≡ Menu

Getting Closer to an Exomoon Detection

Finding moons around extrasolar planets is an invigorating quest. After all, at least three moons around gas giants right here in our own system — Europa, Enceladus and Titan — are considered of high astrobiological interest. What about gas giants in the habitable zone of some distant star? The image below shows what a moon of such a planet one might look like, as imagined by astronomer Dan Durda (Southwest Research Institute). Could such worlds be?


As we learn more, bear in mind that the hunt for ‘exomoons’ has already begun. The CoRoT spacecraft is searching for transit timing variation signals (TTV) — variations in the time it takes a planet to transit its star — as described by Sartoretti and Schneider in a 1999 paper. David Kipping (University College London) has been developing a second method called transit duration variation (TDV), which works in conjunction with the first. The TDV signal is brought about by velocity changes as the planet/moon ‘system’ is observed over time, the result of both planet and moon orbiting a common center of mass.

Dr. Kipping now offers a further take on these two effects, which should be able to detect and characterize an exomoon when used in tandem. In the new paper, the astronomer breaks transit duration variation itself into two parts, one based on velocity (V), the other on what he calls the transit impact parameter (TIP).

…an exomoon around a transiting exoplanet should induce a transit duration variation effect with two dominant components. One of these components is due to the moon altering the velocity of the host planet, which we label as the V-component. The second constituent is due to the impact parameter of the transiting planet varying as a result of the moon’s presence, which we label as the TIP-component.

The analysis of these combined effects should allow astronomers to tell the difference between moons in a prograde or retrograde orbit, because the TIP component acts constructively with the V-component in prograde situations and destructively with it in retrograde orbits. The effect is to boost the exomoon detection method via TDV by about ten percent in magnitude. All this helps us understand more about how the moon might have been formed and tells us something about the stability of its orbit.

The key point is that we have signals thrown by these effects that should be observable today, and will certainly be so as our instrumentation improves:

We therefore propose that it should be possible for future observations to not only detect an exomoon and determine its mass, but also provide a confident deduction of the sense of its orbital motion. Although this determination will likely require photometry at the limit of planned missions, it seems likely that once an exomoon is detected a more in-depth investigation would be able to answer the question of orbital sense of motion conclusively.

There are caveats here, including the fact that the calculations are effective for a planet-moon system in which the plane of the two objects’ orbits is aligned with the star-planet orbital plane — large exomoon inclinations would be disruptive. But it’s interesting to note that in Kipping’s view, exomoons at low inclination angles should be observable in the lightcurve during any planet-moon eclipse. That would be an exciting confirmation of the first detection of a moon around a distant extrasolar planet.

The paper is Kipping, “Transit Timing Effects due to an Exomoon II,” accepted for publication in Monthly Notices of the Royal Astronomical Society and available online.

Comments on this entry are closed.

  • kurt9 April 27, 2009, 13:08

    The problem with the habitability of these exomoons is that gas giants tend to have rather intense radiation (Van Allen) belts around them. Jupiter’s radiation belt is so intense that it may not even be possible for “biological” humans to visit the Jovian system. Saturn has these belts, but they are somewhat less intense than those of Jupiter. This suggests that the intensity of these radiation belts is proportional to the size of the gas giant. The larger the gas giant, the more intense the radiation belts. Biological life (at least as we know it) is not going to get anywhere near a gas giant that is 3 times the size of Jupiter.

  • FrankH April 27, 2009, 15:23

    Other problems with moons around gas giants in a HZ:
    1 – the gas giant probably migrated from the colder parts of its star system… so the moons will have a large amount of water. If the moons are not massive enough to hold onto water in a HZ (at the very least 2x the size of Mars) all that water will go away and we’ll end up with a smaller rocky, uninhabitable moon.

    2- there’s at least one article that has simulated moon growth around gas giants. They found that the maximum size is around Mars sized or smaller. Mars is too small to be habitable, even though it’s at the outer edge of the Sun’s HZ.

    So even if the gas giant has a fairly benign radiation environment (like Saturn, thanks to its rings) it’s pretty unlikely to have a suitably sized, habitable moon.

    Gas giants in a star’s HZ are just bad news for Earth-like life.

  • kurt9 April 27, 2009, 16:28

    Good points, FrankH.

    Jupiter and Saturn are tell-tale for your point #2. Jupiter has 3 times the mass of Saturn, yet its largest moons are not any bigger than Saturn’s Titan. It just has four of them, rather than one.

    Titan has a thick atmosphere, despite being smaller than Mars. Titan keeps its atmosphere because it is too cold for it to excape. If the giant migrates inward, how long can a Mars-size moon hang on to its atmosphere (and water) before it boils off into space and the moon becomes, well, Mars-like? Does anyone have an idea what time scales are involved in the inward migration of the gas giants.

  • bigdan201 April 27, 2009, 18:34

    points taken, but we still shouldnt overlook the possibilities of exomoons. im sure theyre out there, and i also think that we shouldnt make too many sweeping assumptions. a large moon around a gas giant in the HZ could be a location for life, especially if it had an atmosphere like Titan does. it could be outside the radiation belts, it could have higher gravity than the moons in our solar system, or even earthlike gravity/mass, etc. we cant rule out possibilities.. and even if we dont find life on exomoons, an exomoon could be a good destination for space-faring humans in the distant future.

    if some of the exoplanet gas giants weve been finding are more massive than jupiter, then wouldnt it follow from there that they could have more massive moons?

  • Daniel April 27, 2009, 18:35

    its not all moon around jupiter and saturn there is inside of this gas giant radiation belts…. for exemple gaminede (has your own magnetic field) and casllisto its outiside of jupiter radiation belt,and one of this moons gaminede is bigger that the mecury,and some of this extrasolar gas giants a bigger that jupiter,then maybe they form big earth-size moons…

    who kwons! we have a lot suprises in exoplanets sciences,and i think that we go to have much more…

    i think that we can´t understimate extrasolar earth-size moon on the habitable zone for while

  • David Kipping April 27, 2009, 18:45

    Thanks so much for your interest in the exomoon work again! I did not expect this article would feature on your site but I am very pleased that you believe your readers will find this work of interest.

    I think the article sums up the results very well and I enjoyed reading it. In regard to the misaligned moons issue, I would offer that any misaligned moon, like Triton around Neptune for example, would generate a strongly enhanced TDV signal. Indeed, in many ways, misaligned moons are the ideal case but one cannot build a model around such exceptional circumstances. Triton would effectively pull Neptune to lower and higher latitudes with respect to moving across the Sun’s face during a transit. This effect is still in effect for moons where the orbit is co-aligned to the planet-star plane, but much less pronounced.

    Then again, it would seem that Triton and indeed our own Moon formed through more unusual circumstances than the Galilean moons. For example, Canup and Ward (Nature, 2006) find that there is an upper mass limit on moons of around 2*10^(-4) of the host planet, where it is understood the moons formed from the giant planet accumulating gas and rock-ice solids from solar orbit. However, if the moon is captured then there is no restriction on the potential mass, except for dynamical stability. Barnes & O’Brien did a nice job in 2002 of showing that a habitable 1 Earth mass exomoon could be dynamically stable around a Jupiter-like planet for any main sequence star masses above about 0.45 solar masses. Retrograde and highly inclined orbits can also be explained by captured moons and so perhaps it is not so outlandish to consider detecting these more unusual bodies since they offer such large signals.

    Thanks again for your interest!

  • Darnell Clayton April 27, 2009, 20:03

    While radiation may not be a problem (as a Moon with a strong magnetic field could easily repel its planetary parents radiation belt) the atmosphere factor might be.

    We would also have to deal with the duration of daylight, especially with one side receiving a lot more than the other (for a lot longer), which may not benefit too many organisms.

  • kurt9 April 27, 2009, 20:11

    …upper mass limit on moons of around 2*10^(-4) of the host planet…

    Given that Jupiter is about 300 times the mass of the Earth, a gas giant big enough to make an Earth-sized moon will have to be 15-16 Jupiter masses, which makes it a small brown dwarf. What is the discovery rate of brown-dwarfs compared to gas giant planets? Is there an upper limit to the size of a planet that is likely to form in a solar system? Perhaps the star has to be of very high metallicity to make such a big planet.

  • Dave Moore April 27, 2009, 20:51

    Kurt9, I wouldn’t be so quick to dismiss life in a high-radiation environment. Deinococcus radiodurans lives in the water of nuclear reactors.

    Also, the Galilean satellites don’t have an atmosphere to moderate the surface levels of radiation. And even if the surface levels of radiation were high, the levels some 10 of meters below an ocean or solid surface would be a lot lower.

    Life would simply evolve with more effort devoted to genome maintenance. Deinococcus radiodurans can surrvive having its DNA shattered (not just the odd break or defect). It is polyploid, and by comparing its 4 copies of DNA with each other, it can reassemble DNA and start growing again in 24 hours.

  • spaceman April 28, 2009, 7:18

    The detection of exomoons would be a fascinating development in extrasolar planet science. I predict that such a discovery will happen quite soon via the microlensing technique. In any case, I have a question about the color of the gas giant planets that these exomoons may be orbiting. In our solar system Jupiter and Saturn, and the gaseous atmospheric components of Uranus and Neptune are different hues of the color blue. Is it possible, or even likley, that there are gas giants that are green colored.

  • ljk April 28, 2009, 9:18

    I bet Europa’s ice crust is very good at blocking out most of the radiation
    from Jupiter. Though the moon may need a little to get life evolving.

    Experts like Richard Greenberg of Unmasking Europa are saying that
    not only may there be life on Europa, but complex life at that.

    We so need to LAND on that moon.

  • Benjamin April 28, 2009, 10:14

    I recently listened to Marty Hidas talk in a lecture at the University of Sydney about his work on exoplanet transits, and one of the things he mentioned is that there is a certain periodic variation in the brightness of a star due to the phase of planets around it: that is, some amount of light is reflected from the planet, and at different points in its orbit it will reflect different amounts toward the Earth, depending on its size (determined easily by a transit), orbital parameters (transit + Doppler shift) and albedo. This is used to work out things like the atmosphere of a planet, and has popped up once or twice on this blog, I believe, for that reason.

    I wonder if a very sensitive instrument, more advanced than, say, Kepler or CoRoT, might be able to detect a variation in that due to moons around the planet in question.

    (Re-reading the above I tend to think that this is likely to be difficult to the point of impossible, and I perhaps shouldn’t try and post after midnight or I’ll say more stupid things!)

    It seems more likely that the moons will alter the transit parameters by wobbling the planet, but still, we’re talking about some very careful measurements over many years.

  • kurt9 April 28, 2009, 11:34

    It is polyploid, and by comparing its 4 copies of DNA with each other, it can reassemble DNA and start growing again in 24 hours.

    I knew a biochemist in the late 80’s who proposed using biotechnology to make this capability available for us humans. This seems a useful capability to have if you plan to live in space.

  • bigdan201 April 28, 2009, 22:23

    thats right, ganymede does have a magnetosphere.. another reason why we cant be too quick to cast doubts on life or habitability on exomoons.

  • Bernd Henschenmacher May 3, 2009, 13:57

    There is another possibility, which could make even mars-sized moons habitable. If the moon orbits its host planet in a slightly noncircular orbit, tidal heating might provide an energy source for plate tectonics to develop. So the conditions on these moons could be more earthlike than marslike. Given the fact, that ganymede possesses its own magnetic field, larger monns could develop stronger magnetic fields. So even gas giants with smaller masses could harbor habitable moons. Caleb Scharf wrote an article about the prospects of habitable moons. You can read it: http://arxiv.org/abs/astro-ph/0604413

  • Ronald May 4, 2009, 5:35

    Assuming that about 0.2 – 0.3 earth mass seems to be the lower limit for a planet to hold on to a ‘decent’ atmosphere, according to the above-mentioned ratio, that would require a roughly 3-5 Mj gas giant.

    Question is, assuming a correspondingly dense atmosphere resp. strong magnetosphere, how far from the gas giant would such an exomoon need to be to experience roughly earthlike radiation levels? (for instance in comparison with Ganymede, Titan, Europa, …)

  • spaceman June 3, 2009, 6:17

    Here is a link to a microlensing presentation given by astronomer Joachim Wambsganss kipac-prod.stanford.edu/collab/seminars/acks/talks_spring_2009/090226/at_download/file

    in which the actual detection of an exomoon via microlensing is being hinted at?

    Apparently, as mentioned on page 46, an exomoon has been detected and they are just waiting for some reason to publish the paper describing it!

    Has anyone else heard of this?

  • ljk June 20, 2009, 14:43

    A Survey for Satellites of Venus

    Authors: Scott S. Sheppard (Carnegie, DTM), Chadwick A. Trujillo (Gemini)

    (Submitted on 15 Jun 2009)

    Abstract: We present a systematic survey for satellites of Venus using the Baade-Magellan 6.5 meter telescope and IMACS wide-field CCD imager at Las Campanas observatory in Chile. In the outer portions of the Hill sphere the search was sensitive to a limiting red magnitude of about 20.4, which corresponds to satellites with radii of a few hundred meters when assuming an albedo of 0.1. In the very inner portions of the Hill sphere scattered light from Venus limited the detection to satellites of about a kilometer or larger. Although several main belt asteroids were found, no satellites (moons) of Venus were detected.

    Comments: Published in July 2009 (Sheppard, S. and Trujillo, C. 2009, Icarus, 202, 12-16.)

    Subjects: Earth and Planetary Astrophysics (astro-ph.EP); Instrumentation and Methods for Astrophysics (astro-ph.IM); Solar and Stellar Astrophysics (astro-ph.SR)

    Cite as: arXiv:0906.2781v1 [astro-ph.EP]

    Submission history

    From: Scott S. Sheppard [view email]

    [v1] Mon, 15 Jun 2009 21:40:40 GMT (32kb)


  • ljk July 1, 2009, 14:37

    Exomoon simulations

    Authors: A. E. Simon, Gy. M. Szabó, K. Szatmáry

    (Submitted on 30 Jun 2009)

    Abstract: We introduce and describe our newly developed code that simulates light curves and radial velocity curves for arbitrary transiting exoplanets with a satellite. The most important feature of the program is the calculation of radial velocity curves and the Rossiter-McLaughlin effect in such systems.

    We discuss the possibilities for detecting the exomoons taking the abilities of Extremely Large Telescopes into account. We show that satellites may be detected also by their RM effect in the future, probably using less accurate measurements than promised by the current instrumental developments. Thus, RM effect will be an important observational tool in the exploration of exomoons.

    Comments: 5 pages, 2 figures with 9 figure panels, accepted by EM&P

    Subjects: Earth and Planetary Astrophysics (astro-ph.EP)

    Cite as: arXiv:0906.5442v1 [astro-ph.EP]

    Submission history

    From: Gyula Szabo [view email]

    [v1] Tue, 30 Jun 2009 08:56:01 GMT (1626kb)


  • bigtex January 6, 2010, 23:25

    Is it true that the current theory as to why Mars no longer has water/thick atmosphere is because it’s core cooled off in a few hundred million years, and then it lost its magnetosphere and then its atmosphere was evaporated over time?

    If that’s correct, then perhaps if an exomoon is “safely” within the magnetosphere of its gas giant planet, perhaps it can maintain an atmosphere without needing a magnetosphere of its own…