensive study since scientists realized it was surrounded by a large circumstellar disk. The planet Beta Pictoris b was first imaged in 2003, a world more massive than Jupiter that orbits its host every 20 years. You can see its movement in the time-spaced images below.


Image: Infrared images of the planet β Pictoris b obtained in 2003 (a), 2009 (b) and 2010 (c), showing the planet’s movement in an orbital plane that is nearly edge-on as seen from Earth. The host star is in the central part, but its light has been suppressed to show the fainter planet. The white dots in b and c denote previous positions of the planet. Faint blobs are optical effects. It is not possible to tell from these images whether the planet is orbiting towards or away from us, but {Ignas] Snellen and colleagues’ spectroscopic observations clearly indicate that the planet is currently in a part of its orbit where it is moving towards us. Credit: ESO.

Note in the description above that the planet’s orbital plane is close to edge-on from our perspective. It’s not close enough to make a transit possible, but what Plait talks about is
the next best thing. Drawing on a paper by Jason Wang (UC-Berkeley) and colleagues, Plait explains that the region around the planet called its Hill Sphere will pass in front of the star from our perspective. The Hill Sphere is the area around an astronomical body in which its gravity dominates. In other words, within the Hill Sphere, a moon could be retained by the planet.

Nobody explains such concepts as well as Phil Plait, so I’ll give him the floor here, drawing directly from his essay:

The size of the sphere depends on the mass of the planet, the mass of the star, and the distance between them. For example, the Earth’s Hill sphere reaches out to about 1.5 million kilometers. The Moon, orbiting 380,000 km away, is well inside that, so its motion is mostly influenced by the Earth (some people like to say the Moon orbits the Sun more than it does the Earth, but those people are wrong). Weirdly, Pluto’s Hill sphere is much larger than Earth’s, but that’s because it’s so far from the Sun that an object can orbit Pluto from farther away and still be heavily influenced by it.

What emerges with regard to Beta Pictoris b is that its Hill Sphere is 160 million kilometers in radius. We get no transit of the star by the planet itself, but by August of 2017, the planet will be at its closest approach to the star and the Hill Sphere region will transit. We’ll be able to look for debris or exomoons. A large moon passing in front of the star would be the first entry in the exomoon catalog.

But even if we get no exomoon detection, bear in mind that we may make other interesting observations. This young planet is still being born, and it may well contain a circumplanetary disk of its own, or even a ring system that is the residue of planet formation. “The transit of β Pic b’s Hill sphere,” Wang et al. write, “should be our best chance in the near future to investigate young circumplanetary material.” We’ll also learn a lot more about how Beta Pictoris b perturbs the circumstellar disk, a window into early solar system formation.

All this is good material for my next walk and the conversations sure to follow. The paper is Wang et al., “The Orbit and Transit Prospects for β Pictoris b constrained with One Milliarcsecond Astrometry,” accepted at the Astrophysical Journal (preprint).


Comments on this entry are closed.

  • Harry R Ray August 12, 2016, 13:10

    In 1981, there was a 1-2% dip for several days that ORIGIONALLY was attributed to the PLANET transiting. Now that this is proven NOT to be the case, has the original data been re-analysed to determine whether we might have caught the very outer edge of the outermost ring transiting instead?

  • Ashley Baldwin August 12, 2016, 13:54

    Funny things exomoons. Something we all more or less take for granted but tantalisingly at the extreme edge of current detection methods and as yet hypothetical only though hopefully David Kipping will turn some up out of either the Kepler data as techniques improve or alternatively PLATO next decade. ( David despite his brilliance is still a young man !) . The other big name associated with exomoons is Rene Heller, of McMaster University in Canada a till recently but now of the Max Planck Institute in Gottingen ,Germany.

    Heller has been simulating exomoons for years , generally around gas/ice Giants. Essentially there are two types. Regular and irregular. Regular moons form from the same area of a protoplanetary disk as their parent planet and irregular moons are essentially captured individual components of passing binary planetary systems , believed to be common in a solar systems early life.
    Heller has modelled moon formation around varying sizes of gas giant from Jupiter size up to twelve times larger. The biggest moon he could produce this way was Mars sized and of course as it’s likely that gas giants form beyond a system’s ice line , were icy in consistency . Should that gas giant subsequently migrate into its stars habitable zone the moon would melt and Heller postulates there may be many potentially habitable Mars sized ocean moons out there waiting to be discovered .

    Other important issues with gas giant moons , regular or irregular , is there proximity to the planet. Too close and gravitational tides will lead to run away heating and a dead moon as well as the planet regularly eclipsing the moon and thus leading to extended dark/cold periods . To this effect his research has shown that dependent on the gas giant , any exomoon, to be habitable would have to orbit at-least 5-20 planetary radii out from the planet to minimise these potentially harmful effects. Fair enough. The issue then is what about magnetic field atmospheric protection? Probably important . Gas giants have big magnetic fields but from Juno’s journeys and Europa too we know the potential risk of radiation belts around such planets to say nothing of the harmful effects described above. Getting past these by orbiting further out is fine , but that will take any moon beyond the protective envelope provided by its planet’s own magnetosphere . The question is then can that moon produce its own magnetic field ? Ganymede does , but this is likely due to a subsurface briny ocean . That said , it’s important to remember that although moons are tidally locked , they are locked to their planet NOT their star. So any moon orbiting a gas giant within a solar type sun’s habitable zone will likely rotate ,in relation to its star, far more quickly than its orbital period . So it’s possible that a magnetohydrodynamic field could be produced by related circulation in its outer liquid metallic core ( if it has one ) if there is enough convection in its mantle . Maybe there are other ways thinking about Ganymede.

    Mars sized is not very big. We know that being much smaller than Earth led to Mars losing its internal heat and magnetosphere relatively early and thus its atmosphere ( thanks also to its low mass/gravity)

    As stated , irregular moons ( such as Triton) are likely captured components of a binary planetary system. Unlike regular moons they are not necessarily limited to Mars and could easily be bigger, Earth sized even. The question is whether such large moons could be captured and it’s here that relative closing velocities of the gas giant and binary system matter. The less the velocity , the greater the likelihood of an ( intact) capture . Lower velocities will occur with smaller gas giants with smaller gravity ( perhaps even as it is migrating inward to the habitable zone where a rocky binary system conveniently awaits ) , perhaps Neptune sized best ? Triton is likely a stray icy KBO, but any inner system planetesimals are likely to be rocky , just perfect to make an Earth like moon that is sufficiently close to rotate enough to produce a magnetic field yet not be subject to harmful tidal heating or eclipses .

    It’s worth remembering that gas giants are targets for cometary and asteroid impacts thanks to their large gravity , so even a habitable exomoon would be subject to a greater bombardment than Earth itself though ironically , as with Jupiter’s “grand tack” it may be that the very inward migration that allows a gas giant to pick up a habitable large moon may help clear away a lot of the debris left over after planetary formation and help reduce future risk of collisions.

    So we raise a glass to David and Rene and hope for an Avatar or Star Wars style paradise moon ……

    • Steven Torry Rappolee August 12, 2016, 23:47

      It should be noted that Robin Canup at SwRI wrote about Constraints on moon size during Gas Giant formation and also came up with maximum sizes.Mars-sized moons being a rare exception.


      Irregulars need more study, how often would that produce a habitable world around a gas giant?
      More interesting would be how often a habitable exomoon is formed from the impact with an ice giant or smaller world such as a super earth or mini-Neptune?
      Such numbers should they become known would add to the equation that must include the rare earth hypothesis.

  • Michael August 12, 2016, 16:24

    I hope the first indirect evidence of exomoons is around a planet of Tabby’s star :)

    • ljk August 15, 2016, 10:52

      Does Tabby’s Star have any detected exoplanets? I was not aware of them.

  • Steven Torry Rappolee August 12, 2016, 23:50
  • Ronald August 13, 2016, 11:11

    After having browsed though some of the work by Heller (referred to by Ashley Baldwin above) and by Canup & Ward (referred to by Steven Rappolee above), I am not too optimistic about habitable moons;
    – The max. mass of an in situ (regular) moon is about 0.01 – 0.02 % of the gas giant mass. This means that even a very large gas giant would have a roughly Mars sized moon at the most, and gas giants are rare in the HZ. On top of that there are the problems of radiation belts and tidal heating. Neptune class gas planets and gas dwarfs (mini-Neptune) are very common, but they won’t have large regular moons.
    – Captured (irregular) moons, like Triton, could be larger, but how common will they be, especially in the inner system, in view of the three body problem?
    – The third possibility, an impact-formed moon around a Neptune class or mini-Neptune, mentioned by Rappolee above, is interesting, but I wonder whether such large planets can have any impact moons.

  • Ronald August 13, 2016, 11:27

    Which *does* bring me to a very fascinating idea, just an idea, and more the realm of SF: a true binary planet system, consisting of a gas dwarf (mini-Neptune) and a terrestrial planet, formed by impact.
    Is this physically possible? How likely is it? Is there any SF with that theme?
    Since the most common planetary system seems to be the compact system of ice giants (Neptunes) and gas dwarfs (mini-Neptunes), this issue may actually have some relevance.

    • Steven Torry Rappolee August 15, 2016, 0:07

      My best guess is that the Mini Neptune with an Impacter might be the most common scenario and one that could lead to a habitable planet.
      let us say that this happens 3% of the time or 5%.You would think this type of binary planet would be as common or more “common” than the Earth-Moon system.
      So for SciFi art we want to see a Mini Neptune floating in our sky :)

      • Michael August 15, 2016, 11:06

        The problem with gas and icy giant collisions is that they are made mostly of gas and ices, any collisions release large amounts of volatile materials which are hot and won’t condense fast enough.

        • Steven Torry Rappolee August 18, 2016, 3:53

          I agree with you but would this be the case for super-earths and mini-Neptunes? or objects in between.Such objects might have sizable rocky cores
          Such objects might differ if formed close in or in orbit about red dwarfs
          Uranus might be an example of a collision without a happy ending in that its moons are small

  • Phil August 13, 2016, 22:07

    Somewhat OT, but is this just hot air, or has the Pale Red Dot program leaked?



  • Paul Gilster August 14, 2016, 8:31

    I’ve seen the stories, which all seem to come from one source. I don’t cover stories on embargo breaks, though, because they’re often misleading. In any case, once the data on any Proxima planet are released publicly, you can be sure we’ll take a close look. Right now the timing on that is indefinite.

    • Harry R Ray August 15, 2016, 9:49

      I have been monitoring this DAILY! Their website reported positive reports from referees over a month ago. I guess it all depends on whether the paper has finally been accepted. AND: Whatever happened to Kipping et al’s MOST observations which were SUPPOPSED to be completely reduced by the end of June? In my WILDEST “Centauri Dreams” I can envision a JOINT ANNOUNCEMENT!!!

      • Harry R Ray August 15, 2016, 10:28

        David Kipping tweeted on August 13: “I know Proxima is being talked about a lot this weekend.-I can’t comment on the #MOST transit campaign until a few weeks time!”

      • Eric Goldstein August 15, 2016, 10:49


        David Kipping had this to say:
        “I know Proxima is being talked about a lot this weekend- I can’t comment on the #MOST transit campaign until a few weeks time.”
        Source: https://twitter.com/david_kipping/status/764581374099685376

        I admit I also playfully imagined a joint announcement from all the proxima centauri search teams.

      • ljk August 15, 2016, 11:00

        According to Pale Red Dot via their Twitter account, they are saying the rumor of an exoplanet circling Proxima Centauri did not come from them:


        So if not PRD, who else is monitoring our nearest (known) stellar neighbor? Could it be these guys:


        The recent CD article on PRD:


        • Harry R Ray August 15, 2016, 15:07

          Don’t forget the two “mesolensing” events with HST! Although, I seriously doubt that a “habitable zone” detection(even one in the OUTERMOST HZ)could be made via this technique, and any earth-mass planet they could detect would lie outside the HZ. My GUT FEELING, though, is it is KIPPING! Just the way he worded his tweet has “EMBARGO” written ALL OVER IT! You just DON”T embargo non-detections. GREAT CAUTION IS URGED HERE, because if the planet candidate is at the same distance as indicated by the PRD public data, the transits must line up with the RV amplitudes, OR, if it is at a DIFFERENT DISTANCE, the system has to be proven STABLE!

          • Harry R Ray August 15, 2016, 15:13

            I ALSO forgot Tomei, Voght, Butler et al! If they were to COMBINE the PRD public data with THEIR OWN, they could “scoop” PRD as they have scooped the HARPS teams SO OFTEN IN THE PAST, with many unfortunate FALSE POSITIVES in the past.