Planetary Habitability 3: Moons

Again assuming a star system containing a habitable zone (HZ) in the first place, there might be a situation in which life can evolve even if all the planets orbiting inside the HZ are gas giants: Namely, if at least one of them has at least one large, rocky moon.

I don't know much about the problems associated with moons in this respect. Does anyone feel up to supplying an overview-post along the lines of the two previous threads (Stars, Planets)?

edited 21st Dec '10 12:00:10 AM by kassyopeia

It would be interesting if you had an Endor or Pandora type of thing going on - teh planet is a gas giant, but the moon is sufficiently-sized to be a planet in its own right - only that it's orbiting said gas giant, instead of havin it's own orbit.

Sure make the sky quite interesting to look at...

One problem that you might have is that at least with respect to Jovian gas giants are that their moons are tidally locked to their parent planet. Which means that at least part of this moon is never going to see much if any in the way of daylight. When on the parent planet's dayside you're going to be facing away from the sun but as you go around to the night side the planet blocks the sun. You might get twilight at the first quarter and third quarter positions, I'm not sure. It would probably depend on the relative size of the parent planet in the sky.

Luna is slightly odd case because its the largest satellite with respect to it's (non-dwarf) parent planet and it's orbit is inclined relative to Earth's rotation. Larger planets tend to make their moons behave better.

This is not to say it's not possible, just that it's going to have some quirks that you won't find on an independent planet.

[...] teh planet is a gas giant, but the moon is sufficiently-sized to be a planet in its own right - only that it's orbiting said gas giant, instead of havin it's own orbit.

Anyone know if the satellite could then have a satellite of its own? I mean, it's obviously possible in theory, but could such a situation actually come about? I have a dim memory of someone asking this question in another thread, but don't remember which one.

One problem that you might have is that at least with respect to Jovian gas giants are that their moons are tidally locked to their parent planet.

I wonder if that's really an inevitability, though. Naively, one just has to make the moon sufficiently large, its original rotation sufficiently fast, and its orbital distance from the planet sufficiently large, to avoid tidal braking of an extent that would have led to full lock after "only" a few billion years. The question is if any such combination of sufficient values is allowed by the lunar formation processes.

Which means that at least part of this moon is never going to see much if any in the way of daylight. When on the parent planet's dayside you're going to be facing away from the sun but as you go around to the night side the planet blocks the sun.

Yes, I see what you mean. If the moon is close to the planet, the eclipse-phase starts right after sunrise and ends right before sunset. If it is far from the planet, the eclipse-phase only lasts from shortly before to shortly after noon, but the day (= month) is very long. Either way, bad news for life.

( quoting from the sister-thread)

[...] you need some analogue to Jupiter that has a sufficiently strong gravitational effect to pull asteroids and comets into it otherwise your planetary biosphere is going to keep getting reset by constant heavy bombardments.

Do you know whether Jupiter makes things better or worse, in this respect, for its own moons? If it's worse, it's not a deal-breaker, I think, it just means we still need a major gas giant orbiting outside the HZ to protect the minor gas giants and their rocky moons inside.

edited 23rd Dec '10 12:21:29 AM by kassyopeia

Well if the moon is too close to a gas giant, magnetic and tidal forces will probably make it uninhabitable.

edited 23rd Dec '10 7:42:50 AM by storyyeller

What would be the effects, specifically? Are there examples of such in the solar system?

Sure. Just look at Io

Ah, Io. Good titanium mining there. I hate spiders...

Saturn doens't have such a strong field, and it has Titan, which at least has an atmosphere. There's also Europa, which may have liquid water. Place the Jovian system (or whatever you call Saturn's system) in our own HZ and perhaps you could find life on Europa (or Titan).

By the way, yes, it is totally possible that life would develop on a moon and not on a planet. Planets are basically moons that orbit the sun, already. In reality, the last best hope for finding extraterrestrial life in our solar system is on the moons of gas giants (e.g. Jupiter). Endor and Pandora are solid speculations; there are some logistic problems (like the size of the native fauna, omg!), but they don't have to do with whether life exists there at all.

Not an astrophysicist, but as far as I know, there is no fundamental difference between a planet and a moon. The former is a word for a ball of rock (gas, etc...) that orbits a star, and the latter is a word for a ball of rock (etc...) that orbits a planet. The ecology / gravity / etc. would have specific concerns, though.

edited 23rd Dec '10 3:27:29 PM by rodneyAnonymous

Sure. Just look at Io

Thanks, read the wikipedia article just now. If I interpret the information correctly, Io has certain peculiarities that further exacerbate the problems necessarily caused by proximity to the planet. Or, to put it another way, there're lots of things that can go wrong with moons.

Place the Jovian system (or whatever you call Saturn's system) in our own HZ and perhaps you could find life on Europa (or Titan).

It's "Saturnian" - not, as I half-suspected, "saturnine", fortunately.

I'm pretty sure none of the large solar-system moons would work, because they're all tidally locked. I don't think there's any combination of distances from the Sun and planet, respectively, which would result in acceptable day-night temperature differences without running into the other problems discussed above.

Not an astrophysicist, but as far as I know, there is no fundamental difference between a planet and a moon.

Yes, I think so. The definition is certainly based solely on orbital characteristics, not on physical ones. There may be some fundamental differences between the two types of bodies that stem from the different formation conditions, but even those only apply if the moon in question was created in situ and not captured at some point later on.

Quoth self in post #4:

A question about stationary orbits about gas giant moons in another thread reminded me of this question, and I think I actually learned enough in the meantime to answer it, sort of. I'm going to call the gas giant's moon "moon" and the gas giant's moon's moon "satellite" below, in an attempt to confuse both myself and you, dear reader.

- Firstly, the size of a moon's Hill sphere, i.e. that region in which its gravity dominates and in which it can thus (in principle) sustain satellites is given simply by d_hill ~ (m/M)^(1/2) D, where m and M are the masses of the moon and gas giant and D the orbital distance of the moon about the gas giant.

- Secondly, there is a rule of thumb that says that gas giants can't have moons in excess of one thousandth of their mass. Don't ask me to explain that one, I have no idea what it's based on. For what it's worth, the four Galilean moons combined, relative to Jupiter, and Titan, relative to Saturn, are closer to one ten-thousandth.

- Thirdly, there is another rule of thumb that says that for an orbit in a three-body system to remain stable in the long term, it needs to be ten times closer to its focus than it needs to be to remain just stable in the short term.

If I put those together and call the region in which the planet can sustain long-term orbits the "effective sphere", then its size should be roughly given by

d_eff ~ (1/10) d_hill ~ (1/10) ((M/1,000)/M)^(1/2) D ~ D/300

We now have the maximum orbital distance of the satellite in terms of nothing but the orbital distance of the moon. I've thought of three ways to get an upper bound for the latter, based on the assumption that we want the moon to be habitable, per the thread's title.

- Firstly, one can simply look at the Jovian and Saturnian systems and observe that there are no major moons beyond ten million kilometres. Instead, there's irregular moons in irregular orbits and with irregular rotations. Doesn't sound like a good place to live.

- Secondly, for evolution to have time to occur, the sun can't be much bigger than our Sun. And for the gas giant to remain a planet and not become a brown dwarf, it can't be much bigger than ten Jupiters. That means a mass ratio of (2*10^30 kg)/(10*2*10^27 kg) ~ 100, and if we want the moon to be in the HZ, the gas giant needs to be at a distance of ~1 AU from the sun, which gives us everything we need for a second "effective sphere" calculation, for the sun/gas giant/moon system:

D_eff ~ (1/10) D_hill ~ (1/10) (M/(100 M))^(1/2) (1 AU) ~ 0.01 AU ~ 10^6 km

- Thirdly, all of Jupiter's and Saturn's major moons are in tidal lock, so it seems reasonable to assume the same here. However, the day-length can't be much longer than an Earth-day for the temperature fluctuations not to become excessive (where "excessive" would mean anyting approaching 100 Kelvin, the difference between melting and boiling water). Re-using the maximum mass from above, the orbital period (and thus day-length) is given by

(2 pi)^2 D / P^2 = G M / D^2 ===> P ~ 2 pi (D^3 / (G M))^(1/2)

Trying out the previously obtained values for D of ~10^7 km and ~10^6 km, that yields 10 weeks and 2 days, respectively, lending support to the lower value.

So, assuming habitability, a reasonable guess for d_eff would be (10^6 km)/300 ~ 3,000 km, which is obviously no good at all for a satellite orbit about an Earth-sized moon. The higher value for d_eff from the first method gives 30,000 km, which I think would be more or less alright for Earth - it's "officially" still inside the atmosphere (exosphere), but the density at that height is close enough to that of the interplanetary medium for that not to have much practical significance.

All in all, I'm thinking a third-order solar system orbit (sun -> gas giant -> moon -> satellite) may be marginally compatible with the moon being habitable, if one tweaks the various input values to or a little bit beyond their commonly accepted limits. Hope someone finds this useful.

edited 27th Aug '12 1:17:26 PM by kassyopeia

It's entirely possible to find a gas giant around the liquid water zone. Why, we once discovered a gas giant with a mercury like orbit.

You're going to run into tidal issues what with the shear gravity of a jovian planet but that's a good thing. Tidal stress stirs the pot a bit. Mixes up the trace elements into the water.

Hmm...maybe a near solar jovian moon is a more ideal planet for life than a stand alone planet. Maybe.

If the moon is in tidal lock with the gas giant, the only tides result from eccentricities in the orbit which modulate the height of the stationary tidal bulge. At the orbital distances derived in the previous post (a few million kilometres) and the usual level of eccentricity (a few percent), you don't get to have really impressive tides. Ten times the lunar tides on Earth would be the best (worst) you could expect, I'd estimate.

ETA: If one does manage to give the moon a satellite per my previous post, that one would have to be very close and, if it's of any significant size and not just some non-spherical hunk of rock, would be the more likely dominant source of tides.

ETA2: The main remaining question mark regarding the habitability of gas giant moons, for me, is an issue Knightof Lsama raised in one of the companion threads to this one ("Planetary Habitability 2: Planets"): Gas giants have a much greater capacity to capture passing asteroids and comets than rocky planets do. Does this put their rocky moons at greater risk of extinction-level impacts? I haven't a clue, myself.

edited 28th Aug '12 9:22:48 AM by kassyopeia

From what I can gather, the inner solar system was more violent but became calmer as the orbits opened up and material either collected into a small number of bodies or was thrown to the outer solar system. So in the case of a jovian moon the collisions would be more frequent but quiet down as time wore on.

@Regarding the moonception question: I asked an actual astrophysicist about this, and I believe he said that while it's possible, if absurdly unlikely under Newtonian theory for a recursive moon to have a stable orbit, in Einsteinien theory, it's not possible.

I read the 'pedia article on Callisto the other day - one of Jupiter's four major moons and, due to not partaking in the tidal resonance pattern of the other three, geologically inactive. It mentions that "[t]he ancient surface of Callisto is one of the most heavily cratered in the Solar System. In fact, the crater density is close to saturation: any new crater will tend to erase an older one." I'm thinking this pretty much implies that the shielding effect a Jovian-scale outer planet has on the inner solar system doesn't apply to its own moons. But that still leaves the possibility of having a Jovian-scale inner planet with habitable moons as long as there is second Jovian-scale planet somewhere farther out as well. Or so it seems to me, at least.

The "absurdly unlikely" answer meshes nicely with the back-of-the-envelope calculations I showed above. Did they explain why Relativity would rule out such a thing? I see a qualitative difference between primary and secondary orbits (those about stars and those about planets), but secondary and tertiary orbits (about planets and about moons) don't seem so very different, conceptually speaking.

That was actually a long time ago, I really can't remember. I'm not even sure about the "Einsteinien theory forbids it part", though I believe that's what he said. I remember he drew out the math, but it was really complicated stuff and I hadn't taken an actual physics class yet.

edited 19th Sep '12 6:05:05 AM by Archereon

There's not any qualitative difference between different types of orbits - they're all the result of inertia and gravitational attraction balancing out.

I also can't immediately see why Relativity would rule out fourth-or-deeper-level bodies, apart from the fact that the scale changes involved would be ludicrous. (e.g. stars are orbited by planets are orbited by moons are orbited by Big Dumb Objects.)

For anyone curious, the big change Relativity makes to orbital mechanics is gravitational radiation, which has an approximate equation here. (The other big change is that gravity has a finite speed, but I have no idea how you'd get to the work in terms of the math.)

edited 20th Sep '12 10:09:48 AM by Yej

As I remember, it had something to do with the conditions necessary for the recursive moon to be in a stable orbit had to be so perfect that even the slightest disturbance would destabilize the orbit. Or a slight change to the math.

Different in that you can generally ignore the rest of the galaxy/universe when considering planetary orbits about a star, whereas you cannot forget about the rest of the solar system when considering lunar orbits about a planet. That's what I meant by "qualitative". Apologies for confusing you.

At this scale, the (quantitatively) big change is that geodesics in a Relativistically curved spacetime are only approximated by the traditional Keplerian ellipses predicted by the Newtonian approach. At sufficiently small scales or over sufficiently long timescales, the differences become measurable, though, as is famously the case for the perihelion precession of Mercury.

As far as I know, gravitational radiation is always negligible for the sorts of orbit in question, simply because non-gravitational phenomena like electromagnetic radiation pressure and the Poynting-Robertson effect provide higher-order corrections. Whether that still applies to something extreme like a Jovian-scale planet on a Mercury-scale orbit, I can't say.