How different would life be (that is calendar, seasons, tides, solar days and other natural events) on a moon of a moon of a planet? Is this setup even possible?

  • 3
    $\begingroup$ I have my doubts that this is possible. Planets have moons because they are so far away from the Sun that the influence of the Sun on the moon orbit is not enough to disrupt it. But the planet may be so distant because the Sun is really massive; a moon so far from its planet would be not pulled strongly enough to orbit the planet. The only possibility would be that the secondary moon is a really small moon, which would not have a significant gravitory field and would make life and colonization difficult. $\endgroup$
    – SJuan76
    May 2, 2015 at 8:02
  • $\begingroup$ possible duplicate of How would having multiple moons affect tides? $\endgroup$ May 2, 2015 at 15:50
  • $\begingroup$ Have you looked at the scores of other questions about life on a moon? $\endgroup$ May 2, 2015 at 15:51
  • 1
    $\begingroup$ is it possible was discussed here not to long ago. You ought to look up the results of that, and then choose a specific geometry and then ask about what that would mean to the inhabitants. $\endgroup$
    – JDługosz
    May 2, 2015 at 20:32
  • 3
    $\begingroup$ related: Do moons have moons? $\endgroup$
    – SF.
    May 2, 2015 at 22:51

4 Answers 4


The largest moon in our solar system is Ganymede. Its mass is about $1.5 \cdot 10^{23}~\text{kg}$, comparable to Mercury ($3.3 \cdot 10^{23}~\text{kg}$) and Mars ($6.4 \cdot 10^{23}~\text{kg}$). So, although Ganymede has no moon, it is large enough to have a moon. It also has a molten core and some atmosphere.

Moons are normally much smaller than what they orbit, not counting Pluto, since it is no longer considered a planet. The heaviest moon compared to its planet is our moon, which is about $\frac{1}{80}$ the mass of Earth ($7.35 \cdot 10^{22}~\text{kg}$ vs. $5.97 \cdot 10^{24}~\text{kg}$).

Assuming a moon $\frac{1}{50}$ the mass of Ganymede, you would have $3 \cdot 10^{21}~\text{kg}$ worth of moon - we'll name it Jimmy. Jimmy's surface gravity will be around $8%$ of Earth's gravity, so Jimmy won't have any atmosphere left. It won't be very Earthlike in other ways either.

Unless Jimmy's orbit is quite close to Ganymede, it will not be a stable orbit over the long term as it will be pulled at by Jupiter and the other planets. Jimmy will be tidally locked to Ganymede, but in close orbit, its day will also be fairly short. $24~\text{h}$ would be entirely possible. Frequent total eclipses would be likely as well, from Ganymede as well as Jupiter. Jimmy will have strong Jovian tides twice be per as well as much weaker solar tides.

Jimmy's seasonal durations will match Jupiter's orbit. But having no atmosphere, your probably won't really care since you will have to spend all of your time inside.

Jimmy is probably about as large as a moon of a moon would ever be, not considering moon captures and catastrophic moon splitting events. And with Jupiter close by, Ganymede will not be successful retaining such moons long term.

Even though Jimmy is close enough to Ganymede for Ganymede to retain Jimmy, the orbit of Jimmy will not be very stable. Over a large period, Ganymede will eventually lose Jimmy. As Jimmy will eventually reach an orbit that favors capture by Jupiter. But I would expect a run of over 100 million years to be possible, if you can avoid resonances that destabilize Jimmy's orbit more quickly.

  • $\begingroup$ About seasonal durations, you may want to compare Are there seasons on Luna?. $\endgroup$
    – user
    May 3, 2015 at 14:47
  • $\begingroup$ Seasons would likely exist primarily in a technical sense only, i.e., the most likely orbital plane would tend to be close to the orbit of the Jovian planet. As a practical matter, seasons would be pretty much a non-issue, just as they are on our moon. $\endgroup$ May 4, 2015 at 0:38

A moon orbiting another moon is probably not a stable configuration.

"Tidal forces from the parent planet will tend, over time, to destabilize the orbit of the moon's moon, eventually pulling it out of orbit," says Webster Cash, a professor at the University of Colorado's Center for Astrophysics and Space Astronomy. "A moon's moon will tend to be a short-lived phenomenon."


To have any chance of stability, the distances between the planet and the first moon would have to be rather large, so that the planet would not have much gravitational influence on the moon's moon. Probably the planet needs to be also far away from the sun, so that the sun does not disturb the large orbit of the moon around the planet; and the planet has to be huge to hold onto the moon at that distance.

That means tides will be mostly influenced by the first moon, not the planet. Seasons, day cycle, etc will also be largely unaffected by the planet, aside from the possible occasional solar eclipse.

One major problem is the atmosphere: To be able to hold onto an atmosphere, the moon's moon needs to be either big enough to have enough gravity, or be far away from the sun, since the solar wind will strip its atmosphere away otherwise. Our moon is too small to hold an atmosphere at our distance from the sun, and it is one of the biggest moons in the solar system. Titan, bigger and further out, can hold onto an atmosphere, but its distance from the sun means it is quite cold. A moon orbiting titan would have to be rather small, close to titan, and titan would have to be much further away from saturn, to allow a stable orbit

So in summary, such a moon would probably have to be far away from the sun and its planet, and rather small, to avoid getting ripped apart or deorbited by tidal effects. The temperature would be rather cold due to the distances to the sun. If the planet is a super Jupiter / brown dwarf, the moon could get some additional heat from the thermal radiation of the planet. But not too much, because close to the planet, tidal effects would forbid a stable moon-moon orbit.


The largest possible gas giant is around 12 Jupiter's in mass (this is roughly where you start hitting the difference between a gas giant and a small star). Let's call this theoretical planet G1.

G1 is big enough that it could have another gas giant as a moon - say one similar in mass to Jupiter, let's call it G2. If G2 was far enough away from G1, it could have its own moon system, with moons similar in size and composition to earth.

Now the problem with this setup is that in order to keep the orbits stable, G2 has to be pretty far from G1 so that G2's moons aren't perturbed too much. And because of that, G1 and G2 need to be really far away from their parent star so that G2 doesn't get perturbed.

The end result is that while you might be able to get a planetary/moon/moon setup with an earth-like mass and composition, you can't put it in the habitable zone of the star because that's too close for it to be stable.

  • $\begingroup$ It won't be in the solar habital zone, but would rely on tidal heating like Europa etc. $\endgroup$
    – JDługosz
    May 2, 2015 at 20:34

Fun Facts

In our solar system, all planets have moons except Venus & Mercury. The larger planets exhibit an odd commonality in that the mass of their moon system $ m_{moon system} $ ~ $ \frac {m_{planet}}{10000} $.

For other SE questions, I calculated the minimum mass of a body that could retain water in its atmosphere when its temperature was similar to Earth's at $ m_{body} $ ~ $ \frac {m_{Earth}}{3} $.

The closest a moon comes to the mass of its parent body is the Pluto - Charon system with a mass ratio of $ m_{Charon} $ ~ $ \frac {m_{Pluto}}{10} $.

Interestingly the maximum mass of an object in a planet's or moon's L4/L5 without making those locations unstable is $ m_{small} $ ~ $ \frac {m_{large}}{10} $

What would my system look like?
Moon of a moon:

Mass ~ 2 x 10^24 kg ~ 1/3 Earth
Radius ~ 4,400,000 m
Density ~ 5.5 g/cm^3


Mass ~ 2 x 10^25 kg ~ 3x Earth
Radius ~ 9,500,000 m
Density ~ 5.5 g/cm^3


Mass ~ 2.2 x 10^29 kg ~ 100x Jupiter
Radius ~ 331,800,000 m
Density ~ 1.24 g/cm^3

There are several problems with this. As mentioned above,

  1. At 80x Jupiter's mass you start getting fusion in the core - but we can fudge the above rule and just use a planet that has approached but not reached the fusion limit.
  2. At Jupiter's mass, adding mass doesn't increase the planet's diameter, it increases its density, so the size is bogus.
  3. Tidal forces will tend to destabilize the configuration, so you'll need the system far away from the planet (tidal forces decrease at a $ \frac {1}{r^3} $ so distance is your friend).
  4. You also want tidal forces from the star to be small, so make the parent star hot and place the primary planet far away from the star.
  5. Because the problem with tides messing up orbits and unstable orbits, you may just wish to put the secondary in the primary's L4/L5. This places them farther apart than they would be if the secondary orbited the primary, but it would make the orbit more stable. Max mass for the secondary would be (as stated above) $ m_{secondary} $ ~ $ \frac {m_{primary}}{10} $

Not the answer you're looking for? Browse other questions tagged .