Long-time Centauri Dreams readers will be familiar with the work of Manoj Joshi and Robert Haberle. Back in the 1990s when both were at NASA Ames (Joshi is now at the University of Reading), the scientists went to work on the question of whether planets around red dwarf stars could be habitable, given the problem of close orbits and tidal lock. Simulating the atmosphere of such a planet, they found even a thin atmosphere would circulate globally, moving enough heat to prevent the air on the darkside from freezing out. The prospect of a planet with oceans and a climate mild enough to support life began to look more promising.
Joshi and Haberle have a new paper out that looks once again at planets around red dwarfs, this time extending the possible habitable zone to a greater distance from the star. M-class red dwarfs are smaller and cooler than G-class stars like the Sun, and emit a much larger fraction of their radiation at longer wavelengths where the reflectivity of ice and snow are lower. The effect is striking, as the paper notes:
The values for snow and ice for a planetary surface orbiting the Sun are 0.8 and 0.5 respectively, which are broadly consistent with the values that are used in climate models. Fresh snow and ice albedos on a planet receiving black body radiation from an object at 3300K are 0.6 and 0.3 respectively, which are significant reductions from the “solar” values.
The black body radiation is an idealised representation of an M-dwarf that is approximately 40% as massive as the Sun. The authors then go on to calculate the albedos for snow and ice on hypothetical planetary surfaces around the stars Gliese 436 and GJ 1214 and find them even lower. What this means is that more of the long-wave radiation emitted by an M-dwarf will be absorbed rather than reflected by an icy surface. The effect is to widen the habitable zone of a planet around this kind of star outwards by anywhere from 10 to 30 percent. What might have seemed a frigid world now looks more hospitable.
Allowing a habitable zone (defined here in terms of liquid water at the surface) to exist farther out from the parent star is significant, although there are no changes at the other end of the HZ:
The effect considered here should not move the inner edge of the habitable zone, usually considered as the locus of orbits where loss rates of water become significant to dry a planet on geological timescales (Kasting et al 1993), away from the parent M-star. This is because when a planet is at the inner edge of the habitable zone, surface temperatures should be high enough to ensure that ice cover is small. For a tidally locked planet this implies that ice is confined to the dark side that perpetually faces away from the parent star: such ice receives no stellar radiation, rendering albedo effects unimportant.
But planets with significant amounts of snow and ice will have higher surface temperatures and the outer edge of the habitable zone is extended. It’s an interesting thought because of the sheer ubiquity of red dwarfs — some estimates of their prevalence run as high as 80 percent of main-sequence stars, so seeing them as astrobiologically friendly would revise our estimates for extraterrestrial life. Just how close the nearest life-bearing planet might be affects our planet hunt astronomically as we choose targets and has significance for any future interstellar probes.
Other issues remain and are under active investigation, from the problem of flares to climate models involving the effects of clouds and water vapor. A good place to get an overview is Tarter et al., “A Reappraisal of The Habitability of Planets around M Dwarf Stars,” Astrobiology 7, pp 30-65 (2007). The Joshi and Haberle paper is “Suppression of the water ice and snow albedo feedback on planets orbiting red dwarf stars and the subsequent widening of the habitable zone,” accepted by Astrobiology (preprint).
Still doesn’t change the fact that class M are flare stars and will wipe out life each time it tries to start up.
To be close enough the planets will be tidaly locked so won`t have a magnetosphere.
We await results from the Gaia
satellite. It was slated for launch in
2013, http://sci.esa.int/science-e/www/area/index.cfm?fareaid=26.
Gaia will provide a reliable catalogue of G dwarfs to
well beyond 500 pc, and early- and mid-type M
dwarfs to 100–200 pc. Unfortunately, the catalogue
will not come out until approximately
2018, at the earliest.
Recent work on stellar flares and the possible adaptability of life on planets near M-dwarfs:
Scalo, et al., “M Stars as Targets for Terrestrial Exoplanet Searches And Biosignature Detection,” Astrobiology 7: 85-166 (2007).
Segura, et al., “The Effect of a Strong Stellar Flare on the Atmospheric Chemistry of an Earth-like Planet Orbiting an M Dwarf,” Astrobiology 10: 751-771 (2010).
And this earlier quote from John Scalo: “What might life be like on a planet orbiting a red dwarf with powerful flares and continuous intense coronal X-rays? One possibility is that most of the biosphere would need to be underground or underwater; another is that the challenging mutational radiation environment would accelerate the evolution of life.”
http://www.spacedaily.com/news/life-02c.html
Regarding magnetic fields on slowly-rotating planets, consider Ganymede (rotation period 7.15 days) and Mercury (rotation period 58.6 days). Slowly-rotating planets can definitely sustain geomagnetic fields. Of course in both of these cases the fields are weak, but these are quite small planets. I haven’t yet seen any definitive statement that an Earth in the HZ would lack a geomagnetic field.
I find the question of accelerated evolution added by Paul in a later comment particularly interesting. Sure, this can lead to novelties that would normally be impossible in environments were high selective pressure should more typically result in a purifying effect, but this increased mutation rate should lead to the degradation of far more genetic material than that it can create.
To me the advantage here is that the radiation dose is highly location dependant (Arsirus’ first sentence seems to have completely missed that different substellar hemisphere locations experience different levels of radiation and half the planet is available for life to hide out in somewhere and repopulate if emergency requires it). Once sexual reproduction is established one can imagine that crossbreeding communities that have experienced different radiation effects might be able to keep both efficiency and novelty. At least that’s my first impression of this situation.
Actually, in general, how confident can we be that these planets are not significantly more conducive to the development of higher life than Earth-like ones?
Arsirus, any organism that can use ionizing radiation would have a significant evolutionary boost from the enhanced flux at such frequencies. As for absent magnetic fields, that’s a geophysical question that modeling has generally said is perhaps mistaken. Tidal-locking also doesn’t necessarily imply a synchronous orbit – Mercury-style end states are as likely.
Surely its way too early to be dogmatic about this one way or another. What we can say is that we are going to find lots and lots of planets around M dwarfs!
That’s interesting in itself. Beyond that – do we really know what percentage of M dwarfs are flare stars, and for how long (I was under the impression that early M dwarfs that were NOT fully convective tended to be quiet)? Its going to be a long time before we know how many M dwarf planets really are tidally locked, rather than being in another orbital resonance. Finally, the magnetic field question, here I find the paper by:
Barnes et al. “Habitability of Planets Orbiting Cool Stars”, arXiv:1012.1883v1
very interesting. It suggests that tidal locking might be less important than planetary mass, size and internal composition. Quote “…the magnetic moment of a planet does not depend on its rotation rate…” p9
I too wait for Gaia, for the new HARPS north (at least that’s going to happen!), and for any new development in IR spectroscopy that will help us to discover systems around cool stars and brown dwarfs (now there’s a really interesting possibility!)
P
@Arsirus
Life on a world that experienced intense solar flare storms will have to adapt. Maybe it will live underground in caves or underwater. Maybe the aliens will have survival methods- like hiding underground or hiding eggs/spores/etc. in shielded locations. I’m sure SF writers can dream up plenty of ingenious solutions to the unique challenge of surviving on a habitable planet orbiting a red dwarf!! And, as andy pointed out, we have not yet seen a definitive statement that a habitable planet in the HZ of a red dwarf will lack a geomagnetic field.
Things do look a bit better already. For one thing, even a thin atmosphere will move heat around enough to keep the back side from freezing over. The HZ of a red dwarf is larger than expected due to the decreased reflectivity of infrared light off of ice. Life is adaptable- there are fish that survive in sub-zero conditions due to having antifreeze in their blood. There are bacteria that live underground and eat rock. Their are bacteria that resist radiation damage. Some form of life could exist on a planet orbiting a red dwarf. In fact, the increased radiation might speed up the development of complex lifeforms.
If red dwarfs can support habitable planets, the estimates for life in the cosmos will have to be changed. Red dwarf stars are the most common type of star in the universe. Still, life on a planet orbiting a red dwarf is likely to be difficult simply due to the variability in their output. Going from massive star-spots to solar flares can’t be fun!!
What do you think life on a habitable planet or moon orbiting a red dwarf might be like for humans? They may not be their by choice- maybe they had to evacuate an ailing starship or flee to an unlikely world to escape the attention of a stellar empire.
I think life would be somewhat difficult. The humans will have to locate their buildings and bases under the ground to avoid dangerous flares and radiation. Earth plants will probably not grow, as most of the light from a red dwarf is in the infrared but Earth plants mostly use the visible region of the spectrum. Maybe genetically altered plants could grow…
Travel on the surface might be hazardous due to radiation and flares. I wouldn’t want to strike it out on foot- I’d use armored rovers with the capability to travel on both land and water. We might have to cross over seas or oceans. Hopefully micro-nuclear power plants are available, or something as good- I don’t think there will by any gas stations on the way. Maybe the rover will have enough shielding if the star flares…
I imagine a strange scene- entrances to dug-in buildings poking out of a rocky landscape scattered with native plants raising black leaves to a dim red sun. In the nearby spaceport, the reddish light reflects off of the burnished metal of spacecraft parked on the packed dirt of a large field. A few human ground vehicles trundle through the rugged landscape. A few figures gather near a spacecraft. A craft rises from the launch areas on the ionized glow of its plasma engines. A few humans survey the whole scene from a high outcropping of rocks. All of them carry personal microcomputers and sidearms- the native lifeforms are both deadly and opportunistic.
On the rotation issue, does this extend the habitable zone enough that the planets might be outside of the tidal-lock radius? At the very least, they could have other orbital couplings than 1-to-1, like Mercury’s 3-to-2. Not to mention that in a red star’s habitable zone, the “year” length could be as little as a week or two. Even a 1-to-1 tide-locked planet around a sufficiently dim red dwarf could have a rotation period of a matter of days.
On the flare-star issue, my understanding was that young red dwarfs were flare stars, but the flaring subsided over time. And I agree with the Scalo quote: on a planet more heavily irradiated than ours, life would probably just evolve better adaptations to radiation.
Earth’s atmosphere blocks a high proportion of incoming IR radiation, a much lower fraction of visible light. The greenhouse effect depends on this difference in atmospheric opaqueness for incoming and out going radiation. If we have a redder star this probably simplistic summation would suggest a weaker GH effect and so a narrower habitable zone.
The HZ of a typical M dwarf (1% solar luminosity) was previously considered to be only about 10% of the HZ of our own sun. So, even with the present extension, that would still be only 11-13% of our sun’s HZ.
Let’s assume the ‘average’ solar type star (around G5-G8) to have 65% of solar luminosity, resulting in an HZ of 80% of that of our sun.
All G stars amount to about 7% of the galactic stellar population, all solar type (from F9 through K2) about 12%.
That means, that, if we take a generous 13% of solar HZ for the average M star and assume an equally generous 80% of stellar population to be M stars, even then the *total amount of M star HZ* is hardly more than the total amount of solar type HZ (less than 10% more).
That combined with the frequency of M flare stars, tidal locking and possible relative paucity of planets near M stars, does not make M stars a particularly favorable class for astrobiology, despite their abundance.
The only real advantage of M dwarfs is their extremely long stable lifespan. But even that advantage may be relative: the fact that the HZ of the brighter solar type stars gradually moves outward means that more planets have a chance of spending sufficient time in the HZ, at least for primitive life to arise. Furthermore, even a very long M stellar lifespan is of little value if the planet itself stops being geologically active, effectively becoming a dead planet.
On the subject of flares from M class stars, I seem to remember reading on this site a subsequent paper that indicated the rate of flaring tended to decline with age – a very brief search has failed to find the reference however. Combining that point with the longevity of red dwarves and the possibilities around underground / underwate biospheres etc. and I suspect they remain potentially of astrobiological interest, but I’d be very interested in any views on that.
My impression, more generally, as an interested observer of this debate, rather than a professional astronomer or astrobiologist, is that the field is moving with astonishing speed at the moment in terms of the data available for assessing the likely frequency of habitable planets in the galaxy or wider universe. Work such as the subject of this post and the wider recent discoveries around red dwarves and the ongoing expansion of the dataset on exoplanets more generally are going to require a complete reworking of earlier models fairly soon.
Ronald: How do you quantify the HZ? 10% of solar HZ in terms of, what, absolute distance? relative distance? area?
Also, what do you mean by geologically active? Plate tectonics? Are you suggesting that lack of plate tectonics means the end of life? Why?
@Christopher L. Bennett: “On the rotation issue, does this extend the habitable zone enough that the planets might be outside of the tidal-lock radius?”
Interesting point and relevant, given tidal locking (speed) is inversely related to the *6th power* of distance. So, 1.3 times as far means almost 5x as long time period before being tidally locked.
Then again, since a typical M dwarf of 1% luminosity would (originally) have its midpoint of HZ at about 0.1 AU, even at 0.13 AU tidal locking will take place rather fast.
With regard to flares, mutations and natural selection: I agree with Rob Henry that intense and suddenly occurring flares, during which the stellar luminosity, and particularly its UV and X-ray, increases to tens of times normal (e.g. in the case of UV Ceti 75 times), will probably erase most life.
I do not agree that this would lead to more (beneficial) mutations. Very abrupt and intense fluctuations are mostly detrimental to life, without it getting much chance to adapt to it. It’s an error to think that higher selective pressure will automatically lead to more or faster evolution. If the selective pressure is too high, it will have an impoverishing effect, or even lead to extincton.
Christopher Phoenix said: “Their are bacteria that resist radiation damage.”
Quite right. Deinococcus radiodurans has proteins that repair radiation damage to its DNA very effectively. If that ability exists here on Earth, it could certainly exist on a more radiation-rich world, and organisms with such repair mechanisms would outcompete those without them.
For that matter, it’s conceivable that we could use gene therapy to give those same repair proteins to humans and enable us to survive in more radiation-rich environments such as red-dwarf planets. Heck, we’d probably need enhanced radiation resistance just to live in space generally, let alone to make a lengthy journey through interstellar space, bombarded by cosmic rays the whole time. If we could surmount the problem of getting there, we could manage to live there.
Eniac: I quantify the HZ as a % of our own HZ, so that would be absolute distance (expressed as AU). Since all (diffuse) irradiation decreases as an inverse square power (i.e. twice distance = a quarter of insolation, half distance = four times insolation), a typical red dwarf with 1% of solar luminosity would have the square root of that, i.e. 10%, of our HZ.
Actually less, because (much) more of it is not visible light but IR, which does not easily penetrate the atmosphere, as Andrew W rightly suggested.
A geologically active, ‘living’, planet: yes, plate tectonics and volcanism, and the interior still sufficiently hot to maintain those. Why? Because without volcanism, not enough of some essential elements, foremost C, are recycled to the atmosphere and biosphere.
Ronald: why do you think the absolute area of the HZ is relevant? The distribution of planetary orbital periods is not uniform. A uniform distribution would result in the expected number of planets between, say 1 and 2 AU being the same as the expected number of planets between 11 and 12 AU, or 101 and 102 AU. This is clearly not what is observed: closer in the planets are packed tighter. So while the absolute area of a red dwarf HZ is indeed smaller than the HZ of a solar type star, it is closer in, so the probability of finding a planet there is not going to be so much different to the probability of finding a planet in a solar-type star’s HZ. On a logarithmic scale they are fairly similar.
To Andy and Ronald, the longstanding argument over the area of the potential habitable zones around red dwarfs predates the availability of any data over the actual distribution of planets among such stars. The real question then is: do we now have enough data to ditch that line of argument?
The combination of high mutation rates and low selective pressure should be disastrous. At the other end, as Ronald has pointed out, very high selective pressure tends to strip a genome its flexibility to remain active outside the (narrow) confines of its present habitat. I suspect though that there is much in between, and even though these effects would retard life in most (if not all) environments, it does not follow that this disproves the contention that that such biospheres experience accelerated evolution. Also, towards the edges of its subsolar hemisphere it seems to me that red dwarf planets would have a combination of plenty of shielding yet sufficient light to allow refuge.
The question of nutrient cycling through geological activity, as brought up by Eniac, is interesting. Though, in Earth’s case, much growth seems rate limited by such cycles, I have never seen an estimate of how much biological productivity would decrease given lower levels of geological activity. From what has been written, I would not be surprised if a world 10% as effective at geochemical cycling as Earth had fully 90% of its biological activity. At these lower levels are there alternatives to plate tectonics? I am very interested if anyone knows any evidence that could prove such speculation wrong.
If you want to know how life might deal with a star producing too much short-wavelength radiation, you don’t need to look further than the equatorial regions of your nearest G-dwarf planet, where “flare” events occur around noon every cloud-free day. The answer seems to be: find some shade. There are relatively few species, even among plants, that expose themselves to Sol’s full glare – but the forest understory is full of ecological niches. Imagine an M-dwarf planet with a good thick forest cover – not to mention the permanent, stationary shadows of mountains and crater walls – and it starts to look a lot more habitable.
Ronald: As Andy points out, simply taking the absolute width of the HZ and assuming it to be proportional to the probability of finding a planet there is clearly wrong, and probably not even close.
Cessation of plate tectonic may stop elements from being recycled, but that works both ways, so it is not at all clear to me whether it would lead to an increase or decrease, or any significant change at all for a given element. Is there good evidence or arguments for any of this? Lack of plate tectonics does not seem to have removed carbon from Mars or Venus, for instance.
For some interesting speculation on the ways in which life on a planet orbiting a flare star might adapt to (avoid or exploit) flares, I recommend Niven’s short story “Flare Time”. I have it in a Niven collection entitled “N-SPACE”. Wikipedia gives this info: “the story is one of several in Medea: Harlan’s World (1985; ISBN 0-932096-36-0) [which] is a collection of science fiction short stories by different authors, all taking place on the same fictional moon. It was an experiment in collaborative science-fictional world-building, featuring contributions by Hal Clement, Frank Herbert, and others. Haven’t seen the collection. Want it!
andy, Eniac: yes, it is reasonable to assume that planetary distribution is not uniform across planetary orbital periods, and planets are packed closer together closer in. However, I expect that to be mainly significant when comparing inner systems to outer systems (as andy is giving as an example: 1-2 AU versus 11-12 AU).
For the inner planetary systems (the HZ regions), as for now (and for want of better data) I expect the differences not to be very great, as is also suggested by Traub, ‘Terrestrial, Habitable-Zone Exoplanet Frequency from Kepler’. See particularly Table 4 and figure 5, which is double-logarithmic, indicating that planet density is indeed somewhat higher toward innermost regions, but not spectacularly so.
Eniac: from what I could find about carbon cycles on earth: sedimentation removes about 150-200 Mt (0.15-0.2 Gt) per year from the atmosphere, biosphere and oceans, which is approx. matched by volcanic eruptions. That is very little in comparison with other C flows, partic. between atmosphere and ocean, atmosphere and land (meaning soils plus vegetation) and with C stores. However, if atmospheric C was not replenished by volcanism, the annual sedimentation would theoretically be high enough to equal all atmospheric C (approx. 800 Gt) in about 4000 – 5000 years.
Mind, I am not saying that all atmospheric C would be removed that quickly (in fact highly unlikely), but just that sedimentation would be a significant remover of C at geological timescales.
The absolute amount of atmospheric C (and other atmospheric elements btw) are indeed very low on Mars. The fact that atmospheric C is so extremely high on Venus is probably because there has been no biosphere nor sedimentation (water) to remove it. Having said this, it would be interesting to observe, or even model, what would happen on Venus with a C-fixing biosphere plus sufficient water, but without regular volcanism (Venus does occasionally have cataclysmic volcanic outflows, though, so-called major global resurfacing events).
@Ronald
Venus’ global resurfacing events are how the crustal stress from underlying convection is relived, I’ll take volcanism and plate tectonics, thank you very much :-)
Venus isn’t quite tidally locked (slow retrograde rotation), it’s magnetic field is a lot weaker than Earths and seems to be solar wind induced, yet Venus still has a massive atmosphere. Mars no longer has an active dynamo, and its atmosphere seems to have eroded. Inferred from the massive signs of water erosion, and Mars’ current atmosphere is too thin to allow liquid water to exist, except perhaps in the very lowest lying regions.
LarryD: Venus does indeed have very irregular and very cataclysmic volcanism (massive outflows), as result of lacking plate tectonics.
Read it again: I never said that Venus is tidally locked. With the rest of what you say I simply agree:
Venus has a very dense atmosphere, but nearly all CO2, as a result of nearly all water having evaporated and photodissociated (which probably also caused its lack of plate tectonics: lack of water as a ‘lubricant’ in the mantle).
Mars is too small to hold on to a significant (O2/N2) atmosphere and as a result of its small mass has cooled down so much that it lacks (nearly?) all geological activity. Planetary mass also sets important limits to long-term habitability, both with regard to atmosphere and geological activity.
See also the recent thread under the post “Habitable, not Earthlike”.
Or… there are very few strategies to avoid lethal effects of X rays and life has another strain to deal with.
@Kalish: Huh? I do not think X-rays have any chance making it through the atmosphere to the surface. What exactly are you thinking of?
X-rays couldn’t make it to the surface, but their energy would be absorbed by the atmosphere and re-emitted as UV, which would reach the surface. Also a powerful x-ray or gamma-ray burst could generate nitrogen oxides that would wipe out the ozone layer.
I believe when X-rays are absorbed and reemitted, the result goes in all directions and only a small part will actually be dangerous UV. You would need some really, really strong X-rays for them to cause lethal effects on the surface. Do you know of any realistic estimates of the luminance needed? Presumably, anything able to cause real trouble would have to have an apparent brightness comparable with the sun, and at interstellar distances that is pretty hard. Ten orders of magnitude type hard. A light year is approximately 10^5 AU, that squared is 10^10, which is how much brighter than the sun an object would have to be at 1 light year distance.
A type 1a supernova is about 5 billion times as bright as the sun. At 1 light year distance, then, it would already be less bright than the sun. At greater distances, its effects on a planet should be quite mild and temporary, compared with other geological events and catastrophes that routinely befall planets, such as climate change or impacts. And, for most of its “life”, the Earth did not have an ozone layer to destroy.
You are not comparing like-for-like here. The supernova’s spectrum contains a greater proportion of high-energy radiation than sunlight does.
Which is less harmful, because it does not penetrate the atmosphere…
The estimates I have read about ‘safe distance’ from a supernova explosion varied from about 5 to 30 ly, probably depending on type of supernova and definition of safe.
According to a couple of studies, the most dangerous type, IA, would have to be closer than 10 parsecs (33 light-years) to affect the Earth.
A real risk at geological time-scales near the galactic center (hey, that’s another recent post), but a very small risk in the (mid-outer) galactic disc.
Which comes first, ozone or life? In most cases I suspect the answer is life since that’s where the free oxygen is likely to come from. This means that life would have arisen and flourished without the protection of an ozone layer. If UV is absolutely antagonistic to early life, something else must have been blocking the UV at that time.
On the other hand the effects where it is absorbed can be quite nasty for the planet, particularly once the photochemistry that it drives is taken into account. The energy has to go somewhere after all… you will also end up getting secondary radiation as a result of the gamma rays interacting with the atmosphere.
Ozone has been detected on Mars and Venus, though not in remotely the same quantities as is present on Earth.
Eniac’s assessments highlight the extreme difficulty in finding a mechanism whereby anything other than advanced land-based life is greatly affected by a supernova even at just one light years distance. Since such life can easily reemerge from the sea then, at worst, we must be looking at planets where there is never a window longer than a quarter billion years between such supernova bursts. Such bad luck must be extremely rare even at the galactic core.
@Andy:
Now you realize that it is the energy that matters, not so much the wavelength. In fact, visual and UV are probably more damaging to life at equal energy than X or gamma rays. Because only they will directly reach the surface. Thus my assessment that anything providing much less energy than the sun will not exterminate life, and that puts the distance at somewhere around 1 ly. For the biggest of supernovas.
@Eniac: wavelength certainly does matter, particularly where photochemistry is concerned, and it is the photochemistry that is one of the major environmental effects of the supernova.
Recall the basic observations that led to the discovery of quantum physics. Einstein got his Nobel Prize for his work on the photoelectric effect after all…
You’ve also got the cosmic ray particles to factor in – these may function as cloud condensation nuclei and trigger a global cooling event.
And what would be the effect of the supernova ejecta? A supernova doesn’t just produce light after all…
@Andy: You are right where photochemistry is concerned. That is a generalization of the ozone layer argument. I do not believe we know whether ozone will increase or decrease as a result of increased x-ray or gamma ray irradiation. I do not believe that either is going to destroy all life on the surface far below, much less below the surface or in the oceans. Nor do I consider it likely that another, more dangerous photochemical hazard lurks. Global cooling events are common, without much sterilization.
Ejecta are another matter, although they, neither, will penetrate the atmosphere. My guess is: Unless ejecta are more energetic than the radiation, they will not do more damage. They would also have to be more energetic than the solar wind. Are they either? I am guessing they will be much more spread out in time than the radiation, which should serve to attenuate them quite a bit.