≡ Menu

Keeping a Planet Alive

I’ve made no secret of my interest in red dwarf stars as possible hosts of life-bearing planets, and this is partially because these long-lived stars excite visions of civilizations that could have a stable environment for many billions of years. I admit it, the interest is science fictional, growing out of my imagination working on the possibility of life under the light of a class of stars that out-live all others. What might emerge in such settings, in places where tidal lock could keep the planet’s star fixed at one point in the sky and all shadows would be permanent?

Some of this interest grows out of an early reading of Olaf Stapledon’s 1937 novel Star Maker, in which the author describes life in the form of intelligent plants that live on such a tidally locked world. For that matter, Larry Niven developed an alien race called the Chirpsithra, natives of a red dwarf who have a yen for good drink and socializing with other species (you can sample Niven’s lively tales of these creatures in The Draco Tavern, a 2006 title from Tor). I tend to imagine red dwarf planet dwellers as something more like philosophers and sages than intelligent carrots or Niven’s incredibly tall barflies.

But no matter. A new paper from Christa Van Laerhoven and Rory Barnes (University of Washington) and Richard Greenberg (University of Arizona) has me absorbed in matters such as how close an Earth-class planet would need to be to stay habitable around a red dwarf. There’s no one answer because of the range of stellar temperatures between different types of red dwarf, but Laerhoven and company are looking at a star with a mass of 0.1 solar masses and a luminosity 1.15 x 10-3 times that of the Sun. Here it turns out that to receive the same incident flux as the Earth, the planet would need to orbit at 0.034 AU.

Now Mercury is about 0.38 AU from the Sun, which gives us a feel for how much cooler such a star must be. We can also note that because of their long lifetimes, many red dwarfs are much older than our Solar System, on the order of twice as old in some cases, and because a transiting Earth-class planet around such a star should be detectable (the transit depth would be huge), it’s possible that the first Earth-like habitable planet we find will be billions of years older than our own. Thus my visions of ancient races of philosophers under a darkened sky.

But maybe not. The Van Laerhoven paper makes the case that planets like these are going to be cooling internally as they age, enough so to cause problems. Plate tectonics are driven by heat, and we’re learning their necessary function in the carbon cycle that allows the planet to avoid greenhouse overheating. Here’s the issue (internal citations omitted):

On an Earth-like body, long before reaching twice Earth’s age, plate tectonics would probably have turned off as the planet cooled, primarily because solidification of the core would terminate the release of latent heat that drives mantle convection. While plate tectonics may not be essential for life on all habitable planets, an equivalent tectonic process to drive geochemical exchange between the interior and the atmosphere is a likely requirement. The necessary amount of internal heat for such activity is uncertain (even the mechanisms that govern the onset and demise of terrestrial plate tectonics are still poorly understood and controversial), but it seems likely that a planet ~10 Gyr would have cooled too much…

So my race of philosophers and poets may have a much shorter time to thrive than the ten trillion years its dim star will live. What we need is an additional heat source, and the possibility in play in this paper is tidal heating, which the paper argues calls for either non-synchronous rotation or an eccentric orbit. Even these are a problem because tidal effects gradually synchronize the rotation and circularize the orbit, but we need them to help us on geological timescales.

The solution may be another planet in the same system, an outer companion that can keep the inner planet’s orbit from circularizing and thus maintain the tidal stresses that heat the planet. The computer models the researchers used allow this effect to keep the inner world habitable for billions of years even when other internal sources of heat have long perished. The paper argues that this effect, while studied here only in terms of two-planet systems, can also come into play in systems with a larger number of planets. From the paper:

…a reasonable fraction of terrestrial-scale planets in the HZ of very old, low-mass stars may be able to sustain life, even though without a satisfactory companion they would have cooled off by now. The requirements on the outer planet are not extremely stringent. For example, one could well imagine a Neptune-size outer planet a few times farther out than the rocky planet with an orbital eccentricity ~0.01-0.02. Not only would such an outer planet yield an appropriate amount of tidal heating to allow life, but the heating would be at a steady rate for at least tens of Gyr.


Image: For certain ancient planets orbiting smaller, older stars, the gravitational influence of an outer companion planet might generate enough energy through tidal heating to keep the closer-in world habitable even when its own internal fires burn out. But what would such a planet look like on its surface? Here, UW astronomer Rory Barnes provides a speculative illustration of a planet in the habitable zone of a red dwarf. “The star would appear about 10 times larger in the sky than our sun, and the crescent is not a moon but a nearby Saturn-sized planet that maintains the tidal heating,” Barnes notes. “The sky is mostly dark because cool stars don’t emit much blue light, so the atmosphere doesn’t scatter it.” Credit: Rory Barnes / University of Washington.

It could be, then, that a planet in this configuration — a terrestrial world like the Earth orbiting a 0.1 solar mass star with an outer companion — could experience enough tidal heating to make it the longest lived surface habitat in the galaxy. Is such a world, as the authors speculate, a possible home for humanity in the remote future, when our own Earth becomes uninhabitable? For that matter, given that such worlds seem made to order for ancient civilizations, shouldn’t we consider them as good SETI candidates? The paper recommends that any search for habitable Earth-scale planets should include a search for outer system companions.

The paper is Van Laerhoven et al., “Tides, planetary companions, and habitability: Habitability in the habitable zone of low-mass stars,” Monthly Notices of the Royal Astronomical Society, published online 12 May, 2014 (abstract / preprint).


Comments on this entry are closed.

  • Lepton August 6, 2014, 12:52


    I personally also have a soft spot for red dwarfs.

    In your article, there are at least three places you mentioned host star size:
    1) a star with a mass of 0.1 solar masses and a luminosity 1.15 x 10-3 times that of the Sun.
    2) Here, UW astronomer Rory Barnes provides a speculative illustration of a planet in the habitable zone of a star about the size of the sun.
    3) a terrestrial world like the Earth orbiting a .01 solar mass star with an outer companion

    Are all of them intentional?


  • Paul Gilster August 6, 2014, 12:59

    No, what you were seeing was a failure of proofreading! My apologies. I’ve gone back into the text to make the needed adjustment. Glad you caught that, as I’ve been short of time this morning.

  • kzb August 6, 2014, 13:46

    SuperEarths, of which there appear to be plenty, should last billions of years longer than Earth simply going by the mass to surface area ratio?

  • jamesmessig August 6, 2014, 14:15

    Red dwarfs could also be used very effectively as stellar engines. Imagine how far a red-dwarf engine could travel through space at say 0.1 C over a period of 10 trillion years, compounded by potential superluminal recessional velocities from the sun due to universal space-time expansion of the intervening space.

    What amazes me is the potential for technological, scientific, philosophical, mathematical, psychological, spiritual and the list goes one, advancements that are possible over the 10 trillion year lifetime of such a star.

  • Greg August 6, 2014, 14:17

    One other possible solution for heating a planets core would be if the planet passes through the red stars magnetic field fast enough to heat it’s core using induction heating. With a life bearing planet being so much closer, it’s possible this can occur.

  • Ron S August 6, 2014, 14:35

    If tidal heating is necessity it is only required until an intelligent species evolves and can invent and choose to apply technology to manage their environment. The one data point we have is that 4.5 Gyr is sufficient.

  • Bill Higgins-- Beam Jockey August 6, 2014, 14:44

    The system you describe sounds a bit like the tiny kingdom of Kepler 42 ( née KOI-961).

    Kepler 42d is not quite cool enough to reside in the habitable zone, but it’s not far away, and our uncertainties in planetary temperatures are large, are they not?

    Sadly, that outer planet you requested does not seem to be available. But perhaps we have not yet learned of every planet Kepler 42 owns.

  • Andrew Palfreyman August 6, 2014, 16:00

    If we accept the assumption that stable conditions for life are possible in such places for at least the current age of the universe, then we can expect a multiplicity of life in our galaxy simply because of the prevalence of red dwarves. Life began on Earth ~10 Gyr postBB and took ~4 Gyr to develop to the current level. If a red dwarf civilisation arose around one of the youngest suns in our galaxy, i.e. ~1 Gyr postBB, and took about the same amount of time to develop as did we to our current level, then such an oldest possible red dwarf civilisation would have a head start on us of about 9 Gyr. For this huge number to work, stability is of course the key.

  • FrankH August 6, 2014, 16:10

    Unless the red dwarf in the illustration is setting in a very dusty atmosphere, it shouldn’t be red. A typical red dwarf is going to look no redder than a light bulb.

  • Chris W August 6, 2014, 16:51

    I too find tidally locked habitable planets a compelling idea. Imagine the psychological, social, and evolutionary ramifications where day & night are spacial rather than temporal phenomena.

  • RonM August 6, 2014, 18:28

    One difficulty I didn’t see discussed in relying on a hypothesized outer companion planet preventing the inner planet’s orbit from circularizing is the whole question of orbit stability over very long time periods, say 10 Gyrs or longer. In other words, the probability of the existence of a life-friendly planet in a truly long term stable orbit around a red dwarf, where the planet in question happens to have a companion in a similarly stable orbit further out (this tidal companion having eccentricity ~0.01-0.02, and meeting perhaps a dozen other requirements) approaches zero real fast.

  • d.m.falk August 6, 2014, 18:59

    Mass is a critical factor in core longevity, as the planet’s own gravity is responsible for keeping the core warm. Not enough mass, and a warm core can’t be maintained. And an active core means an active magnetic field, and an active magnetic field is a planet’s best shield against its sun, and a shelter for life.

    I think that an ideal planet will be between .8 and 1.5 Earth’s gravity, so we should look for worlds of corresponding mass.

    I also am interested in red-dwarf planets. :)

    Two things, though about tidally-locked planets:
    They won’t have moons (just as planets orbit stars, it’s a gyroscopic effect to maintain an orbit).
    They’ll have the equivalent of our moon’s libation, so a habitable planet will have seasons.


  • JBE August 6, 2014, 20:01

    What about a large moon instead of a second planet. How big would such a moon have to be to cause the planet to become tidally locked to it instead of the star, and would the remaining tidal forces from the star acting on the planet, always rotating relative to the star, be enough to keep it hot internally? Such a moon would also allow the planet to enjoy a permanent day/night cycle, although the cycle might be many weeks long.

  • Alex Tolley August 6, 2014, 22:44

    AFAICS, the only value of a red dwarf for an advanced civilization is that it offers a potentially very long term, local energy source. Almost everything else is irrelevant as such civilizations need not be planet bound, so geological and ecological constraints are removed.

    For ecosystems, there are other issues, but it seems likely that bacteria could survive in the lithosphere long after surface life disappeared.

    Another issue is the possible longevity of any civilization, or succession of them. Species will evolve in mere eye blinks compared to such a star’s lifetime. Technological species are likely to self direct their evolution or transformation on extremely short timescales.

    Unless star flight is nigh impossible, the relevance of the longevity of a red dwarf star type for civilization seems moot to me. Either civilization appears, populates its local system and likely migrates to other stars, or it doesn’t. This should happen well within the lifetime of such a star. If it doesn’t, the system will be populated by a star faring species instead.

  • Mephane August 7, 2014, 5:08

    We should also not forget that, when a the inhabited planet around a red dwarf eventually has cooled down too much, by that point any civilization developed there could have already moved out to orbital habitats (e.g. O’Neill cylinders).

  • John Smith August 7, 2014, 8:15

    Apparently, super-earths larger than 1.5 or so Earth masses may not have a “liquid core”-mantle boundary, rather a supercritical fluid region of sorts. That means no magnetic field and then the atmosphere gets eroded, especially since red dwarfs have much more active solar wind (for their size and habitable regions).

    More details here https://www.youtube.com/watch?v=e8ieGwpWeRg (SETI talk about super earths)

  • John Smith August 7, 2014, 8:19

    I forgot to mention that persistent tidal heating from a companion large planet can also change the orbit of the earth-like planet, moving it out of the habitable zone over billions of years.

  • steven rappolee August 7, 2014, 10:30

    I believe it was in Analog magazine years ago, ideas on how to keep plate tectonics going,dump radioactive materials into trenches to keep the mantel heated

  • stephen August 7, 2014, 12:09

    You cite the possibility of only one other planet in the system. Is a red dwarf more likely to fewer planets, or more? The presence of several planets could have a greater cumulative effect.

    Over the course of billions of years, the approach of nearby stars might have a cumulative effect.

  • Ashley Baldwin August 7, 2014, 12:57

    There are certainly papers that say that to be safe from having its atmosphere driven off a planet must be at least .8 AU from its primary (see Habitability of K stars) . Even the strongest dynamo driven magnetic field will be inadequate. Its also true that red dwarf circulating planets are unlikely to have close moons. The key to age that I take from Barnes’ paper is not so much that a planet must have a satellite to maintain its internal heat through tidal forces so much as close by companions. Kepler and RV studies have already discovered compact solar systems, the key is whether or not some of these can have planets in the elliptical orbits that can raise gravitational tides in companion planets a la Io. The outer limit to the HBZ of an M0 dwarf is circa .4 AU, so obviously a companion could be further out. Looking at the equations for tidal locking the critical diatance for avoidance for Earth like lifetimes an M0 is about .5 AU, so maybe a big planet at this distance in an elliptical orbit might just be able to raise tides in a nearer HBZ planet especially if it is subject to tides from larger nearer planets too. From a perspective of main sequence lifetime similar to earth to an orbit outside of .5 for 4.5 gigayears we are looking at spectral dwarfs F6 ( 5 gigayears) to K5, just ( .51 AU) . Another factor is the iron content of any planet . Can we envisage a Super Mercury , small in size but with Earth like mass made up mostly of iron and capable of generating a powerful long term liquid core. Such planets have been mooted.

  • Joëlle B. August 7, 2014, 15:26

    @dmf & JBE & John Smith

    I like the idea of exomoons in red dwarf systems.

    Considering evidence for it being easier to detect sizable exomoons around M dwarfs (cf. https://www.centauri-dreams.org/?p=30670), I’d say that would be a motivating factor in not ruling out the possibility of habitability, even if one finds a tidally locked super-Earth/Earth/non-Earth like planet orbiting–just look for its satellites, which, if ~Mercury-sized+, could give hope?–since said exomoons would become tidally locked to their respective planet, evenly distributing heat and day/night cycle and there’s still a lot we don’t understand about resonant rotation (i.e. the wacky Venus-Earth approaches and Mercury’s 3:2 spin-orbit resonance around Sol?; also, there’s the fact that we can’t jump the gun just yet, being we haven’t thoroughly examined our outer solar system enough to rule out the possibility of life on those planets and moons (and maybe even deep inside of our own satellite or within Mars) irregardless of our ‘current’ understanding of habitability and life.

    Obtaining further information on the conditions necessary for a Ganymede (or larger) sized moon around each type of currently known M dwarf planet mass would be greatly appreciated, if anyone is privied to construct a creative outline or send me in the right direction to figuring it out myself. ^^;

    Additionally, do you know (specifically) about how many Gya it would take for that earth-like planet to move out of the HZ? If it is more than ~4.5 [how long it has taken for us to show ‘interstellar-forward’ technological intelligence], then as the above posters have assessed, I would assume an artificial means of habituation or orbital modification would exist to perpetuate that existence by then if a sentient species was at that point and threatened (since they’d know). And how far would that planet move away from the star, noting it will survive trillions of years without fear of getting eaten-up? Would it eventually go rogue/free-floating if another stellar object doesn’t snatch it?

    Also, might the amount of asteroid/comet (and other protoplanetary disk debris) activity in a system influence the conditions of habitability? It only took one tiny chunk to wipe out 75% of all life on Earth, but it appears that there isn’t a lot of data on M dwarf circumstellar debris. A paper that touches on the issue: http://arxiv.org/abs/1404.1954

    Could this mean that there is a better (or worse) chance of celestial extinction events showing up?

  • Joëlle B. August 7, 2014, 15:53

    Ashley Baldwin said on August 7, 2014 at 12:57
    “Kepler and RV studies have already discovered compact solar systems, the key is whether or not some of these can have planets in the elliptical orbits that can raise gravitational tides in companion planets a la Io.”

    If an elliptical orbit was found in a compact system would that earth-sized planet be less likely (or not) to move out of the HZ?

  • Joëlle B. August 7, 2014, 20:57

    I found the t > 15 Gya time frame to answer my initial question and guess a better way to ask my second question: is it possible for an orbitally eccentric planetary body to be complimentary to geological activity for habitability? I know that eccentric Jupiters are more prevalent than hot Jupiters, though they are more likely to eject any Earth-like planets from the habitable zone, but are there any known scenarios which could enable its satellites to join along in its eccentric journey around its parent star (and not stray away)?

  • Michael August 8, 2014, 3:38

    @Joëlle B. August 7, 2014 at 15:53

    ‘If an elliptical orbit was found in a compact system would that earth-sized planet be less likely (or not) to move out of the HZ?’

    That depends on the mean potential energy distance of the ellipse whether it will move into or out of the HZ.

    If you work out a basic change in potential energy (distance) of a body in orbit around its star you can work out how much power will be liberated to move it, it is quite a lot! Moving the earth by 500 000 km I get a ~6 billion year 84 terra watt heat liberation, that is as much light the earth receives from our Sun per second.

    Now if the planet has a thick atmosphere, ice or ocean it would be more insulated and retain heat better for longer allowing a smaller tidal heating interaction requirement.

  • TitanFlyer August 9, 2014, 0:38

    Wouldn’t systems with multiple red dwarfs (if they are close enough together) be likely to add eccentricity to a habitable zone planet’s orbit, or at least to a larger outer planet? So many systems have more than one star; why wasn’t that possibility discussed?

  • Ashley Baldwin August 9, 2014, 13:27

    Interestingly enough there are numerous studies focusing on viable , circular orbits around red dwarf binaries with the close “contact” type binary attracting more attention because of their capacity to have cirumbinary orbits , so called “p” orbits , within the dual system habitable zone. ( the 2 alpha Centauri members are frustratingly a bit too close to allow HBZ p orbits, but not so far away as to interphere in individual ‘s” orbits. Bit if salt in baby bear’s porridge , but happily ,according to an article in ArXiv on Friday ,not so much as to prevent stable ‘a’ orbits) .In terms of be eccentric orbits in close proximity, tight, M dwarf planetary systems these would be more likely powerful in the outermost members especially if the had been disturbed during the violent planetary formation period if the NICE model is to be believed although that involves gas giant migration ( as is postulated to have occurred in our own system) and gas giants appear rare around M dwarfs.
    We dont just have to limit ourselves to M dwarfs though . With TESS drawing nearer and K2 ongoing they are the flavour of the month at present. Most attention is shared between them and G stars given our own stars spectrum . We hear practically nothing about K dwarfs , the lower k1-3 in particular are very long lived with HBZs far enough out to avoid exposure of their planets to as much of a radiation battering ( and also far enough to avoid becoming tidally locked for Giga years ) . Their long main sequence spectrum ( 20 Gig plus years) is very long and clearly tidal warming to maintain internal heat applies here critically . Its important to remember that maintenance of internal heat is not just to support a radiation protecting dynamo induced magnetic field, but also to keep belching out the volatiles that produce the secondary atmosphere and maintain the all important carbon cycle / greenhouse effect and resultant stable atmospheric temperature. Maybe Heller and Barnes super habitable planets orbit a K dwarf. Quite a few locally so I’m sure PLATO will look at them and having seen the interim proposal reports for NASA’s direct imaging Probe/WFIRST -AFTA mission , all 3 include K spectrum stars for analysis .

  • Ashley Baldwin August 9, 2014, 16:49

    Joelle B and all those interested in tidal locking , its impact on life , habitable zones and change over and in relation to M dwarfs and other stars. There is a paper in .arXiv published in 2012 .” Evolution of magnetic protection in potentially habitable terrestrial planets” , Zuluaga et al . It addresses many of your questions. Figure 3 ( page 7) is particularly illuminating. It adorns my office wall. Strongly recommend a read. Best wishes

  • D August 28, 2014, 14:25

    A planet has been found around Groombridge 34 A!

    The NASA-UC-UH Eta-Earth Program: IV. A Low-mass Planet Orbiting an M Dwarf 3.6 PC from Earth