Habitability Around Nearby Stars

by Paul Gilster on May 10, 2011

My friend Adam Crowl, a polymath if there ever was one, is working hard on Project Icarus and keeping an eye on the exoplanet situation. When you’re working on a starship design, no matter how theoretical, a major issue is the choice of targets, and the study of Kepler planets we looked at yesterday caught Adam’s eye some time ago. We’re not finding as many planets in the habitable zone thus far in the Kepler hunt as we might hope to, given that the ideal would be a habitable world somewhere within reach of near-future technologies of the kind that Icarus represents.

Sure, Kepler’s target stars are much further away in most cases, but the mission is giving us a useful statistical sampling from which we can generalize. Working with the data from Lisa Kaltenegger and Dimitri Sasselov’s paper, Adam thus takes a back-of-the-envelope stab at the galactic population of terrestrial worlds, knowing that Kepler is far from through, as we’re moving into the domain of planets with longer orbital periods. But if you had to go with what we have now, Adam estimates a population of ‘Earths’ somewhere in the range of 150 to 200 million in the Milky Way, assuming 100 billion stars for the galaxy, a figure that may be on the low side.

200 million habitable planets is a huge number even so, but if the total number goes down, so does the chance of finding one of these within, say, 100 light years of our Sun. We’re still shooting in the dark here, but we can only hope that the three ongoing searches of Centauri A and B will yield something interesting, or that we may find other exoplanet candidate stars hosting rocky worlds nearby. On that score, it’s intriguing to see that Gliese 581 d is back in the news, in the form of a new paper by Robin Wordsworth (Institut Pierre Simon Laplace, Paris) and colleagues.

Continuing Saga of a Nearby Red Dwarf

GJ 581 has been a major player in the exoplanet story since the 2007 announcement from the Geneva team of the potential habitability of GJ 581 c. Gliese 581 itself is an M-class red dwarf, smaller and cooler than our Sun, and a variety of studies have suggested that a habitable climate might be possible even on a world that, like GJ 581 c, is tidally locked to a star like this one, presenting the same face to the star throughout its orbit. Subsequent work, though, makes a strong case that GJ 581 c is more like Venus than anything else, while GJ 581 d, further out, may just make it inside the habitable zone, depending on conditions in its atmosphere.

Wordsworth and team are obviously aware of the discussion about the controversial detection of another planet, dubbed GJ 581 g, which appeared to be in the middle of the habitable zone. That one was announced in September of 2010, but its existence remains unconfirmed and in some cases strongly disputed. But GJ 581 d is a known quantity at least in some respects — it’s a ‘super-Earth’ with a minimum mass between 5.6 and 7.1 times that of Earth, and the question is whether a sufficiently dense atmosphere could give the planet a strong enough greenhouse effect for surface habitability, even though it receives 35 percent less energy from its star than does Mars.

The new paper uses global climate model simulations starting with the assumption that GJ 581 d’s climate is dominated by the greenhouse effects of CO2 and H2O, which is the case for the rocky planets with atmospheres in the Solar System. The model is put to work on a planet that may have a permanent night side, a place where volatiles like CO2 and H20 can freeze out on the surface. Even a relatively dense CO2 atmosphere can collapse under these conditions, so the simulations are designed to evaluate whether liquid water at the surface is possible.

The simulations used a global climate model developed specifically for exoplanet studies, assuming CO2 as the main atmospheric gas and drawing on these methods (from the paper):

To assess the influence of water on the climate independently, we considered two classes of initial condition: a rocky planet with no water, and an ocean planet, where the surface is treated as an infinite water source. CO2 was taken as the primary constituent of the atmosphere and H2O was allowed to vary freely, with surface ice / liquid and cloud formation (including radiative effects) taken into account for either gas when necessary. Restricting the composition of the atmosphere to two species in this way allows us to determine conservative conditions for habitability, as it neglects the warming due to other greenhouse gases like CH4 or buffer gases like N2 or Ar…

Greenhouse Warming and its Consequences

The researchers then performed simulations with 5, 10, 20 and 30 bar atmospheric pressure and varied the orbital resonances for both rocky and ocean planets. CO2 turns out to have a strong warming effect in these simulations, which showed that for pressures below about 10 bar, the atmosphere was unstable and began to condense on the poles and dark side of the planet. But denser atmospheres allowed enough heat transport and greenhouse warming to bring temperatures at the surface above the melting point of water, even as CO2 ice clouds frequently formed in the middle atmosphere, similar to the ice clouds we see on Mars.

So is it likely that CO2 partial pressures over 10 bar occur on GJ 581 d? We know nothing about the geophysics of this world, but if a mechanism exists there like the carbonate-silicate cycle we see on Earth, the atmospheric CO2 should stabilize at levels that allow oceans to exist. And of course there are other scenarios, as the authors point out: The planet could have a thick envelope of hydrogen and helium, like the atmospheres of Uranus and Neptune, or it could have no atmosphere at all — remember, GJ 581 is a red dwarf, a star whose frequent flare activity in its youth could have removed the early atmosphere. At this point, we just don’t know.

But the beauty of GJ 581 is that at 20 light years, it’s relatively near to the Earth, unlike most of the stars Kepler is now studying. That means that in the future, we’ll be able to discover which atmospheric scenario applies here by direct spectroscopic observations. The paper goes through the indicators that would help us distinguish between these scenarios, and it’s clear that this is a call we’ll one day be able to make as the technology comes online.

Thus the title of the paper by Wordsworth et al., “Gliese 581d is the first discovered terrestrial-mass exoplanet in the habitable zone,” accepted at Astrophysical Journal Letters (preprint). The caveat is obvious: The fact that a planet is in the habitable zone in some scenarios does not mean that it is actually habitable, but the new paper does tell us that this is a world that merits intense scrutiny, because oceans here are not beyond the realm of plausibility. If that’s the case, Project Icarus may be able to find at least one astrobiologically interesting planet within range of its vehicle after all.


djlactin May 10, 2011 at 9:58

I understand why we should assume tidal lock in such systems, but is it an absolute? Recall that Mercury was assumed to be tidally locked until radar observations in 1965 revealed that it actually rotates 3 times per 2 revolutions. Moreover, the 3:2 spin-orbit resonance appears to be due to interactions with the gravitational influences of other planets, particularly Jupiter (http://en.wikipedia.org/wiki/Mercury_planet), so the presence of other (large) planets in the Gliese 581 system might conceivably have a similar (or greater) effect. I think that somebody (not I, mere biologist) should examine this possibility: specifically, how might the atmospheric dynamics of a planet that is not tidally-locked differ from one that is? How would such difference affect opinions of habitability?

Paul Gilster May 10, 2011 at 10:31

djlactin, the Wordsworth paper does not assume tidal lock for GJ 581 d but investigates it as one possibility. From the paper:

“As it is most likely either in a pseudo-synchronous state with a rotation period that is a function of the eccentricity, or in spin-orbit resonance like Mercury in our Solar System (Leconte et al. 2010; Heller et al. 2011), GJ581d should have extremely low insolation at its poles and possibly a permanent night side. Regions of low or zero insolation on a planet can act as cold traps where volatiles such as H2 O and CO2 freeze out on the surface.”

Also this:

“…at higher pressures we found stable, hot climates with little global temperature variations even in the tidally locked cases, regardless of the choice of microphysical parameters.”

Total tidal lock does present problems for this planet depending on atmospheric density:

“We also investigated the possibility that an ice-covered, tidally locked GJ581d with permanent day and night sides could be locally habitable on the day side due to partial melting. However in this scenario, the day side only stayed warm when the atmosphere was thin and hence inefficient at transporting heat. As a result, the planet’s dark side became cold enough for the collapse of even an N2 atmosphere. An ice planet with only a thin sublimation-driven H2 O atmosphere could have dayside temperatures above 273 K and continual transport of H2 O to the dark side, but the low atmospheric pressure would preclude liquid water except in extremely limited sub-surface regions. Clearly, dense, stable atmospheres offer better prospects for habitability.”

Johnhunt May 10, 2011 at 10:51

Habitability is interesting from the exobiology standpoint in that a habitable exoplanet might contain life. But for a several reasons, I think that it is not particularly useful from an interstellar mission standpoint.

First, if a habitable exoplanet does have bacterial life, then we’ll probably know that before we launch through the optical analysis of its atmosphere. I don’t know that a flyby mission would add much more information.

Secondly, if habitable exoplanets are so common that we find one within 20 light-years of Earth, then there would likely be millions of Earths with bacterial life. That many life-bearing exoplanets and apparently none went on to intelligent life with a desire to contact us? Possible, but I don’t think likely.

Thirdly, it is likely that we’d have to fly further to reach a habitable planet than an otherwise interesting but uninhabitable planet. But, although such a planet would not be habitable, it would be “inhabitable” in that it would be a perfectly good location to use technology to paraterraform and to plant the seeds of life if we were able to decelerate.

However, after having said all this, if the first true interstellar missions were to launch thousands of nanocraft for not much more than the price of the first launch (i.e. huge infrastructure cost and low relaunch cost) and if those nanocraft were able to travel decent percentages of the speed of light (e.g. 0.5 c), then I could imagine having a habitable planet on the list of early targets.

kurt9 May 10, 2011 at 11:55

At 5 to 7 Earth masses, GJ 581 d is probably a Neptune like planet rather than a rocky world.

philw1776 May 10, 2011 at 11:57

Assuming a resonant spin like Mercury, what would such a rotating planet be like? Close to an M star for habitability, the tides would be HUGE. Oceans and even lakes would roll over the planet. Also, the tidal stresses inside the planet should create an Io like scenario with tidal stress vulcanism galore. Better that the planet be peacefully tidally locked with Hell surrounding the sub-stellar point but outward a ring of possible temperate habitability. The winds conducting heat from the lit side to the dark side should be something however.

Paul Gilster May 10, 2011 at 12:56

kurt9 writes:

At 5 to 7 Earth masses, GJ 581 d is probably a Neptune like planet rather than a rocky world.

I think this is exactly right.

Rob Henry May 10, 2011 at 17:53

If Gliese 581d is a rocky planet that is much more massive than Earth, wouldn’t its levels of volcanism be much higher than ours? How high would they have to be to create (or return from the frozen night side) a sufficient quantity of carbon dioxide to keeps things going.

Adam May 10, 2011 at 20:46

Hi All
If Gl 581d is a sub-Neptune, with some H/He residual atmosphere after the early EUV baking, then it might be an Ocean Planet, if it has cooled sufficiently for the water to condense. According to Raymond Pierrehumbert & Eric Gaidos such planets are to be expected near the far edge of the Hab-zone and beyond, just like Gl581d…
Hydrogen Greenhouse Planets Beyond the Habitable Zone
…andy has linked to it previously, but I thought it worth repeating in this context. Of course such planets – with either H2 or CO2 greenhouse atmospheres on a huge scale – aren’t “Earth-like”. Those will probably be restricted to a very narrow annulus around stars. I don’t think anyone has much improved on Mike Hart’s results of 0.95-1.05 AU in the +30 years since he did those models. Earth-like, with low CO2/H2/CH4/CO etc, aren’t so easy to form and maintain. I suspect, though, the task might be somewhat easier with differing levels of N2, land/sea and moderate CO2 levels. The Earth-like zone might stretch 0.8-1.2 AU IMO, though we’re talking from Arrakis to Hoth style planets.

Michael Simmons May 10, 2011 at 23:09

I’ve read the Rare Earth book (available on Amazon as a kindle book) but its out of date and doesn’t have a enough detail on basic planetary parameters.

Is their any good article/book that details the various parameters that make a planet habitable.

Can a tidally locked planet have a earth strength magnetic field?

My understanding is that without a magnetic field, hydrogen in the upper atmosphere will leak into space thorough the bombardment of solar wind.
That hydrogen comes from the splitting of water vapour in the upper atmosphere and hence water over billions of years is lost.

How large does an planet (without a magnetic field) need to be such that hydrogen atoms hit by solar wind particles do not exceed escape velocity?

henk May 11, 2011 at 5:52

Do not worry. Alpha centauri will likely have a habitable planet

Tobias Holbrook May 11, 2011 at 7:40

If it is an ocean planet, how will this affect the feedback cycles, given the thick layer of high pressure ice that would cover the rocky “crust”? A 200km thick layer of ice would seem able to impede the cycling of CO2 through the carbonate-silicate cycle – with no dry land, would weathering even occur at all? Such a planet might not even have plate tectonics…

Ocean planets seem to be ruled out as naturally habitable, for a variety of reasons…

Scott G May 11, 2011 at 8:40

Using Adam’s higher-end estimate of 200 million “Earths” in the Milky Way, I was curious to see what the average distance between them would be, so this is my attempt at a back-of-the-envelope calculation to find out:

Volume_galaxy = pi * radius^2 * thickness = 7.854 trillion cubic light years
(where: radius = 50,000 light years, thickness = 1,000 light years)

Volume_per_planet = Volume_galaxy / (200 million planets) = 39,270 cubic light years

Distance_per_planet = Volume_per_planet^(1/3) = 33.99 light years

So that’s about 40 light years (on average) between “Earths” in the Milky Way. But remember that the galaxy is not a homogeneous disk; there are obvious areas with a high and low densities of stars, so this really doesn’t say much… but, nevertheless, it’s fun to think about.

Scott G May 11, 2011 at 9:39

**Small Correction — in the last paragraph of my previous calculation, I meant to say the average distance between Earth-like planets is 34 light years (not 40).

But, to refine the calculation one step further, you might want to take into account the fact that most stars in the galaxy reside inside its spiral arms. Thus, you might assume the galaxy’s volume is, say, 60% of its nominal value (ignoring the empty regions of space between the arms), which would give an average distance between other “Earths” of 28.66 light years. This value is probably more representative of the true average.

tesh May 11, 2011 at 10:15

henk May 11, 2011 at 5:52
Do not worry. Alpha centauri will likely have a habitable planet

Is this from access to “insider information” of ongoing searches??????

Ronald May 11, 2011 at 11:07

Fascinating topic, my favourite! And good work, Adam!

Ref. Adam: Kasting and others estimate the HZ (defined by liquid water) in our solar system from 0.95 to anywhere between 1.2 and 1.5 AU. As I stated in a previous thread, the outer edge seems to be fuzzy and possibly far out, but there is general concensus with regard to the inner edge, at 0.95 AU.

The mentioned estimate of 150 – 200 million terrestrial planets in the HZ is remarkably on the same order as some previous estimates (and, I may add, a guesstimate by my self) ranging from 50 – 200 million, see for instance the post plus comments: http://www.centauri-dreams.org/?p=11625&cpage=1#comments

I have a few questions to Adam:
– In your estimate, did you take the observational bias toward short orbital periods into account, in other words did you correct for the limited time that Kepler has been operational, or may we expect this number to go up with longer observational time (and orbital periods)?
– What is your definition of a terrestrial planet, particularly with regard to mass?
– How wide did you define the HZ?
– Did you consider only solar type stars? Or what spectral types?
– Did you consider only the galactic disc, or even a galactic HZ, or the whole MW galaxy?

andy May 11, 2011 at 15:10

Adam: I see you noticed that hydrogen-rich habitable planets paper as well. Certainly makes for interesting reading, though I am sceptical about whether there are going to be high enough chemical concentrations in the vast oceans of such worlds to allow for life to arise there. For an earlier consideration of such planets, see this paper about the possibility of liquid water oceans in ice giant planets. Turns out Neptune is too hot and too dry for liquid water at the present time.

Of course, if we define liquid water as the sole criterion to determine habitability, the habitable zone effectively has only an inner boundary: sufficiently massive planets containing sufficient water will end up with liquid water (e.g. as an internal ocean) for some fraction of their lifetime.

Rob Henry May 12, 2011 at 7:10

Andy, are your really sceptical that oceans on super Earths would be too chemically poor for abiogenesis in their primordial states, with comets and meteorites still reigning down? I suspect that your actual concern is that potential biogeochemical cycles would be insufficient on these worlds to support higher life.

Ronald May 12, 2011 at 8:06

Scott G: I don’t think it is correct to say that the space between the spiral arms are largely empty, on the contrary. The spiral arms are so conspicuous because of the presence of relatively large numbers of very bright and large stars (O, B, A, giants). But there are also huge numbers of stars in the regions bewteen the arms and even in the halo.
To refine your calculation I think it would be better to consider the galactic (thin, intermediate) disc, perhaps even the Galactic Habitable Zone within it, roughly extending from 6 – 12 kpc from the galactic centre. And then only the solar type stars (for instance F9-K2) within that.
That limits your number of habitable planets even further (I and others came to very roughly 50 million orso) but at the same time concentrates them more.
But going by your rough estimate for now, I actually find 30 – 40 ly between nearest two truly habitable planets (not just terraformable, terrestrial) rather encouraging. This would imply that we will be able to image and spectroanalyze at least one or a few even within the foreseeable future.

andy: “if we define liquid water as the sole criterion to determine habitability, the habitable zone effectively has only an inner boundary”.
Yes! This is what I also meant in my previous comment (and in other posts on this topic). The outer boundary, also according to various authors pratic. Kasting, is fussy and possible far, since it largely depends on planetary characteristics. The inner boundary, however, seems sharp and rather well-delimited, because it is mainly determined by solar UV driven photo-dissociation of H2O and resulting runaway greenhouse. The real problem on Venus is not excess CO2, but rather lack of water. Strikingly, nearly all authors (except Dole 1964) put the inner boundary of our HZ at 0.95 AU.

Scott G May 12, 2011 at 10:54

Ronald — In light of your comment, I decided to try this again using one other approach; here it goes…

Still assuming 200 million other “Earths” in our galaxy of 400 billion stars, we have 1 Earth per 2000 stars. From Wikipedia, the local density of stars in our region of space is 0.00352 stars per cubic light year (the number comes from a count of 64 stars within ~16 light years).

So, the distance from us in which we’d expect to see another terrestrial planet within the habitable zone is one that gives a sample of 4000 stars — because the occurrence of “Earths” is 1 per 2000 stars and we’re already including ourselves in the sample. If we call this distance D, then we have:

D = [ (4000 stars) / ((0.00352 stars per cubic LY) * (4/3) * pi) ]^(1/3)
D = 64.74 light years

(The formula is from the volume of a sphere.)

andy May 12, 2011 at 13:01

Rob Henry: no, my actual concern is what I actually stated: that chemically-speaking these worlds are poor prospects for abiogenesis, particularly if when considering water-soluble chemistry. Ocean planets appear to lack many of the features that would tend to lead to higher concentrations of chemicals: you don’t have shorelines where tides can wash in and out, and the silicate core where you would otherwise get hydrothermal systems is at the bottom of the ice mantle. Meteorites don’t seem to me to be a particularly viable way to get around this: the ones that get broken up into small pieces would end up getting spread through a vast volume of water, the big ones would presumably end up sitting on the icy ocean floor doing not very much (rocks don’t tend to dissolve all that much, and an inert pebble is rather different to a hydrothermal system).

Rob Henry May 12, 2011 at 18:56

Andy: I find the details of your objections to abiogenesis in super Earths fascinating because each contention seems both plausible and possible and yet open to caveats.

Atmospheric conditions on these worlds should be more amenable to the build-up of a prebiotic soup than the early Earth (if a complex dilute soup forming in an ocean is still believed possible). As for the concentration of these materials, the formation of ice has been equally mooted as a potential mechanism, and perhaps my ignorance of the nature of high pressure ices allows me to believe that their continual formation in more turbulent parts of the host planets depths could make up for the lack of shoreline here (if indeed our primordial planet ever had a shoreline either).

As for life originating in hydrothermals, the minerals from which the mantels of these planets forms should start with an enormous inventory of water, even if most of it is chemically bound. I find it hard to believe that all this can be extracted as the mantel forms, or outflowed from it so rapidly as to preclude the potential for abiogenesis.

Even if this desiccation of the mantel does occur extremely rapidly these planets must somehow release the radiogenic heat from their interiors. Ongoing complex life may or may not require ‘plate tectonics’ but surely the best possible solution for the possibility of one-off processes such as abiogensis would be like that on Venus where huge amounts of geochemical energy are released all at once. During such global resurfacing episodes it is hard to conceive of a way deep ices could chemically insulate the rest of the planet from rich infusions.

Actually, after all that, I am not really sure that life on Earth began on Earth at all, and this would make the issue moot anyway.

Ronald May 13, 2011 at 11:45

Scott, again interesting work and the results are in the same realm.

I have just done my own back-of-the-envelope calculation, taking just solartype stars (defined as F9-K2) in the Galactic Habitable Zone, the GHZ defined, according to Lineweaver and Reid, as a the part of the Galactic Disc between 6 and 10 kpc from the galactic centre, thus assuming a width of the GHZ of 4 kpc (13,000 ly) and a thickness of 3000 ly, giving a torus (belt) with a volume of 6.5 * 10^12 cubic ly.
I loosely follow RECONS and NStars in an average stellar density of approx. 0.003 per cubic ly, giving a total of about 20 * 10^9 stars in the GHZ.

I then assumed a modest 50 million habitable planets around sunlike stars within this GHZ.

That is 1 in 400, giving (using Scott’s formula) an average distance D between two habitable planets in the GHZ of 32 ly.

Again, in the same range.

spaceman May 13, 2011 at 21:44


Just because an earth-sized planet is within its stars Habitable Zone, does this necessarily mean that it will be habitable? Is it not possible that there may be plenty of dry rocks sitting smack in the middle of the HZ of many of these millions of stars? I am not trying to be negative…just wondering.

Tobias Holbrook May 15, 2011 at 9:31

Re. ocean planets – would there possibly be an oxygen rich atmosphere present, even without life? I’m thinking that radiation would break apart the water molecules, and the hydrogen would escape to space… this could possibly supply a suitable oxidiser to any organism that can concentrate a reducing agent in itself, and so provide an easy source of energy. Even if it doesn’t, it would possibly nudge such planets into the Biocompatible bracket…

Rob Henry May 15, 2011 at 17:43

Tobias Holbrook, oxygen is such a poison to the process of life that the consensus is that a planet that contains it would not be conducive to abiogenesis in zones that were not protected from it. Ozone might make dry land based life possible, but there would be none of this on ocean worlds anyway.

As for the inducement to higher life, I note that reducing sugar or formaldehyde with hydrogen releases only a third the energy of its oxidation with oxygen, but hydrogen diffuses four times quicker than oxygen. Since the rate of delivery of respiratory gas is seen as the limiting step in the release of metabolic energy in higher life, I put it to you that equimolar quantities of either oxygen or hydrogen in water would be equally good at providing for complex life. In the ocean this should occur when the partial pressure of hydrogen is two to three times higher than oxygen, an easily met requirement given the typical hypothetical atmosphere for a super Earth.

Sometimes I feel that if we are not the only sentience in the galaxy, all the rest will be hydrogen breathers who have long believed that higher life can not exist on planets with oxygen in their atmospheres.

ljk May 18, 2011 at 3:22

Case builds for habitable alien planet

Zina Deretsky, National Science Foundation

The orbits of planets in the Gliese 581 system are compared to those of our own solar system. The Gliese 581 star has about 30 percent the mass of our sun, and the outermost planet is closer to its star than we are to the sun. Gliese 581d might be able to sustain liquid water on its surface.

By John Roach

The case is building about the habitability of a planet orbiting a red dwarf star about 20-light years away from Earth, according to a new climate modeling study.

The planet, Gliese 581d, is one of a handful of planets orbiting the star Gliese 581. When it was discovered in 2007, astronomers thought it was likely too cold for liquid water, and thus life.

The new study, accepted for publication in the Astrophysical Journal Letters, suggests high concentrations of carbon dioxide in its atmosphere could keep things warm enough for liquid water to be sustained at the surface.

The finding falls on the heels of a similar atmospheric modeling studies published that have reached a similar conclusion.

Full article here:


Greg Stephens May 25, 2011 at 11:18

What searches are you referring to? I hear mumbles about organizations looking here but there never seems to be any results. Is there a reason?

“We’re still shooting in the dark here, but we can only hope that the three ongoing searches of Centauri A and B will yield something interesting,”

Paul Gilster May 25, 2011 at 13:38

Greg, check this story for the details on the three Alpha Centauri searches, all now active:


Greg Stephens May 25, 2011 at 15:39

Thanks Paul..Love the blog. So no results for in three years. It seems like the nearest system would be the highest priority. It seems to be the only one even theoretically starting a mission towards in the next 50 years or so.

Comments on this entry are closed.

{ 3 trackbacks }