≡ Menu

On Habitable Worlds and Their Moons

One of the problems with building a backlog of stories is that items occasionally get pushed farther back in the rotation than I had intended. Such is the case with an article in Astrobiology Magazine that talks about how much of a factor a large moon may be in making a planet habitable (thanks to Mark Wakely for passing the link along). It’s an interesting question because some have argued that without our own Moon, the tilt of the Earth’s axis, its ‘obliquity,’ would move over time from zero degrees to 85 degrees, a massive swing that would take the Sun from a position over the equator to one where it would shine almost directly over one of the poles.

The resulting climate changes would be severe, potentially affecting the development of life. The thinking is that just as the direction of the tilt of a planet varies with time — astronomers say that it ‘precesses’ — so does the orbital plane of the planet. The gravity of a large moon like ours affords a stabilizing effect by speeding up the Earth’s rotational precession and keeping it out of synch with the precession of the planet’s orbit. When rotational and orbital plane precession are synchronized, the obliquity begins to change chaotically. The Moon’s job, then, is to keep the two out of synch, minimizing the kind of fluctuations that would play havoc with life on the planet.

Image: The image shows Earth’s axial tilt (or obliquity), rotation axis, plane of orbit, celestial equator and ecliptic. Earth is shown as viewed from the Sun; the orbit direction is counter-clockwise (to the left). Credit: Wikimedia Commons/Astrobiology Magazine.

Jason Barnes (University of Idaho) and colleagues are behind the latest work (presented at the most recent American Astronomical Society meeting) which suggests another interpretation, arguing that the effect of the Moon on the Earth’s obliquity has been overstated — the Moon is not in fact crucial for the development of life. We can hope this bears out, because estimates of how many terrestrial planets will have a substantial moon get as low as one percent. Most of these worlds, then, under previous thinking, would experience huge changes in their obliquity, pointing toward a ‘rare Earth’ conclusion.

But that thinking is under challenge. Recent work by Sebastian Elser (University of Zurich) argues that the chances of large moons for such planets are as high as 10 percent. And Barnes’ team contrasts the gravitational effects of the Moon with those of other planets orbiting the Sun. The conclusion: The Moon does provide some stability to our planet, but the pull of Jupiter and, to a lesser extent, other planets orbiting the Sun would tend to keep the Earth’s obliquity swings in check. In fact, Barnes has determined that the Earth’s obliquity without a moon would vary only ten to twenty degrees over half a billion years. That’s enough to cause major climate changes, but it would “…not preclude the development of large scale, intelligent life,” Barnes adds.

I was intrigued by the team’s findings about planets with retrograde motion in their orbits. With or without a moon, the obliquity variations of a planet in this configuration — spinning in the opposite direction from their star — should be smaller than those orbiting in the same direction as the star. Barnes believes that the initial rotation direction of a planet should be random, voicing his suspicion that “whatever smacks the planet last establishes its rotation rate.” If this is true, the odds on retrograde precession are 50/50, lengthening the odds for relatively modest obliquity.

So we get help from the spin of a planet where it is retrograde, and also the combined gravitational effects of other planets in the system that help to reduce the planet’s tilt. If Barnes and team are right, then worlds lacking a large moon are still very much in the running for the development of life, a stability that he reckons may account for 75 percent of the rocky planets in the habitable zone. We’re a long way from confirming that idea, but it’s refreshing to hear this assertion that a large moon may not be a sine qua non after all, given how little we know about exoplanetary moons and the likelihood of their emergence in the right size range.


Comments on this entry are closed.

  • andy August 18, 2011, 13:59

    Talking of habitable worlds, there’s a couple of interesting papers that came up on the arXiv today:

    The HARPS search for Earth-like planets in the habitable zone: I — Very low-mass planets around HD20794, HD85512 and HD192310

    A Habitable Planet around HD 85512?

    The planet in question, HD 85512b has a minimum mass 3.6 times the Earth’s mass. It seems quite strongly irradiated but there may be scenarios under which it could support habitable conditions. Certainly impressive progress is being made in the field of radial velocity planet searches!

  • kurt9 August 18, 2011, 14:43

    Mars is the test of this. Mars obliquity is currently around 24 deg, about like Earth. Does Mars obliquity oscillate between zero and 85 deg? Or does it oscillate over a narrower range? There should be evidence of this in Mars’ geology.

    Mars is also the test of plate tectonics and if a big moon forming impact is necessary to initiate plate tectonics and if a big moon is necessary to sustain plate tectonics. I think plate tectonics is necessary for a bio-sphere (this is the only part of “Rare Earth” I think is correct).

    We can learn a lot about exoplanets by studying Mars and Venus.

  • henk August 18, 2011, 16:48

    maybe but we know for 100% that without a moon this planet will be different. Maybe other kind of life evolve on such a planet that can protect themself agains the strong climate. you can see that animals adept if the climate change on this planet. Some people think that every planet with life will be a beautiful world like this one. Other worlds can be very ugly. where a human never want to life and that the animals are very strange. It does not have to be a starwars planet. They always make a earth clone.

  • Rob Henry August 18, 2011, 18:43

    Kurt9, I think that you are being too harsh on “rare Earth” factors and too lenient on the “rare Earth hypothesis”. Earth has many quirks that have been legitimately promoted as outside possibilities for being important to the development of intelligent life. All that we can reject with high confidence is that most of them turn out to apply as per Brownlee and Ward.

  • Istvan August 18, 2011, 22:23

    It’s nice to see some solid peer-reviewed research work substantiate this line of thinking, but I’m amused to observe that I’ve been reading science fiction for decades in which more than one author has assumed (based on I know not what earlier suppositions) that Earth’s “oversized Moon” was a key factor in the emergence of life on or sustained habitability of our world.

  • Alex Tolley August 18, 2011, 23:21


    I attended a lecture on Mars at SETI a year or so back where the speaker indicated that there was evidence of very large obliquity changes in Mars rotation.

  • Enzo August 19, 2011, 6:26

    Strictly speaking, the following is slightly off topic, but it does affect the type of life that can exist on a earth-like planet and so, in some ways, its habitability.
    The other day I was wondering how the size of the super-earth can affect the depth of an ocean.
    The first consideration is that, due to its higher gravity, a super-earth should be flatter than Earth. This seems pretty straightforward but the implication is that any ocean would be wider. You can imagine the effect on Earth if the difference between the lowest and highest point was reduced by a few kms.

    The other thing I did was to try to calculate the relationship between depth of an ocean and the radius of a planet that had a given % of its volume (called x) as liquid water. That gives :
    4/3*pi*((R+h)^3-R^3) = x*4/3*pi*(R+h)^3

    where h is the depth of the ocean and R the radius of the planet without the little oceanic skin. The term on the left is the volume of the ocean and one on the right the amount of liquid water as % of the whole volume of the planet (x). Now I know that this is an approximation as not all the water would be sitting on the surface but some might be part of mantle minerals etc. but I’m no geologist.
    The equation above is a bit messy but it can be greatly simplified under the assumption that h is much smaller than R, which is a very reasonable assumption.
    Under this assumption the term on the right (R+h)^3 = ~R^3
    For the one on the left (R+h)^3-R^3 = ~3*h*R^2 as 3*h^2*R and h^3 would be much smaller than 3*h*R^2.
    After eliminating 4/3*pi from both sides, the initial equation becomes :
    3*h*R^2 = ~ x*R^3 => h =~ x*R/3
    In other words, h is proportional to the radius R of the planet.
    This also means that, if a super-earth has the same % of its volume as surface liquid water as Earth but a radius 1.4 times bigger, the ocean would be 1.4 times deeper. I believe that if we consider % of mass instead of volume the ocean will be even deeper, provided that the density of the super-earth is also higher.

    These two considerations make me think that super-earths should be more “oceanic” than Earth unless they are drier. And that affects their potential biosphere. The mitigating factor is that a super-earth with 1.4 times the radius of Earth will also have nearly twice its surface and so, less landmass in % would not necessarily mean less than Earth’s. Still, the gut feeling of super-earths as more “oceanic” remains and speculation is all you can do without much more data…..:-)

    By the way, I remember a sci-fi novel (David Brin ?) where something similar was discussed but regarding Earths, not super ones. I read this some 15 years ago and, while I don’t remember the details, the conclusion was that Earth was rather dry and alien Earths might be more likely to be water worlds.
    Does anyone remembers something like that in one of David Brin’s novels ?

  • Ronald August 19, 2011, 6:42

    kurt9: is Mars really a good test? I think that a major, possibly *the* major shortcoming of Mars is its too small mass (hardly 11% of Me) for a terrestrial planet, which is probably the main reason for its present lack of plate tectonics and could also enhance its instability.
    So Mars may be a test indeed, but rather mass-wise than moon-wise.

    Venus may be a better test for this issue: its present tilt is only about 3 degrees, it has a mass of 0.82 Me. And its rotation is retrograde. Could this be a reason for its small tilt? Is Venus’s historical tilt variation known?

  • kzb August 19, 2011, 7:51

    I don’t understand why this variablity of tilt is meant to be so bad. Agreed WE would not like it, but that’s because we’re used to Earth.

    In fact, the extreme climate changes would be a GOOD thing for driving evolution. The most common habitable world in the Galaxy may be a tidally-locked super earth with an M-class star. An ecosystem on such a world would see no significant environmental changes over billion year timescales. No challenges, therefore no evolution.

    So a bit of chaos thrown into the system via large climatic swings could only be a good thing surely?

  • Ronald August 19, 2011, 8:00

    Enzo, interesting. Following your line of reasoning, then there has to be a threshold R for a super-earth (for a given density and water %) beyond which it will always be entirely ocean covered. For a roughly earthlike water % and ‘typical’ density of discovered super-earths, at which R would this be, is it possible to give an indication?

  • Ronald August 19, 2011, 10:26

    The publications mentioned by andy are very fascinating indeed, worthy of a dedicated post (Paul). I browsed through the first one, some highlights:

    – It concerns the preliminary HARPS radial velocity results for 10 very stable and more or less solar type nearby stars, Alph Cen B and Tau Ceti among them. All ten have shown no signs of any giant planets, despite observational bias toward giants in the RV method.

    – Three of them (HD20794, HD85512 and HD192310) show a total of 6 low-mass planets, from super-earths to Neptune class ‘subgiants/ice giants’.

    – HD85512, though hardly solar type (K5V star), is very interesting because it has a 3.6 Me planet at 0.26 AU with an estimated equilibrium temperature of 298 K, that is 25 Celsius, about southern France right now!

    – HD 20794, this star is the nearby 82 Eridani at 19.8 ly, solar type G8V with a luminosity of 0.65 * solar. It is also very interesting because it has 3 low-mass planets: 2.7, 2.4 and 4.8 Me, at 0.12, 0.20 and 0.35 AU resp. (the 2nd one being uncertain). Furthermore, all three have near zero eccentricity. The only annoying thing is that even the outermost one is still too close for comfort, with an estimated eq. temp. of 388 K (215 Celsius). Can there be another low-mass planet further out in the HZ?

    – Conclusion: “This result already confirms previous indications that low-mass planets seem to be very frequent around solar-type stars
    and that this occurrence frequency may be higher than 30%”.

  • kzb August 19, 2011, 11:20

    Re the ocean theory:
    the amount of water may be a completely random event. Most of it is delivered by comets. The frequency of comet impacts may be only weakly correlated with planet size, being controlled mostly by the configuration of outer gas giants and cometary material in the system under consideration.

  • Erik Anderson August 19, 2011, 12:16

    I too have sensed that the chances of large moons orbiting terrestrial planets may turn out to be more likely than it seems. Surely an overabundance of worldlets in the very early Solar System facilitated all sorts of possibilities. But I’m not so sure how well the Earth would do without the Moon. The Sun and Jupiter sure didn’t help Mars, which is known to have toppled over from the chaos of those tidal interactions. On Earth, moreover, plate tectonics perpetually shift the balance of oceans and continents–so there’s enhanced risk of toppling from internal factors. And remember too that throughout the entire Earth’s past history, the moon was always closer than it is today–extremely so early on–so the stabilizing effect was always greater than it is today. Another benefit to early life on Earth from the moon is that ocean tides were tremendous in scope, which vigorously stirred up the primordial soup. Recommended reading: “What if the Moon Didn’t Exist?” (book title)

  • kurt9 August 19, 2011, 16:38

    “I attended a lecture on Mars at SETI a year or so back where the speaker indicated that there was evidence of very large obliquity changes in Mars rotation.”

    Well there you have it. If Jupiter can stabilize Earth’s obliquity in the absence of the moon, it can certainly do so for Mars. That it does not suggests that it would not do so for the Earth either.

    “Venus may be a better test for this issue: its present tilt is only about 3 degrees, it has a mass of 0.82 Me. And its rotation is retrograde.”

    This is another way of saying that Venus’s obliquity is 177 deg.

    There are lots of parameters that must be met for an Earth-analog to be truly Earth-like. The day-night cycle should be 18-30 hours. The atmosphere should be reasonably close to 1 atm pressure, with 20% Oxygen partial pressure. Other Earth-analogs may have a CO2-rich atmosphere with a pressure of, say, 4 atm with almost no Oxygen. Or it may have the equivalent of 1 atm of Oxygen (which would be poisonous to us). The day-night period could be as long as a month or as short as 10 hours. Either of these would suck as well. The obliquity could vary wildly as well. I think simple statistical calculations can calculate the percentage of Earth analogs being really Earth-like (such that we could live on them unenhanced biotechnologically). Such planets are probably quite rare.

    All of this in addition to Earth mass and having enough water, but not being a waterworld. It is just a hunch, but I think waterworlds are very common. Way more common than our Earth.

  • andy August 19, 2011, 18:01

    Well there you have it. If Jupiter can stabilize Earth’s obliquity in the absence of the moon, it can certainly do so for Mars. That it does not suggests that it would not do so for the Earth either.

    kurt9: of course you are neglecting that Mars is a less massive planet, has a very substantial north/south asymmetry (which will almost certainly do some very interesting things when you start perturbing the planet), experiences a weaker solar tide, and is in a different region of the solar system so experiences a different set of secular resonances (don’t assume that the relative contributions of various perturbations to the obliquity variations of Mars and Earth are going to be the same!) etc. etc.

  • Istvan August 19, 2011, 18:31

    @Ronald Thanks for the preliminary scan of those reports. Very interesting stuff indeed. Lack of evidence for a giant-mass planet at Tau Ceti is… odd, as based on all the “hot Jupiters” being found by RV I was starting to expect jovians all over the place. No evidence for a jovian yet at aCen B seems hardly a shock at all, however, due to the closeness of the binary, and the fact that a high-mass planet very close in probably could have been spotted by now.

    Any data giving us a better picture of the statistical frequencies of terrestrials (rather than jovians), I call good and useful news.

  • henk August 20, 2011, 8:43


    that is what i always like about those people. We know it is very unlike. But If this planet is like this, than it might be habitable. I think they need to change habitable zone in the liquid water zone. We do not know if a planet in the habitable zone is habitable. but we do know that a planet in the habitable zone can have water

  • Phil August 21, 2011, 0:33

    I thought the HARPS paper was a bit cagey about a Cen B actually, theres clearly some unusual RV action there, which they suggest is involved with the complicated dynamics of the system, and that will be the subject of a forthcoming paper. The subtext however, if there is one, is in the tiny RV variations that have confirmed especially the HD20794 planets – this is one super sensitive instrument. I wouldn’t rule out something very interesting in this next paper regarding a Cen B!

    Incidentally would someone like to comment on what seems to be the anomalously low mass of HD 20794 in that paper? Is it just the low metalicity?

    Anyway I’m heading OT so I’ll shut up now…

  • henk August 21, 2011, 11:02


    what paper talks about alpha centauri unusual RV action. Can you link that paper ?

  • andy August 21, 2011, 13:06

    Also another point regarding the obliquity variations of Mars, sure there is evidence for a substantially different obliquity in the past, but that doesn’t necessarily mean that the obliquity change was caused by these variations and not some other mechanism, e.g. large impacts, which Jupiter wouldn’t have a hope of stabilising against. There’s certainly good evidence that Mars has suffered some pretty major knocks in the past.

  • Paul Gilster August 21, 2011, 13:12

    henk, I think the paper Phil is talking about is “The HARPS search for Earth-like planets in the habitable zone: I – Very low-mass planets around HD20794, HD85512 and HD192310”:


    We’ll be talking about this paper further in the not so distant future.

  • henk August 21, 2011, 16:42


    Thanks, but they do not talk about alpha centauri, but still it is very interesting. Do they already found those planets or are it candidates just like in the kepler data ?

  • Paul Gilster August 21, 2011, 17:04

    henk, look on pp. 2-3 of the paper for the Alpha Centauri comments. And no, there have been no detections of planets around any of the Alpha Centauri stars yet, and no candidates have been established. All this may change soon, because there are three ongoing searches for planets in this system.

  • NS August 21, 2011, 17:31

    “No challenges, therefore no evolution.”

    The challenges may be biological though rather than e.g. climatic. It’s been suggested for instance that it was the advent of predators that drove evolution towards multi-cellular life. Had it not been for that life might have stopped evolving with a few single-celled autotrophic species.

  • Rob Henry August 21, 2011, 18:38

    Andy’s last comment sent a chill down my spine. I am afraid to do the calculation, yet I feel certain that an impact large enough to significantly change the axial tilt of Mars, could easily sterilise all life from Earth. If this has occurred several times on Mars since the late heavy bombardment ended, then Earth’s escape (given its larger capture area) can’t be all Jupiter’s doing. We must have been extremely lucky.

  • Ronald August 22, 2011, 6:30

    A bit further to the HARPS paper on the RV results for 10 very stable and more or less solar type nearby stars, that have shown no signs of any giant planets, and Istvan’s recent comment (19 August, 18:31): “Lack of evidence for a giant-mass planet at Tau Ceti is… odd, as based on all the “hot Jupiters” being found by RV I was starting to expect jovians all over the place”.

    I checked the metallicity of these ten researched stars (Simbad/Vizier and other sources) and found that 8 of them have lower metallicity than the sun, in fact 7 out of these 10 have considerably lower metallicity. The only two exceptions are Alpha Centauri B (which is also the only close binary in the set) and Delta Pavonis (which is the only star moving off the main sequence to subgiant stage IV).

    A small table of these stars with spectral type and metallicity is here (I hope it shows reasonably non-distorted, mind that metallicity data can vary, so these are tentative):
    Star Spectral Metal*solar
    HD1581 (Zeta Tucanae) F9V 0,65
    HD10700 (Tau Ceti) G8V 0,35
    HD20794 (82 Eridani) G8V 0,40
    HD65907A G2V 0,49
    HD85512 K5V 0,47
    HD109200 K0V 0,50
    HD128621 (Alpha Cen B) K1V 1,60
    HD154577 K0V 0,26
    HD190248 (Delta Pavonis) G5IV 2,10
    HD192310 K3V 0,91

    Remarkably this seems to confirm well what was recently written by Greg Laughlin on his systemic site, following post:
    Quote: “First, among host stars with masses similar to the Sun that harbor giant planets, there’s a strong preference for metal-rich stars. This is the classic planet-stellar metallicity effect. Second, among low-mass stars, there’s a dearth of giant planet candidates. This is the known giant planet-stellar mass effect.”
    Low-mass star is defined here as below 0.8 solar mass (roughly K1).

    So concluding: solar type stars with low metallicity show a paucity of giant planets, and so do low-mass stars. Vice versa, giant planets, and particularly the very large close-in giants (giant Jupiters, hot Jupiters) seem to ‘prefer’ the solar and larger stars with high metallicity.

    The only exceptions here are Alph Cen B, which is a close binary, so not surprising there are no giant planets there, and Delta Pav, which is an oddball.

  • Ronald August 22, 2011, 6:45

    One more final remark with regard to my previous little analysis: it is remarkable and maybe surprising that we may appear to live in a galaxy (universe?) with no lack of terrestrial planets in the broadest sense (including super-earths), and ‘subgiants’ (Neptune class), but possibly rather a paucity of gas giants.

  • Enzo August 22, 2011, 8:04


    Your question is not easy to answer because it depends on the roughness of the surface. For Earth, using Wikipedia as a source, x = ~0.001196. That gives an h = ~ 2.5 Km. That makes sense because, if you divide by 0.71 (the % of the Earth surface which is ocean), you get 3.58 Km, close to the ocean’s average depth.
    Now, if you think of doubling h, you would be adding at least 2.5 Km (a bit more as there is a small % of Earths surface higher than 2.5Kms).
    However, a super-earth should have lower mountains, but how much lower I have no idea. My educated guess is that a super-earth with radius 2x Earths and the same % of volume as water would probably be almost a complete waterworld. If the super-earth had the same % of mass as Earth of water, then the ocean would be even deeper.
    Looking at this graph here, almost pure silicate super-earths would have to be fairly massive to reach 2x Earths’ radius :

    Also, interesting post on Systemic’s blog on how the water is gradually expelled from the mantle :

  • Enzo August 22, 2011, 8:18


    Water deliver might very well be a random event, but this doesn’t change the fact that a super-earth receiving the same % of volume of water as Earth (or, worse, % of mass), would have a deeper (and wider) ocean than Earth.
    This is due to two different factors and that seem both difficult to dispute even though flatness might be hard to quantify.

  • andy August 22, 2011, 13:04

    Remember that RV detection does suffer from inclination degeneracy, if we see the systems pole-on we don’t get an RV signature. Anyone aware of an inclination measurement for the Tau Ceti debris disc?

  • kurt9 August 22, 2011, 18:51

    The Kepler-11 results suggests anything bigger than 1.5 Earth diameters is probably not a rocky world.

  • Enzo August 22, 2011, 21:18

    @ andy

    According to the web page (scroll down to the bottom), Tau Ceti is almost pole on orientation :

    That’s no good for RV.

  • Phil August 23, 2011, 0:15

    Good point Andy – if Tau Ceti IS highly inclined, maybe it needs to be a priority astrometric target! Isn’t this going to be GAIA’s remit?


  • Ronald August 23, 2011, 6:05

    kurt9: “The Kepler-11 results suggests anything bigger than 1.5 Earth diameters is probably not a rocky world”.

    I am not sure whether this is a response to Enzo on his ocean worlds or to me (on metallicity and planet types), but: true, however, such a larger planet would probably still have a rocky core and hence be a super-earth, still a terrestrial planet in a broader sense. But with a very dense atmosphere.

  • kzb August 23, 2011, 7:13

    @Enzo, yes clearly if a world receives the same % of water as did Earth then you are not wrong. But I’m questioning that very premise.

    There should be a correlation between planet mass and amount of water, but I believe that correlation will be very weak.

  • andy August 23, 2011, 13:18

    Good point Andy – if Tau Ceti IS highly inclined, maybe it needs to be a priority astrometric target! Isn’t this going to be GAIA’s remit?

    It is becoming quite clear that we need astrometry to make more progress with planet-finding: one thing which is becoming clear with Kepler is that intrinsic stellar noise is going to make finding Earth-size planets much more difficult: there are now suggestions that it won’t succeed at finding Earth-size planets in the habitable zone without an extended mission. Radial velocity detection is also hampered by stellar activity. Astrometry by contrast is less affected by stellar activity, so represents a good bet for finding nearby low-mass planets, and gives you the true masses as an additional advantage.

    I think we are really going to regret the loss of SIM-Lite over the next few years…

  • kzb August 24, 2011, 13:26

    I was disappointed to read in the paper that Tau Ceti shows no sign whatsoever of a planetary system. In fact they even use it as a calibrator to assess the background noise for other detections.

    However perhaps we can be encouraged by Enzo’s post concerning its pole-on orientation to us. Is this orientation primarily why Tau can be used in a calibrator in this way?