≡ Menu

Problems with Red Dwarf Habitable Zones

Why all the fuss about red dwarf planets? We’re seeing so much ongoing work on these worlds because when it comes to terrestrial-class planets — in size, at least — those around red dwarfs are going to be our first targets for atmospheric characterization. A ‘habitable zone’ planet around a red dwarf throws a deep transit signature — small star, big planet — so that we can use transit spectroscopy to puzzle out atmospheric components. Getting an actual image would be even better, and modifications to the VISIR instrument at ESO’s Very Large Telescope, a project Breakthrough Initiatives is involved in along with the ESO, could eventually yield such.

We’ll know a great deal more about the possibilities as new missions come online, but for now, researchers are doing their best to apply models to what we know and deduce what surface conditions may be like around stars like TRAPPIST-1 and Proxima Centauri. Some of these results are not auspicious if it’s life we’re looking for. I’m looking at two papers from Chuanfei Dong (Princeton Plasma Physics Laboratory/Princeton University) that assess potential habitability, and in both cases there are significant reasons to question its likelihood.

Image: Princeton’s Chuanfei Dong. Credit: Princeton Plasma Physics Laboratory.

The assumption here is that an atmosphere must exist for long timescales to allow habitable conditions on the surface, and those long time-frames are precisely what is in question. Published earlier this year in the Astrophysical Journal Letters, the first of these papers develops models of the stellar wind, that outflow of charged particles that, in our own Solar System, defines the heliosphere around our Sun. This paper deals with the situation at Proxima Centauri b. The second paper, not yet published but available as a preprint, extends the study to the TRAPPIST-1 system, with results that are equally challenging for life.

While we have tended to focus habitability studies on factors like surface temperature (and this, in turn, is much dependent on atmosphere), Dong and colleagues are concerned about the effects of the stellar wind and atmosphere retention. Their simulations, performed using magneto-hydrodynamical (MHD) modeling that was originally developed for Venus and Mars, allow them to compute ion escape losses that would be expected at Proxima Centauri b.

Specifically, the Proxima paper examines the electromagnetic erosion of the atmosphere given the photo-chemical effects of the stellar wind, finding that the stellar inflow can ionize atoms in the planetary atmosphere, allowing these electromagnetic forces to sweep them into space. The result: Potentially severe atmospheric loss that would deplete the atmosphere of evaporated water, a cycle that could eventually leave the planetary surface dry.

A sufficiently high stellar wind pressure could cause extensive atmospheric loss, making any surface-based life that emerged a short-lived phenomenon. Says Dong:

“The evolution of life takes billions of years. Our results indicate that Proxima Centauri b and similar exoplanets are generally not capable of supporting an atmosphere over sufficiently long timescales when the stellar wind pressure is high. It is only if the pressure is sufficiently low, and if the exoplanet has a reasonably strong magnetic shield like that of the Earth’s magnetosphere, that the exoplanet can retain an atmosphere and has the potential for habitability.”

Image: Is the stellar wind capable of reducing a planetary atmosphere to the point where surface life is impossible? Credit: NASA/JPL-Caltech.

The paper finds that ion escape rates at Proxima b are two orders of magnitude higher than the terrestrial planets of our Solar System, assuming that the planet is unmagnetized. But even in the presence of a magnetosphere, ion escape rates are still higher than any we see in our system’s planets. The same issues apply at TRAPPIST-1, as noted in the new paper, which implicates stellar wind values as the primary driver in ion escape:

…as seen from our Solar system, the ion escape rates for Venus, Mars and Earth are similar despite their compositions, sizes and magnetic field strengths being wildly dissimilar (Lammer, 2013; Brain et al., 2016), thereby indicating that the ion escape rates may be more sensitive to stellar wind parameters; this is also partly borne out by the atmospheric ion escape rate calculations for Proxima b (Airapetian et al., 2017; Dong et al., 2017). Second, we observe that the inner planets of the TRAPPIST-1 system could have experienced significant losses of H2 and water over fast timescales (Bolmont et al., 2017; Bourrier et al., 2017), leaving behind other atmospheric components…

Dong also notes that planets close enough to be in a red dwarf’s habitable zone are likely tidally locked, producing constant bombardment on the star-facing side that would intensify the effects of atmospheric loss whether or not the planet has a protective magnetosphere. Earlier work has suggested that tidally locked planets are unlikely to have more than a weak magnetic field.

I’m focusing now on the TRAPPIST-1 paper because it’s the latest work on this matter, and it amplifies what was found in the earlier Proxima Centauri b work; I give citations for both papers below. Considering TRAPPIST-1 in light of stellar winds, the researchers argue that TRAPPIST-1h, viewed purely from the perspective of atmospheric loss, is the one most likely to have retained its atmosphere, but this is not a world where liquid water is possible on the surface. Dong and colleagues believe that TRAPPIST-1g thus represents “…the best chance for a habitable planet in this planetary system to support a stable atmosphere over long periods,” as the stellar wind effects diminish with distance.

It’s not a pleasant picture for those hoping for clement conditions on other planets around TRAPPIST-1 or Proxima Centauri. Oceans may once have existed there, but this work suggests that their surfaces today are probably dry. While the two papers focus on Proxima b and the TRAPPIST-1 worlds, Dong notes that the newly discovered planet around Ross 128 may have better prospects, as its star appears to be quieter than Proxima Centauri or TRAPPIST-1.

We should also note that an atmosphere can be replenished by outgassing, a reminder that analyzing an atmosphere over billion-year timeframes demands, as the paper notes, “…an in-depth understanding of the interplay between source and loss mechanisms.” Another issue: Stellar properties evolve, so that atmospheric escape rates change. This may not work to life’s advantage, however, for pre-main sequence M-dwarfs, according to Dong’s simulations, would produce even stronger stellar wind effects upon a young planet’s atmosphere.

The implications for other planetary systems seem clear:

For a given star, the mass-loss rate is fixed, implying that the escape rate is lower for smaller planets orbiting at greater distances. Hence, we suggest the following strategy for prioritizing studies of multi-planetary systems. If more than one exoplanet resides within the HZ of a given star, it may be more prudent to focus on the outward planet(s) since the atmospheric escape rates are likely to be lower. Similarly, when confronted with two planets with similar values of Rx and a [radius and semi-major axis], we propose that searches should focus on stars with lower mass-loss rates and magnetic activity.

The examination of the effects of stellar winds will proceed with soon-to-be-launched missions like the James Webb Space Telescope, allowing us to put theoretical work to the test. Given the surprises we’ve consistently found even with our interplanetary probes, Pluto being the most recent example, we can fairly confidently expect to modify our views with each new exoplanet atmospheric characterization. For now, though, these continuing studies raise serious questions about red dwarf planets as a clement venue for life.

The papers are Dong et al., “Is Proxima Centauri b habitable? — A study of atmospheric loss,” Astrophysical Journal Letters Vol. 837, No. 2 (10 March 2017) (abstract/ preprint); and Dong et al., “Atmospheric escape from the TRAPPIST-1 planets and implications for habitability,” accepted at Astrophysical Journal Letters (preprint).


{ 35 comments… add one }
  • Ashley Baldwin December 1, 2017, 13:32

    Recommend prioritisation ?!! Please . Beggars can’t be choosers . In terms of rocky planets , JWST and the ELTs represent the technological edge for the foreseeable future and betwen them will at best only be able to do meaningful atmospheric characterisation ( or not ) on a few dozen nearby ( likely mid to late ) M dwarf systems . Certainly for “hab zone” planets – however hypothetical . Any next generation space or OWL telescope is likely more than twenty years away and some .Optimistically .

  • ljk December 1, 2017, 13:41


    The paper’s conclusion in the abstract:

    Our results suggest that active M dwarf hosts may comprehensively destroy ozone shields and subject the surface of magnetically-unprotected Earth-like planets to long-term radiation that can damage complex organic structures. However, this does not preclude habitability, as a safe haven for life could still exist below an ocean surface.


    • Ramses Ramirez December 2, 2017, 8:48

      That Tilley et al. statement is strange in light of the group’s previous work (e.g. Luger and Barnes, 2015) and that of my own (Ramirez and Kaltenegger, 2014).

      It is very difficult for habitable zone planets around active M-dwarf stars (like TRAPPIST-1 or AD LEO) to even have an ocean to begin with. They would have likely lost it in a runaway greenhouse during the pre-main-sequence phase, when the star was much more radiatively active than today. Such planets would have received radiative fluxes and energies that dwarf even those Venus had received.

  • galacsi December 1, 2017, 14:11

    How is the speed hence the strength of the stelar wind compared with the speed of the solar wind ? A slower wind could be much less of a menace to the atmosphère.

  • ole burde December 1, 2017, 15:57

    One factor would work in the opposite direction : If the planet has a very deep ocean to begin with , such as hundreds of kilometers deep , there could exist a longterm balance between evaporation and atmospheric loss to the solar wind …and longterm balances has a strange tendency of balancing them selves even more by causing secondary mecanisms which stabilizes the system even more …..this would eventualy have an unhappy end , but on Mars it took at least a billion years to loose most of its surface water …so the AGE of a red dwarf planet could in some cases be an all-important variable , and even more so if the oxygen acumulates in the atmosphere after the Hydrogen is blown away …so , if proxyma centaury b had 10 times more water to begin with than Mars , it might still be alive ..

  • djlactin December 1, 2017, 17:17

    “The habitable zone” concept assumes implicitly that life lives on the surface, exposed to solar flux. Ain’t so. Much of the Earth’s ecosphere is subterranean, often deeply so: https://microbewiki.kenyon.edu/index.php/Deep_subsurface_microbes
    I am an accredited (R) scientist, but my (unscientific) intuition is that life will occur wherever sufficient (Gibbs) free energy is available.

    • Ramses Ramirez December 2, 2017, 19:07

      It is certainly true that subsurface life may be thriving on some planets (even Mars), djlactin. However, the habitable zone is defined in terms of surface conditions because there is no (known) way in which subsurface life can be detected remotely, at least not with our current technology (and if there is, I’d love to know!). Current working definitions of the habitable zone assume that surface biota may leave detectable atmospheric biosignatures.

  • P December 1, 2017, 18:21

    Bit of a shame that Prox is an M5.5 dwarf and not an M0 or M1 dwarf. I often think we need 2 spectral classes to cover the red dwarfs given the enormous difference between an M0 dwarf and an M8-9 ultracool dwarf. Perhaps a good boundary might be between stars with radiative cores and fully convective ones?

    Anyway, what about secondary atmospheres formed by long running vulcanism? Maybe early red dwarfs stripping dense primordial atmospheres isn’t such a bad thing, assuming they are replaced once the star quietens down?


    • andy December 2, 2017, 10:08

      Spectral classifications are just that, classifications of the star’s spectrum. Any transition between spectral classes would have to be linked to some observable feature of the spectrum, rather than properties of the stellar interior.

    • Ramses Ramirez December 2, 2017, 19:20

      We do have a bit of a spread in known M-dwarfs.. For instance, TRAPPIST-1 (~M8) , Prox Cen (M5.5), AD LEO (M3.5)..etc. But, we will also need to learn more specific details about these stars and that is (unfortunately) a bit hard to come by.

      Some of us do argue that perhaps desiccated M-dwarf planets could get replenished somehow, perhaps through later water-bearing comet/asteroid impacts. However, if such a mechanism works as did the late heavy bombardment in our solar system did, it would require a largish planet (perhaps at least a Neptune-sized planet, maybe a Jupiter) to trigger such impacts. Plus, it is thought that only small water reservoirs could be brought in this way. Thus, the planet may actually end up looking more like Dune than the Earth. Is such a “Dune planet” habitable? That is an open question.

  • Rangel December 1, 2017, 19:55

    Interesting, though i still believe the rate of atmosphere depletion may be more complex than we know so far, well we will see once we gain data about it’s atmosphere.

    Still TRAPPIST-1g being the most promising for retaining an atmosphere is nice, and indeed it is the bigger one, unfortunately that planet seems to be too distant from the confortable zone of the HZ, though there is several possibilities for temperatures depending on the atmosphere, there still a chance the centre of the daylight zone is warm enough for life to arise, though it’s low.

    Also even ‘g’ being the most promising for an atmosphere, i am pretty sure ‘f’ still in a good position and i believe has a better chance to actually be habitable on the dayside. ‘e’ planet is the perfect target regarding it’s position on the HZ, maybe we are lucky enough and it retained it’s atmosphere long enough? We will see.

  • Brad in Dallas December 1, 2017, 21:48

    We could hope for a planet with the right combination of a dozen or more factors—hab zone, atmosphere, gentle sun, magnetosphere, large satellite to stabilize tilt, etc etc., or we can learn to build Island 3 habs that can be built around any star that will power solar panels, and choose the shielding, centrifugal g-force, atmosphere and orbit we want. Gravity wells are for prisoners.

    • Craig A. Glesner December 3, 2017, 20:16

      “Gravity wells are for prisoners.”

      I want that on a t-shirt!

    • Greg in Nederland December 4, 2017, 19:39

      Do you think that’s what this fuss is all about? People looking for a planet to occupy? Before we ever get to that, we’ll have large colonies in space. People will be happy there. Living on a planetary surface will be considered an extreme outback experience. Any planets nice enough to colonize will likely already have been set aside as a planetary nature reserve (which I believe is what our Earth already is).

    • ijv December 4, 2017, 21:45

      When it comes to interstellar travel, I’ve never understood this obsession with gravity wells. Assuming currently known physics, even an interstellar crawl of 0.1% c, implies a dV of 600 km/s.

      So, if little furry creatures from Proxima Centarui arrived in our solar system tomorrow, dying to stretch their little furry legs after a 4000 year long trip. Their response to any questions about gravity wells is likely to be a long the lines of “I make (disgusting alien noises) at your puny gravity wells!”

  • Ramses Ramirez December 2, 2017, 8:30

    In my view, the biggest issue with M-dwarf planetary habitability is for planets to keep their habitability intact throughout the superluminous pre-main-sequence stellar phase. However, this is very difficult because the X-ray and EUV fluxes would have literally been orders of magnitude higher billions of years ago than they are today. For this reason, it is quite possible that none of the TRAPPIST-1 planets located in the traditional HZ today have substantial surface water inventories. TRAPPIST-1h, which sits outside the traditional HZ, is the best bet (in my opinion) because it would have received the least amount of radiation. As the authors mentioned, perhaps our proposed mechanism of volcanic hydrogen (Ramirez and Kaltenegger, 2017) may have helped foster habitable conditions on that planet. Even then, TRAPPIST-1h would have had trouble retaining its atmosphere or water eons ago with a much more active young star.

    • Ashley Baldwin December 4, 2017, 14:32

      I thought the resonant chain seen with the TRAPPIST planets was suggestive of formation much further out and later inward migration ? This could help keep some of the planets out of harm’s way during the worse of both the pre and post ZAMS active phases . In conjunction with the greater starting load of volatiles received so far out it would also support their ongoing retention .

      • Ramses Ramirez December 5, 2017, 2:08

        Hi Ashley. The idea that the TRAPPIST-1 planets had formed much further out and came inward later after stellar conditions became more clement is just a suggestion. Moreover, this “habitable evaporable core” hypothesis relies on a very special set of circumstances to try to argue for habitability. It is just as (if not more) likely that these worlds had formed in-situ and lost their volatiles when the star was very young. The radiation environment around an M8 like TRAPPIST-1 would have been especially severe, possibly supporting harsh conditions in its “inner stellar system” for over 2 billion years. Thus, even if we assume that the TRAPPIST-1 planets had migrated in later, the migration needed to have occurred very late (which is hard with the gaseous disk long gone) in order for some of those worlds to be potentially habitable today.

        • Ramses Ramirez December 5, 2017, 2:15

          One exception to what I just said is if the migrating planets had soooo much water (tens of percent of their planetary mass) that even a runaway greenhouse lasting many millions of years wouldn’t completely desiccate them (e.g. Levi et al. 2017). However, can such ocean world planets (that lack a solid surface) be habitable? That is open to much debate.

          • Ashley Baldwin December 5, 2017, 14:27

            Thanks Ramses. As ever.
            I’m sure that’s right. The recent 55 Cancri planetary data would seem to suggest that “evaporated cores ” do atleast exist if not necessarily habitable ones .

            It’s ironic that the nomenclature debate is over whether the Red dwarf “class” should include late K dwarfs as well as M dwarfs . In terms of habitability atleast I would posit that there should probably be a classification dividing line somewhere between early and late M dwarfs . Specifically between those that have convective envelopes (behaving more like the larger Sun like stars that the whole hab zone concept best suits ) and those that don’t. Circa M3 perhaps . This would push the traditional hab zone out to a far more respectable 0.2AU or further and meaningful magnetic breaking would help reduce chromospheric/ flare/XUV activity et al within a reasonable time.

            • Ramses Ramirez December 5, 2017, 21:35

              Ashley, the 55 Cancri planetary data also suggest that such planets may actually be “lava worlds”, terrestrial planets engulfed in magma oceans (e.g. Hammond and Pierrehumbert, 2017). To me, this seems more likely.

              I would just add that LHS 1140 seems relatively quiet (at least today) and it is a M4.5. But I do think that studies of the habitable zone should be very clear that there are big differences among stars not only on a spectral class level, but individually.

              • Ashley Baldwin December 6, 2017, 19:28

                I think you make an excellent point. An oft forgotten benefit of Kepler and indeed precision time series photometry has been the seminal insight into stellar activity via extended asteroseismology. Know the star to know that planet, absolutely ,but there is enormous benefit from just knowing a lot about a lot of very different stars . I think this will prove even more crucial with PLATO and I sincerely hope it can last considerably longer than its four year primary mission. Ideally looking at the same field , not just to refine the ephemerides of transiting planets but to give a better insight into longer term stellar cycles. I’m sure there will be plenty of surprises . It would have been very interesting to have had uninterrupted observational data of “Tabby’s star” over say an eight year period.

      • Nicky December 11, 2017, 2:41

        You made a really good point here, since planets formation plays a huge role in determining habitability. But you have misunderstood the process of inward-migration. Planets could only migrate before the gaseous disk dissipates during the interaction with gases, which is well before the end of PMS (pre-main sequence), probably within the first 10 myr. Regardless the migration process, the planets still would have to go through the PMS.

    • Nicky December 11, 2017, 4:48

      Hi Ramirez. I would like to argue that XUV and PMS (pre-main sequence) in TRAPPIST-1 case might have only minor effect on the water reservoirs of HZ planets. Based on the current planets formation theory, TRAPPIST-1 planets were formed near or outside the snowline (where volatiles condense into solids) in the protoplanetary disk, and they migrated to current orbits. The HZ planets probably have very high water mass fractions (something like 50% for outer planets and 10% for inner planets) and low density, because the building blocks were largely composed of condensed volatiles (ice). Though the planets mass is not well constrained, they sum up to low average densities, which means large fraction of water is retained. But any detectable life in these ocean worlds is doubtful, and life origin is probably already a problem.

  • Michael December 2, 2017, 10:47

    If the rotation rate of the RD is slow then the violent outbursts are less, so if a large planet is in orbit it could rob some of the rotation momentum. Perhaps RD’S will need a large massed planet in orbit to tame the fiery tempered star to allow habitable inner planets.

  • Michael Fidler December 2, 2017, 11:29

    Ah, but the universe is far smarter then we are, after all it created us!
    Here is an example:

    Aliens in our midst.
    “The ctenophore’s brain suggests that, if evolution began again, intelligence would re-emerge because nature repeats itself”.

    “Leonid Moroz has spent two decades trying to wrap his head around a mind-boggling idea: even as scientists start to look for alien life in other planets, there might already be aliens, with surprisingly different biology and brains, right here on Earth. Those aliens have hidden in plain sight for millennia. They have plenty to teach us about the nature of evolution, and what to expect when we finally discover life on other worlds”.

    • Alex Tolley December 3, 2017, 18:55

      It is rather hyperbolic to call the nerve net of a ctenophore a brain. Granted comb jellies evolved early, preceded by sponges, and that they evolved neurons, but to suggest that this indicates intelligence always evolves is going way beyond the evidence. Only humans evolved the intelligence to create the civilization we have. Other groups, both with and without backbones have been around for tens of millions of years but never evolved the intelligence needed to create the cultural “takeoff” that was the basis of civilization. We still don’t know whether it was a fluke that put our hominin branch into the ascendant.

  • Michael December 2, 2017, 18:05

    I have mentioned this before that if a large hurricane forms on the star facing side it could push water vapour high into atmosphere where ozone offers less protection allowing it’s breakdown.

  • Adam Crowl December 3, 2017, 7:11

    Does seem to steer the Quest for Life back to FGK stars – on average. No doubt there are red dwarf habitable planets, just not the majority of red dwarf “temperate zone” planets.

    • Alex Tolley December 3, 2017, 18:59

      Should it turn out that life can emerge in the subsurface oceans of icy moons, that should not be a problem around RDs.

      I’m very skeptical that life can emerge on such moons ( it is just more titillation to get funding for probes), but the point is that such environments should exist around RDs and be impervious to the conditions that could ruin a world in the HZ for terrestrial life.

    • Ronald December 7, 2017, 8:01

      Indeed, it seems like that, with M dwarf planets being plagued by so many plagues and increasingly so the more we discover about them: tidal locking, super-flares, stellar wind resulting in atmospheric and water loss, pre-main sequence misbehavior, …

      That’s why I have been a solar chauvinist for some time now. Not strictly solar twin (which may not even be the optimal), but solar type broader sense (roughly F7/9 – K2/3). I think it is no coincidence that we are orbiting a quiet G-star.

  • Rob Flores December 5, 2017, 3:53

    My guess is that Icy moons with large Radioactive cores
    will be as rare as Twin Earth’s. Although one might classify
    Titan as one of such worlds, it falls short. I can’t see an energy
    source that would drive complex cellular life below the Ice Crust on Titan

  • Ronald December 5, 2017, 4:23

    Two related topics in the news:

    – There Could be Hundreds More Icy Worlds with Life Than on Rocky Planets Out There in the Galaxy:
    That would most likely be microbial life and rather poor in energy and nutrients.

    – “The Genesis Project: Developing Ecospheres on Transiently Habitable Planets“:

    Seeding potentially suitable planets with photosynthetic Prokaryotes (cyanobacteria) and simple Eukaryotes.
    That is what I call ‘Terraseeding’.

Leave a Comment