≡ Menu

Time, Tides and Habitability

Keep your eye on Gliese 581. Not that the news is necessarily good for our hopes for habitability around that star — in fact, a recent paper suggests quite the opposite. The red dwarf exploded into the public consciousness with the announcement that one of its planets — Gl 581 c — could conceivably support clement temperatures and water at the surface, at least in places. But in exploring that possibility, we’re getting a case study of world-class science at work, analyzing data, offering hypotheses, broadening options. It’s an exciting process to watch.

Gl 581 d is now being analyzed for habitability, while Gl 581 c begins to appear less and less likely as a home to life. It may take decades and new space-based observatories for the issue to be resolved, but we now have a new take on Gl 581 c, embedded in a broader study of tidal evolution as planetary systems evolve. The study has implications not just for rocky worlds but for planetary formation in many scenarios.

The work of Brian Jackson, Richard Greenberg and Rory Barnes (University of Arizona) draws on a key fact: A planet’s orbit can be greatly affected by tides that the planet raises on its star, and on the tides the star raises on the planet. In fact, tidal distortion and orbital evolution work together, tidal forces producing internal heating at the expense of orbital energy. Thus many close-in planets probably formed further out from their host star than their present position. In a typical case, say the authors, tidal heating increases as a planet moves inward and then decreases when the tides circularize the orbit and shut down the heat mechanism.

But each case will be different, the strength and timing of these effects determining a planet’s properties. The team’s intention is to construct heating histories for planets whose radii have been measured, sometimes with results that vary from theory. And that gets me back to Gl 581 c, for in terms of planets with masses less than ten times Earth’s, such heating could have played a role in the planet’s geophysical development. The Arizona team finds that the contribution of tidal energies on two ‘super Earths’ — Gl 581 c and GJ 876 d — should produce a heat flux with profound implications:

Among terrestrial-scale planets, we find that tidal heating may have dominated the geological and geophysical evolution of the planets and control their current character. The tidal heating rate for GJ 876 d may be orders of magnitude greater than the magnitude considered by Valencia et al. to be geophysically significant. For Gl 581 c tidal heating may yield a surface flux about three times greater than Io’s, suggesting the possibility of major geological activity.

Three times that of Io? Gl 581 c looks less hospitable all the time. The case of GJ 876 d is even more extreme. This ‘super-Earth’ of 5.89 Earth masses hasn’t been in the habitability picture because its two-day orbit keeps it far too close to its star for liquid water to exist. But while the planet has been considered vulnerable to tidal stresses, I don’t think anyone was prepared for what the Arizona team found:

…radiogenic heating of GJ 876 d might have been adequate to initiate plate tectonics, but our results indicate that tidal heating may have been a major contributor to the geological and geophysical character of the planet. Tidal heat has provided an important component of the heat budget for this planet, perhaps the dominant component during at least the past ~108 yr. The tidal heating rate would be so large, in fact, that GJ 876 d is unlikely to be a solid, rocky body.

I’ve only focused on two super-Earths here, but the paper also offers interesting takes on planets like HD 209458 b, whose radius is larger than predicted, and HAT-P-2 b, whose radius is well below prediction. Tidal heating histories may help us understand these apparent anomalies. The paper is Jackson, Greenberg and Barnes, “Tidal Heating of Extra-Solar Planets,” accepted by the Astrophysical Journal (abstract).

Comments on this entry are closed.

  • dad2059 March 14, 2008, 11:48

    Hmmm, doesn’t look well indeed for the Gliese planets. Tital forces were an early worry for red dwarf star planetary systems to begin with weren’t they Paul? Mainly if there were rocky worlds, they would be tidal locked with one face toward the primary.

    What if one was a little further than the classical ‘sweet spot’? If the orbit isn’t too elliptic and the tidal forces was just enough for plate tectonics, the generated heat might make conditions somewhat livable(?)

  • Administrator March 14, 2008, 12:29

    dad2059, you’re right, we’ve had concerns voiced here about tidal forces and red dwarfs before, although the tidal lock isn’t necessarily a show-stopper when it comes to habitable conditions on at least part of the planet’s surface. It appears to me after reading the current paper that we have a long way to go in characterizing these forces and their potential effect over time. I know that Paul Shankland at the US Naval Observatory, a major player in the M-dwarf planet hunt, thinks these stars will offer up many potentially habitable planets, so I wouldn’t write anything off yet, except perhaps GJ 876!

  • Athena March 14, 2008, 12:57

    Gl 581 c, we hardly knew ye!

    Seriously, though, we need improvements that will allow not just identification but also observation of earth-like planets. Our science and culture will be totally altered (and, I venture to hope, for the better) by such advances.

  • ljk March 14, 2008, 14:45

    Radiative Thrusters on Close-in Extrasolar Planets

    Authors: Daniel Fabrycky

    (Submitted on 13 Mar 2008)

    Abstract: The atmospheres of close-in extrasolar planets absorb most of the incident stellar radiation, advect this energy, then reradiate photons in preferential directions. Those photons carry away momentum, applying a force on the planet. Here we evaluate the resulting secular changes to the orbit, known as the Yarkovsky effect. For known transiting planets, typical fractional changes in semi-major axis are about 1% over their lifetime, but could be up to ~5% for close-in planets like OGLE-TR-56b or inflated planets like TrES-4.

    We discuss the origin of the correlation between semi-major axis and surface gravity of transiting planets in terms of various physical processes, finding that radiative thrusters are too weak by about a factor of 10 to establish the lower boundary that causes the correlation.

    Comments: 4 pages, accepted to ApJL

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0803.1839v1 [astro-ph]

    Submission history

    From: Daniel Fabrycky [view email]

    [v1] Thu, 13 Mar 2008 18:35:03 GMT (26kb)


  • Dave Moore March 14, 2008, 15:56

    Tidal heating may improve the chances of habitability on sub-Earth sized planets. One of the reasons Mars is so unihabitable is that it is geologically dead.

    If you look at what we know of Earth’s and Mars’s history, the biggest factor in maintaining habitability appears to be is geological activity. It helps maintain a degree of stasis against external forcing of the environment and it also maintains a chemical disequilibrium, which can provide an energy source for life.


  • Adam March 15, 2008, 6:45

    Hi Dave

    Tidal forces could provide useful energy for keeping planets warm – especially if those planets are moons! Look at Io and Europa, then imagine them a bit bigger. Could their tidal-heating help sustain a deep greenhouse atmosphere on a Mars or Earth sized moon?

  • andy March 15, 2008, 7:11

    This poses the question of just what kind of planet Gliese 876d actually is. If it’s rocky, perhaps there is a global magma ocean? What about on a planet where the main component by composition is icy material?

    In addition, there’s the issue of how much energy goes into volcanism and how much energy goes into moving tectonic plates around on the planet’s surface.

    One thing seems fairly certain – given the combination of high gravity and extensive melting, these planets are not places to expect significant vertical relief.

  • philw1776 March 15, 2008, 8:35

    Once again, in the recent hype over M star planets folks forget how narrow the HZ is for a typical M red dwarf several THOUSANTHS or less of the sun’s luminosity. Tidal show stopers aside, the extremely low probability of finding a rocky planet in such tiny HZs works against even the large #s of M dwarfs.

    Sadly, F stars have large HZs potentially containing multiple ‘Earthlike’ planet candidates, but their shorter lifetimes would tend to migrate the HZ outward too quickly for multicellular life to evolve, assuming (a big one) that Earth’s history is typical.

  • Adam March 15, 2008, 16:25

    Hi phil1776

    The fact we’ve found several planets in, or close to, the red dwarf CHZ, might mean there’s not a random distribution of planetary orbits. A lot of cosmogonists expect some kind of scaling dependent on the star’s mass, thus making planets in the CHZ more likely than chance alone.

    The enhanced tidal forces might allow liquid water biocompatibility even further out than anyone has so far guessed. Definitely a process to watch out for as the tally of red dwarf planets goes up.

  • Dave Moore March 15, 2008, 16:41

    I wasn’t so much thinking of moons but of a planet somewhere between Mars and Earth’s mass, tidally locked, and having its orbital eccentricity pumped by a larger planet.

    There is evidence that tidally locked planets may be habitable; the heat transfer is enough to keep the atmosphere from freezing out. And tectonic activity is certainly likely to help in this regard.

    I could see tidal heating helping habitability in the following ways:
    i) Creation and maintenance of a magnetic which would help prevent atmospheric loss (Though this argument is questionable as the planet’s orbital period and hence its rotational period would be in weeks, and Io, which rotates every 1.7 days doesn’t have a magnetic field for some reason.)
    Anyhow, the primary, being a red-dwarf, would put out a lot less UV, which would slow atmospheric loss.

    ii) Once the carbonate cycle thinned the atmosphere so the planet froze, a more active volcanism would reverse this process more quickly. Even if the planet froze over completely, active volcanism would maintain subsurface oceans and keep the overlaying ice cap thin enough so that cracks may open to the surface allowing photosynthesis.

    I also think that tidally locked planets, because of their extremes, would have a broader HZ than planets with more equitable conditions over their entire surface. Because it’s difficult to freeze the sub-solar point of a tidally locked planet or melt the antisolar point, they would resist runaway greenhouse or freezing out.

    Actually, given the differences or this type of planet to Earth, I think rather than use the term “Habitable Zone” which implies humans could live there, I’d use the term “Life Bearing Zone.”

    Because Red dwarfs are very common and I assume smaller planets more common than larger ones, then this sort scenario has a reasonable probability.


  • philw1776 March 15, 2008, 18:38

    “The fact we’ve found several planets in, or close to, the red dwarf CHZ, might mean there’s not a random distribution of planetary orbits. A lot of cosmogonists expect some kind of scaling dependent on the star’s mass, thus making planets in the CHZ more likely than chance alone.”

    Although the science isn’t solid I also believe distribution’s NOT random and that it DOES scale with mass somewhat, but do the math. Let’s assume that the sun’s HZ for an Earthlike planet to survive several BILLION years without total freeze over or heat runaway allowing the evolution of complex multi-celled life is say 30 million miles, a decent window to have a terrestrial planet form within. However, given this the HZ of a typical M dwarf of less than a solar mass is under 1 million miles. Luminosity and SQRT. So, the fact that there are several tens of times more Ms than Gs is more than mitigated by the low probability of a terrestrial style planet hiting this really tiny HZ.

    And that’s forgeting the troublesome tidal problems raised in this discussion.

  • andy March 17, 2008, 5:01

    If instead of jumping in with maths we take a look at the data…

    If philw1776 is correct, then planets in the habitable zones of M dwarfs would be vanishingly rare, but instead we have two systems with planets in the HZ (both of which may have up to two planets in the HZ), out of a total of about eight planetary systems known around M dwarfs. This strikes me as staggeringly unlikely if such planets are as vanishingly rare as philw1776 makes out.

  • Dave Moore March 17, 2008, 20:11

    I took a look at the average planetary spacing around the habitable zone in an attempt to get a quantitative answer to the likelihood of planets falling into a red dwarf’s HZ. The calculations I did are very ad hoc and rough and there isn’t a statistically significant sized sample to conclude anything; nevertheless, I think you can get some idea by looking at the situation.

    If you take a look at the sun and assume like philw1776 did that the HZ is 30 million miles wide and if you look at the average planetary spacing around Earth (40 million miles) then you could assume that in any given sunlike system the would be a 75% chance of a planet falling into the habitable zone.

    If you look at a M3/M4 dwarfs like GL 581 and GL 876, the HZ is from 3 to 4 million miles out. The planetary spacing at about this distance, if we look at the GL 581, GL 876 and 55 cancri sysyems (and assume there is at least one extra planet between GL 876 b & c) is in the order of 4 million miles so the likelihood of a planet falling in the HZ falls to 25%. While this is less frequent than for sunlike stars, it is not vanishingly rare.


  • andy March 18, 2008, 8:54

    In fact it is probably better to do this analysis in logarithmic space rather than linear: a spacing rule of the form a=A exp(Bn) fits both our own solar system (with the planets in positions n=1,2,3,4,6,7,8,9) and 55 Cancri (1,2,3,4,6).

    The value of B, which represents the spacing of the planets in logarithmic space seems to range from about 0.6 to 1 for most of the known systems (including the red dwarf systems). The logarithmic width of the habitable zone is unaffected by scaling for luminosity. taking an HZ of 0.95 to 1.6 AU in our solar system corresponds to a logarithmic width of about 0.5. More liberal definitions of the HZ could take this figure up to 0.8 or more… thus systems with 1-2 planets in the habitable zone are quite plausible and should be fairly frequent, even around M dwarfs.

    For reference, the planetary configurations I investigated for the M dwarf systems, together with the error on the linear regression in logarithmic space and the value of B:
    Gliese 581 (1,2,3): err=0.27; B=0.9
    Gliese 581 (1,2,4): err=0.02; B=0.6
    Gliese 876 (1,3,4): err=0.24; B=0.8
    Gliese 876 (1,4,5): err=0.08; B=0.6

  • ljk May 7, 2008, 22:33

    Ground-based detection of sodium in the transmission spectrum of exoplanet HD209458b

    Authors: I.A.G. Snellen, S. Albrecht, E.J.W. de Mooij, R.S. Le Poole (Leiden Observatory)

    (Submitted on 7 May 2008)

    Abstract: [Context] The first detection of an atmosphere around an extrasolar planet was presented by Charbonneau and collaborators in 2002. In the optical transmission spectrum of the transiting exoplanet HD209458b, an absorption signal from sodium was measured at a level of 0.023+-0.006%, using the STIS spectrograph on the Hubble Space Telescope. Despite several attempts, so far only upper limits to the Na D absorption have been obtained using telescopes from the ground, and the HST result has yet to be confirmed.

    [Aims] The aims of this paper are to re-analyse data taken with the High Dispersion Spectrograph on the Subaru telescope, to correct for systematic effects dominating the data quality, and to improve on previous results presented in the literature.

    [Methods] The data reduction process was altered in several places, most importantly allowing for small shifts in the wavelength solution. The relative depth of all lines in the spectra, including the two sodium D lines, are found to correlate strongly with the continuum count level in the spectra. These variations are attributed to non-linearity effects in the CCDs. After removal of this empirical relation the uncertainties in the line depths are only a fraction above that expected from photon statistics.

    [Results] The sodium absorption due to the planet’s atmosphere is detected at >5 sigma, at a level of 0.056+-0.007% (2×3.0 Ang band), 0.070+-0.011% (2×1.5 Ang band), and 0.135+-0.017% (2×0.75 Ang band). There is no evidence that the planetary absorption signal is shifted with respect to the stellar absorption, as recently claimed for HD189733b. The measurements in the two most narrow bands indicate that some signal is being resolved.

    Comments: Latex, 7 pages: accepted for publication in Astronomy & Astrophysics

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0805.0789v1 [astro-ph]

    Submission history

    From: Ignas Snellen [view email]

    [v1] Wed, 7 May 2008 07:44:22 GMT (79kb)


  • ljk July 7, 2008, 23:04

    Tides and the Evolution of Planetary Habitability

    Authors: Rory Barnes, Sean N. Raymond, Brian Jackson, Richard Greenberg

    (Submitted on 4 Jul 2008)

    Abstract: Tides raised on a planet by its host star’s gravity can reduce a planet’s orbital semi-major axis and eccentricity. This effect is only relevant for planets orbiting very close to their host stars. The habitable zones of low-mass stars are also close-in and tides can alter the orbits of planets in these locations.

    We calculate the tidal evolution of hypothetical terrestrial planets around low-mass stars and show that tides can evolve planets past the inner edge of the habitable zone, sometimes in less than 1 billion years. This migration requires large eccentricities (>0.5) and low-mass stars (<0.35 M_Sun). Such migration may have important implications for the evolution of the atmosphere, internal heating and the Gaia hypothesis.

    Similarly, a planet detected interior to the habitable zone could have been habitable in the past. We consider the past habitability of the recently-discovered, ~5 M_Earth planet, Gliese 581 c. We find that it could have been habitable for reasonable choices of orbital and physical properties as recently as 2 Gyr ago.

    However, when we include constraints derived from the additional companions, we see that most parameter choices that predict past habitability require the two inner planets of the system to have crossed their mutual 3:1 mean motion resonance. As this crossing would likely have resulted in resonance capture, which is not observed, we conclude that Gl 581 c was probably never habitable.

    Comments: 31 pages, 10 figures, accepted to Astrobiology. A version with full resolution figures is available at this http URL

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0807.0680v1 [astro-ph]

    Submission history

    From: Rory Barnes [view email]

    [v1] Fri, 4 Jul 2008 05:47:27 GMT (287kb)