≡ Menu

Habitability Around Red Giants

The prospect of habitable planets around red giant stars fires the imagination, enough so that quite a few readers forwarded me the link to a recent paper looking at this question. I’m reluctant to speak for others, but I suppose a major reason we’re so interested (and I, too, had flagged the paper as soon as it popped up on the arXiv server) is that it changes our view of habitable worlds once again. Not long ago it was only the G-class, Sun-like star that seemed a likely abode of life. Then we started looking hard at M-dwarfs. Do we now extend the search to massive red giants, the descendants of stars once like our own?


Image credit: NASA, ESA and A. Feild (STScI).

Werner von Bloh (Potsdam Institute for Climate Impact Research) and team show that the possibility is real. We’ve long known that life on a planet in Earth’s orbit would not survive the swelling of the Sun, even if it did not actually engulf the planet. But life on Earth would actually die out long before that event, if for no other reason than that carbon dioxide would be removed from the atmosphere as the planetary core cooled, slowing the volcanic activity needed to replenish it. The weathering of silicates on Earth’s continents creates the carbon dioxide sink within which the CO2 gradually accumulates.

We’re looking, then, at the end of photosynthesis. At an age of roughly 6.5 billion years, long before the Sun enters its red giant phase, the Earth becomes uninhabitable. But as a planet like ours gradually dies, dormant ‘super-Earths’ — planets more massive than the Earth but with similar chemical and mineralogical composition — may be heading toward an entirely different fate. A super-Earth well beyond 1 AU in a Sun-like star’s system could be thawed into life during the star’s red giant expansion. Frozen earlier in the system’s history and well beyond what von Bloh defines as the ‘photosynthesis-sustaining habitable zone’ (pHZ), the planet would have sustained no weathering and consequent loss of CO2.

Warm such a planet up by the swelling of the central star and weathering begins, with a CO2 equilibrium being established. A super-Earth dominated by oceans would hold onto its CO2 the longest because of the reduced amount of land surface. The numbers here make it clear that land surface is crucial in the length of time a planet can sustain photosynthesis. The paper discusses how long a planet might take to transit the pHZ:

For planets of a distinct size, the most important factor is the relative continental area. Habitability was found most likely for water worlds, i.e., planets with a relatively small continental area. For planets at a distinct distance from the central star, we identified maximum durations of the transit of the pHZ. A comparison of planets with different masses revealed that the maximum duration of the transit increases with planetary mass. Therefore, the upper limit for the duration of the transit for any kind of Earth-type planet is found for most massive super-Earth planets, i.e., 10 M⊕ , rather than 1 M⊕ planets, which are rendered uninhabitable after 6.5 Gyr…

The best case scenario is an ocean-dominated world not far beyond the present orbit of Mars, a ten Earth-mass world that could maintain itself in the star’s habitable zone for a whopping 3.7 billion years. Searching for habitable planets, then, we may want to add red giants to the mix, another surprising development that points to the widening range of astrobiological investigations. And maybe we should start thinking in terms of two life cycles around stars like our own, the first as the star burns through the main sequence, the second a perhaps lengthy efflorescence of life on a previously frozen world.

The paper is von Bloh et al., “Habitability of Super-Earth Planets around Other Suns: Models including Red Giant Branch Evolution,” in press at Astrobiology and available online.

Comments on this entry are closed.

  • Ronald December 15, 2008, 12:22

    “A comparison of planets with different masses revealed that the maximum duration of the transit increases with planetary mass.”

    Doesn’t this maximum duration of transit time also depend on the spectral type of the mother star? I mean, a more or less solar type star that spends more time on the main sequence, say a late G or early K star, would first allow for a longer (and more stable) HZ transit period (i.e. the first cycle).

    But would it not also allow for a longer transit period in its red giant stage?

  • Ronald December 15, 2008, 12:29

    With ref. to my previous post about late G/early K solar type stars evolving off main sequence into red giant stage: in our galactic neighborhood good examples of these are Delta Pavonis (G5-G8) and Gliese 454 (K0).

  • andy December 15, 2008, 15:32

    Well at some point, decreasing the stellar mass is going to stop increasing the planetary lifespan because the geological evolution becomes the limiting factor rather than stellar evolution. For Earth-mass planets, the transition between the two regimes is slightly above 1 solar mass (at least if you put the planet in the optimum habitable orbit – Earth is closer to the Sun than the optimum orbit, and the last estimates I’ve seen suggest that Earth gets toasted about a billion years before its predicted geological death).

    As for stars with less mass than the Sun, while the giant stage would last longer, there’s also going to be far more time for the planetary cooling to kill off planets that would otherwise have been habitable during the giant stage.

  • kurt9 December 15, 2008, 21:05

    A little off topic. But is there any theoretical limit on the size of a super-earth planet?

  • David December 16, 2008, 5:56


    Limiting size depends on mass, for which an upper-limit is probably around 10 Earth masses since a more massive planet will rapidly accrete gas from the protoplanetary disk and become a gas-giant.

    This assumes the standard core accretion theory of planetary formation which assumes planetary formation occurs within the first few million years from the collapse of the molecular cloud that led to the formation of the central star.

    The size of such a planet depends to some extent on its composition. A 10 Earth mass planet made of Iron will be a smaller than one mostly made of water. Sara Seagar
    has done research on this in the last few years

  • Michael Spencer December 16, 2008, 7:49

    Paul: Well, your headline sure grabbed me! But when I think about what happens as a star progresses towards red giant, intense radiation for very long periods comes to mind, along with very high velocity solar winds that throw high energy particles in every direction. I wonder what would be the effect on the future ‘ecosystem’ of such an onslaught.

  • John M. Dollan December 16, 2008, 10:42

    Would not the repeated expansion and contraction of the red giant during this phase of its life have some serious consequences on any potentially habitable super Earth?

  • Adam December 16, 2008, 19:52

    Hi all

    Several questions to answer.

    Firstly, kurt9, there is a maximum mass in most models of planet formation for “SuperEarths” because above about 10 Earth masses the planet can pull gas in from the surrounding nebula very efficiently, thus becoming a Jupiter mass in ~100,000 years. Thus, even if gas supply is lower, a SuperEarth becomes an Ice Giant or a sub-Jupiter very rapidly with more than 10% hydrogen/helium.

    John Dollan, the expansion and contraction phase occurs after the Red Giant Branch of a solar mass star’s evolution, during what’s called the Asymptotic Giant Branch. Between the two stages is about 100 million years of steady helium burning during the Helium Main Sequence. If a planet survives the RGB and is suitably placed – about Jupiter’s current position – then it would have habitable temperatures during the HMS.

    Mike Spencer, according to a tame astrochemist I asked, the Solar Wind, while much higher in numbers, is also much lower in energy coming off an RGB star. Apparently the Wind speed drops to about the surface escape velocity – so about 60 km/s for a Red Giant 100 times bigger than the Sun. Also Red Giants aren’t noted for flares, though they do often have big Sun-spots. The radiation environment is much more benign than MS stars, though the overall mass-loss rate can erode atmospheres left exposed without magnetospheres.

  • James M. Essig December 17, 2008, 3:06

    Hi Folks;

    Imagine the most luminous red giant star. Such stars can have a luminosity on the order of a million times that of the Sun. This is the equivalent of converting 4 trillion tons of matter into energy per second; enough power, if converted effeciently into ship based kinetic energy to accelerate a Nimitz Class nuclear powered aircratf carrier to a gamma factor of 40 million every second.

    Perhaps ETI on planets in a habitable zone around such Red Giants could develope some sort of electro-magneto-hydrodynamic-plasma drive, an electro-hydrodynamic-plasma drive, an magneto-hydrodynamic-plasma drive, an ion rocket, electron rocket, and/or a photon rocket based propulsion system. The craft could travel around the coronosphere of the red giant, while using the stellar atmospheric and magnetospheric plasma as a reaction mass wherein thrust vectoring would keep the space craft glued to the star. The relative energy of the plasma, starlight, and stellar magnetic fields would increase with respect to the craft as it closed in on C. This would permit a more vigorous interaction between the craft and the stellar plasma thus allowing the craft to draw more energy from the plasma while providing the craft with more thrust inorder to keep the craft glued to its orbital motion around the star. The process would continue until the craft approached C very closely. Eventually, the craft would exit the star by shutting down its thrust vectoring system allowing the craft to leave the stellar system.

    The craft might continually orbit the star with inceasing gamma factors thereby acting as a special relativistic gamma factor time machine for foward time travel. If the value of centripital acceleration increased to a great enough extent, then perhaps very significant general relativistic time dilation could also result.

    The way I see it, red giants might offer such advantages to any ETI and future human civilizations calling planets around such stars home. Aside from offering a source of warmth for terrestrial planets in the habitable zone, such stars could provide a tremendous source of energy for space transporation systems.

    I share your enthusiasm, Paul, regarding red giants and associated habitable zones. This stuff is really neat.



  • Ronald December 17, 2008, 5:47

    Adam, do I understand you well?:

    After Main Sequence, there is first Red Giant Branch (RGB), then about a 100 my of Helium Main Sequence (HMS), then Asymptotic Giant Branch (AGB) in which expansion and contraction takes place.

    If this is the case, then this seems hardly comforting to me: a planet first has to survive RGB, then it *only* has 100 my of HMS, then already there is the stellar hickup period of the AGB.

    Is this the right picture, or am I missing/misunderstanding something here? Where in this time scale are the several gy’s of HZ for a giant earth?

  • James M. Essig December 17, 2008, 11:04

    Hi Folks;

    Does anyone know what the absolute upper mass limit is for stars? I remember reading an article or paper suggesting that stars using ordinary stellar nuclear fusion fuel may have developed near the beginning of the star forming era with masses as great as 1,000 solar masses. Perhaps such a theory has since been discounted.

    Even a blue supergiant mass limit of 150 solar masses staggars the mind.



  • Ronald December 17, 2008, 12:13

    Ref. to James M. Essig (December 17th, 2008 at 3:06):

    “converting 4 trillion tons of matter into energy per second”

    Talking about Dyson swarms!

  • Doug M. December 17, 2008, 15:50

    Leaving the main sequence is a gradual process. It starts with a very gradual reddening, combined with a somewhat faster brightening.

    In the case of our Sun, over the next 6.5 billion years it will stay very nearly the same temperature while expanding slightly and increasing its luminosity to about 2.2 times its current value. (This is normal for main sequence stars as they age; the Sun has already increased its brightness by about 60% since it first started shining 4.5 billion years ago.)

    Then, after those 6.5 billion years have passed, the Sun will start to expand a bit faster while cooling. It will redden down to a surface temperature of about 4900 degrees, which would be about the temperature of a typical late-K orange dwarf. However, this will be very gradual indeed… the initial reddening will take about 700 million years. Over this period, the Sun’s brightness will increase by about 20%, from about 2.2 to 2.7 solar luminosities. This is a sort of stellar menopause, during which the Sun prepares to leave the main sequence.

    Then the long, slow increase in luminosity suddenly accelerates as it departs the main sequence and becomes a subgiant. In the case of the Sun, it will brighten from about 2.7 times its current luminosity to 34 times — about an eightfold increase — in maybe 500 million years.

    Next step: the Sun becomes a no-kidding red giant. Over the next 80 million years, it swells up to 160 solar diameters, cools down to 3100 degrees Kelvin (lower middle M class) and peaks at a whopping 2300 solar luminosities. Adam is correct that this period will last “only” a hundred million years or so.

    The earlier subgiant phase lasts longer, 500 million years. However, the Sun’s luminosity increases eightfold during that period, so it’s probably not long enough for significant evolution to take place on our hypothetical thawed super-Earth.

    (Numbers come from this paper: http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1993ApJ…418..457S. It is from 1993, but should still be broadly correct.)

    Now, a smaller star would go through the whole process more slowly. Start with a late G dwarf and you’ll spend several times as long climbing the subgiant branch and hanging around as a red giant. A couple of billion years as a brightening subgiant, a few hundred million burning helium: that starts to look promising.

    However, if you go past late G, the main sequence lifetime of the star becomes longer than the current age of the universe. D’oh. So, there has probably never yet been a start that’s stayed in the helium-burning “horizontal branch” stage for more than a couple of hundred million years.

    Someone mentioned Delta Pavonis. That’s a star that began life as an early G, very similar to the Sun, and is just leaving the main sequence for the subgiant branch now. So, there might be an icy super-Earth a couple of AU out, just now feeling the first wave of greenhouse warming!

    Doug M.

  • Ronald December 18, 2008, 5:35

    Doug, thanks for your very interesting contribution.

    However, in my knowledge Delta Pavonis (I was the someone that mentioned this star) is or started as a later G star: I have read spectral (sub)types from G5 to G8, which would also explain why it is now, being in a subgiant stage (IV), about 20% brighter than our sun.

  • ljk April 3, 2009, 13:30

    Is there a metallicity enhancement in planet hosting red giants?

    Authors: Pawel Zieliński, Andrzej Niedzielski, Monika Adamów, Aleksander Wolszczan

    (Submitted on 2 Apr 2009)

    Abstract: The Penn State/Toru\’n Centre for Astronomy Search for Planets Around Evolved Stars is a high-precision radial velocity (RV) survey aiming at planets detection around giant stars. It is based on observations obtained with the 9.2 m Hobby-Eberly Telescope.

    As proper interpretation of high precision RV data for red giants requires complete spectral analysis of targets we perform spectral modeling of all stars included in the survey. Typically, rotation velocities and metallicities are determined in addition to stellar luminosities and temperatures what allows us to estimate stellar ages and masses.

    Here we present preliminary results of metallicity studies in our sample. We search a metallicity dependence similar to that for dwarfs by comparing our results for a sample of 22 giants earlier than K5 showing significant RV variations with a control sample of 58 relatively RV-stable stars.

    Comments: 5 pages, 2 figures. To appear in “Extrasolar Planets in Multi-body Systems: Theory and Observations”

    Subjects: Solar and Stellar Astrophysics (astro-ph.SR); Earth and Planetary Astrophysics (astro-ph.EP)

    Cite as: arXiv:0904.0374v1 [astro-ph.SR]

    Submission history

    From: Andrzej Niedzielski [view email]

    [v1] Thu, 2 Apr 2009 12:11:13 GMT (60kb)


  • ljk August 9, 2009, 23:50

    Discovery of a close substellar companion to the hot subdwarf star HD 149382 – The decisive influence of substellar objects on late stellar evolution

    Authors: S. Geier, H. Edelmann, U. Heber, L. Morales-Rueda

    (Submitted on 7 Aug 2009)

    Abstract: Substellar objects, like planets and brown dwarfs orbiting stars, are by-products of the star formation process. The evolution of their host stars may have an enourmous impact on these small companions. Vice versa a planet might also influence stellar evolution as has recently been argued.

    Here we report the discovery of a 8-23 Jupiter-mass substellar object orbiting the hot subdwarf HD 149382 in 2.391 days at a distance of only about five solar radii. Obviously the companion must have survived engulfment in the red-giant envelope. Moreover, the substellar companion has triggered envelope ejection and enabled the sdB star to form.

    Hot subdwarf stars have been identified as the sources of the unexpected ultravoilet emission in elliptical galaxies, but the formation of these stars is not fully understood.

    Being the brightest star of its class, HD 149382 offers the best conditions to detect the substellar companion. Hence, undisclosed substellar companions offer a natural solution for the long-standing formation problem of apparently single hot subdwarf stars.

    Planets and brown dwarfs may therefore alter the evolution of old stellar populations and may also significantly affect the UV-emission of elliptical galaxies.

    Comments: 17 pages, 2 figures, ApJL accepted

    Subjects: Solar and Stellar Astrophysics (astro-ph.SR)

    Cite as: arXiv:0908.1025v1 [astro-ph.SR]

    Submission history

    From: Stephan Geier [view email]

    [v1] Fri, 7 Aug 2009 11:16:57 GMT (86kb)