≡ Menu

Habitable Worlds Around Tau Ceti?

Yesterday’s look at NExSS (the Nexus for Exoplanet System Science), NASA’s new ‘virtual institute,’ focused on the multidisciplinary nature of the effort. The work I’m looking at today, an analysis of the planets around Tau Ceti performed at Arizona State University, only emphasizes the same point. To get a read on whether two planets that are possibly in Tau Ceti’s habitable zone are likely to be terrestrial worlds like Earth, the ASU team brought the tools of Earth science into play, in particular the work of Sang-heon Shim.

Shim is a mineral physicist who worked with astrophysicists Michael Pagano, Patrick Young and Amanda Truitt in the Tau Ceti analysis. His perspective was vital because early work had already suggested that Tau Ceti has an unusual balance between the rock-forming minerals magnesium and silicon. In fact, the ratio of magnesium to silicon here is 1.78, about 70% more than we find in the Sun. That casts long-standing views of Tau Ceti as Sol’s twin into doubt, and raises questions about the nature of the planets that formed around it.


There is evidence for five of these, with two — Tau Ceti e and f — thought to be in the habitable zone. That’s an attractive possibility, for Tau Ceti is nearby at 12 light years, a solitary G-class star like the Sun, and a relatively stable one at that. No wonder it figures prominently in science fiction, its very proximity made a significant plot issue by Larry Niven in his 1968 novel A Gift from Earth, which depicts an isolated Tau Ceti colony that can still receive the occasional cargo from Earth. Isaac Asimov made a Tau Ceti planet the home of the first human extrasolar settlement in The Caves of Steel (1954).

Image: The Sun is at the left in this comparison with Tau Ceti. Credit: R.J. Hall via Wikimedia Commons.

In fact, I can think of few stars that have received so much attention from writers. Might some of the planets there really be habitable? The two planets we are looking at are ‘super-Earths,’ with masses of 4.29±2.00 and 6.67±3.50 times that of Earth respectively. The new work makes the prospect of Earth-like conditions unlikely. In fact, Shim’s mineralogical study indicates that the high magnesium/silicon ratio of the parent star could produce planets unlike any we’re familiar with, as the scientist explains:

“With such a high magnesium and silicon ratio it is possible that the mineralogical make-up of planets around Tau Ceti could be significantly different from that of Earth. Tau Ceti’s planets could very well be dominated by the mineral olivine at shallow parts of the mantle and have lower mantles dominated by ferropericlase.”

Ferropericlase is a magnesium/iron oxide that is thought to be a major constituent of the Earth’s lower mantle, along with silicate perovskite, which is a magnesium/iron silicate. Because ferropericlase is viscous, an abundance of it in the mantle would make mantle rock flow more readily, possibly affecting plate tectonics and volcanism at the planetary surface. The resulting world would pose challenges for the development of life, and certainly for its detection. “Faster geochemical cycling,” the paper notes, “could impede the buildup of biologically produced non-equilibrium chemical species in the planet’s atmosphere.”

The paper describes what it calls a detectability index, or DI, that gauges the ability of a planet to house life and to maintain biosignatures of the kind we hope to detect with new space telescope missions. Tau Ceti planets in the habitable zone might, in other words, be habitable, but unlikely to produce detectable life signs in their atmospheres. Life would not necessarily be absent, but detecting it would require a thorough study of planetary evolution.

Another issue is the length of time a planet spends in the habitable zone. Tau Ceti e’s position is deeply problematic. The authors believe the world is reaching the end of its habitable lifetime and is at best on the extreme inner edge of the habitable zone. Tau Ceti f, meanwhile, appears to be near the outer edge of the habitable zone, but evidently moved into it within the past 1.5 billion years, and probably in much less time than this.

Assume even a billion years in the habitable zone and bear in mind that Earth’s biosphere took roughly two billion year to produce biosignatures that would be theoretically detectable. The DI for this world — our ability to find life if it does exist on Tau Ceti f — would be low indeed. A long habitable lifetime may be in this planet’s future, but that doesn’t help us now:

Even in the most pessimistic case, the planet will have about 7 Gy of habitable lifetime until the end of the main sequence, plus additional time while the star traverses the subgiant branch. From a detectability standpoint, however, f is a poor candidate. At best, the planet has been in the HZ for < <1Gy under these assumptions.

So we have in Tau Ceti f a world where life could exist but would be, at least according to what we know of life here on Earth, probably in an early state and unlikely to be detectable. There is a significant lesson for us in this conclusion, as the paper goes on to note:

The rate of change of L [luminosity] and Teff [stellar effective temperature] as a function of time means that cases similar to f where a planet enters the HZ in the latter part of the star’s life are more common than planets that have been in the HZ since early times. This serves as a reminder that the present “habitability” of a planet does not necessarily indicate that it is a good candidate for detecting biosignatures. The temporal evolution of the system must be taken into account.

So much, then, for the notion that if we detect a planet in the habitable zone, we are safe in assuming life has had time enough to emerge there. Here we have one case of basic mineralogy casting doubt over whether surface conditions are habitable or whether it could be detected by our instruments if present, and another where the movement of a planet into the habitable zone means that biology may flourish there but only in our distant future.

The paper is Pagano et al., “The Chemical Composition of τ Ceti and Possible Effects on Terrestrial Planets,” Astrophysical Journal Vol. 803, No. 2 (2015), Art. 90 (abstract / preprint).


Comments on this entry are closed.

  • Alex Tolley April 24, 2015, 13:06

    We should be a bit careful of using Earth as the gold standard for life and assume that different conditions for life’s emergence will be poorer with different conditions. Reality may turn out to be very different.

    In our own solar system, we are going to know within the next century whether the lack of bio-signatures except for Earth indicates an absence, or hidden, life. I personally think life in the solar system is confined to Earth, and all other worlds are currently sterile. I would be delighted to be proved wrong, however.

    Exoplanet bio-signatures will likely be very rare, but represent the low hanging fruit for discovery. I hope that they stimulate the drive to send fast probes (like the DragonFly Project) to search/characterize for life on the nearer exoplanets.

  • andy April 24, 2015, 13:30

    The masses given for e and f are the minimum masses from radial velocity measurements. If the Tau Ceti planets are aligned with the debris disc which is inclined at 35 degrees from the plane of the sky, the masses are a factor of ~1.7 times greater. The resultant true masses are sufficiently high that it seems more likely that they are mini-Neptunes rather than solid planets.

    The other question is whether these planets exist: according to the discovery paper the HARPS dataset only contained evidence for the innermost 3 planets b, c and d, while the 5-planet solution arises from the combination of the HARPS data with lower-precision data from HIRES and AAPS.

    Nevertheless, the possibilities for planetary composition in the system are definitely worth bearing in mind if an Earth-mass planet is discovered in the system’s habitable zone (e.g. in the five-planet solution there seems to be a stable zone between the orbits of e and f that could perhaps host such a planet).

  • Ashley Baldwin April 24, 2015, 13:39

    The increased gravity and resultant far greater mantle density wouldnt increase tectonics, vulcanism or geochemical cycling, though they might prologue it, which is good especially around a G8 star with a long main sequence lifetime . I’m sure Dimitar Sasselov might have a different view if these planets were found to be big terrestrials. Kite et al 2008 describe the extended “volcanism” of large terrestrial planets. What’s a billion years when your star still has 7 left on the main sequence ?

  • Ashley Baldwin April 24, 2015, 13:42

    Tau Ceti, Please fight on ! Never mind your large planet bombarding asteroid/Kuiper belt , you can deliver !

  • Michael April 24, 2015, 14:45

    @Ashley Baldwin April 24, 2015 at 13:39

    ‘The increased gravity and resultant far greater mantle density wouldnt increase tectonics, vulcanism or geochemical cycling, though they might prologue it…’

    I would tend to agree, magnesium oxide has a higher melting point, higher density and significantly higher thermal conductivity than silicon dioxide and would tend to suppress convection in the upper/lower mantle. How this affects the formation of a magnetic field is unknown as well though as our Earth is thought to generate it’s magnetic field from solidification of the outer core. Heat that can’t get out would tend to form a more liquid core than a solid one I would think.

  • Andrew LePage April 24, 2015, 15:58

    I wrote in Centauri Dreams three months about Tau Ceti in my review of habitable planet candidates.


    Basically, Tau Ceti f doesn’t even make the cut on the PHL’s “Habitable Exoplanet Catalog” (which is usually far too liberal with its definitions of “habitable”). In the case of Taus Ceti e, it is far more likely to be a hot mini-Neptune or maybe, optimistically, a cool super-Venus. Of course, this assumes that these planets even exist in the first place. They remain unconfirmed and there is reason to believe that the RV signatures of the putative planets are the result of surface activity. A similar thing happened last year with the disappearing habitable planets of GJ 581 and 667C.



  • Ashley Baldwin April 24, 2015, 17:59

    With the RV range on ‘f’ it could easily be anything from a 3M terrestrial to a 10M ice giant . The dreaded msinp ! It’s this combined with uncertainty over the RV discovery in the first place that confine it to the ” need more information category”.
    David Spergel , not content with getting a coronagraph on board is hoping to combine the completed Gaia dataset with WFIRST observations to produce in essence a watered down TPF-I /NEAT. It would have a 15 year baseline. From speaking to Mike Perryman of Hipparchus /Gaia fame and a close colleague of Spergel at Princeton , this is just possible though difficult. It will provide accurate astrometric data on planets down to Earth mass , with less than an 18 year period , out to 10 parsecs. Circa 119 suitable stars apparently . Detailed presentation available from January AAS and DS himself 2013. Meantime the 2016/17 VLT ESPRESSO spectrograph has the sensitivity though whether it can overcome Tau Ceti’s stellar “noise” is unclear.

  • Jim Early April 25, 2015, 11:47

    With all that extra magnesium to react with the available oxygen, there may be less oxygen available to react with the iron. This could give a larger iron core and a mantle depleted in iron oxides.
    I believe the masses are high enough that the planet should capture significant hydrogen even if it doesn’t form in the ice zones. It will be hard for any life to generate enough free oxygen to oxidize all the hydrogen and elemental surface iron, a real challenge for any advanced life.

  • Hop David April 25, 2015, 13:08

    Colonizing habitable planets in other star systems is a science fiction staple. These stories are examples of planetary chauvinism, in my opinion.

    A civilization capable of making a generation star ship is also capable of exploiting and settling smaller bodies like the Main Belt asteroids or Kuiper Belt snowballs.

    We can only exploit the top few kilometers of a planetary surface. Heat and pressure bar us from burrowing deeper. In contrast the entire *volume* of a small body is accessible. When it comes to accessible real estate and resources, asteroids, KBOs and Oort objects have a lot more to offer than Sol’s planets and large moons.

    By the time humankind reaches a level where we can even think of traveling to another star system, I suspect planetary chauvinism will have become ancient history.

  • xcalibur April 25, 2015, 15:01

    I’m looking forward to increased capabilities of observing and measuring exoplanets. We’ve made lots of progress, but there’s still plenty of ambiguity. Getting definite results, and finding a habitable exo-earth would go a long way in increasing interest in space.

  • Larry Kennedy April 26, 2015, 18:33

    Hop David: I’m with you on the small bodies. It seems to me that is really the only way to a true spacefaring civilization. Even if someone did a Mars One and succeeded in setting up self sustaining population, I’m not sure I would truly consider that a spacefaring civilization.

  • John April 27, 2015, 3:27

    Small body colonies make a lot of sense to me too, although the issues of human adaptation to micro gravity (be they solved technologically or biologically) will need to be adressed.

    I wonder if our solar system might actually be sub-optimal for a space faring race – perhaps a system that never formed major planets and instead retained remnants of its protoplanetary disk as massive asteroid belts and comet clouds might be a more attractive home, once we’ve (if we!) become truly at home in space.

  • Zanstel April 27, 2015, 10:23

    In my opinion, terraformed planets are very important as the role of “true biospheres”. Ring worlds are probably beyond truly engineering. Big colonies will not be enough to have a biosphere like a planet. As humen cities will be great, but no as biospheres.

  • Terraformer April 27, 2015, 10:26

    Given the prospect of a world moving out of the habitable zone whilst a world has recently moved into it… how would panspermia affect things? If life originated on the inner habitable planet and was carried through space to colonise the outer one (similar to what has been suggested for Mars and Terra), then life could get a foothold on f much quicker, which would affect detectability..?

  • RobFlores April 27, 2015, 11:58

    The Culture of small body colonizers will be substantially different, because the governance will look very different than Western Democracies, consequently the rate of their technological advancement will be different.

    Assuming a colonizing attempt was launched at a relatively near-by solar system that has a marginally suitable. Somewhat Like Mars but with enough atmosphere so that water is stable but you cannot be outside unprotected for more than 20 mins (mostly due to solar radiation slightly higher CO2 levels) . Other colonists wish to colonize dwarf planets/large asteroids.

    In general the more technological support is needed for basic survival the more likely, a colonizing site will lean towards oligarchy/police state. This a consequence of maintaining the technology to enable human habitation, and to keep the critical systems from damage through internal disturbances (which would kill everyone). Things will be heavily monitored/controlled. Also generation after generation of technologists will have to be assured by directive most likely. So not too many letters and arts majors there. That is contrast to the Planet bound settlers who only need very basic technology to survive, and are more free range, in poultry parlance.
    I think the technological race between the Soviets and US in the post war is somewhat illustrative to the effects of differing governance on technology advances vis-à-vis Planets VS Small body colonists.

  • Michael April 27, 2015, 14:32

    @RobFlores April 27, 2015 at 11:58

    ‘Somewhat Like Mars but with enough atmosphere so that water is stable but you cannot be outside unprotected for more than 20 mins (mostly due to solar radiation slightly higher CO2 levels) . Other colonists wish to colonize dwarf planets/large asteroids. ‘

    Venus type worlds could also be colonised via cloud cities, CO2 makes a very nice lifting gas. Energy and surface area wise Venus has multiples of liveable space than Earth.


  • Gregory Benford April 27, 2015, 15:10

    Stan Robinson’s fine new novel has a starship venturing to Tau Ceti, with many ecological problems afoot: AURORA.

  • Larry Kennedy April 27, 2015, 15:14

    @ John If we ever move into the galaxy I definitely think that hot young stars with lots of small bodies will be the gold standard. Of course assuming no FTL our descendants probably won’t wait for one to come around. Assumes of course they plan on moving on in a few million years.

    @Zanstel I think your biosphere point is a good one. Of course maybe thats a good reason to make planets biological reserves. I do think spreading the biosphere is a big reason for spreading out. Whether the AIs will think so I don’t know. Also biospheres may come in many sizes, especially if you include advanced human technology as part of them.

    @ RobFlores You make a good argument. Of course I’m not sure it applies any more to colonies than it would to most planets, at least not before a great deal of terraforming is completed. On the other hand I wouldn’t lay odds that the necessary tech will be anymore confining to our descendants than oil refineries and air conditioner factories are to us.

  • Rob Henry April 27, 2015, 16:43

    RobFlores, a world advanced enough for interstellar travel is one where joe citizen can enrich uranium and make smallpox in his garage. It would have to be a police state everywhere except on colonies that had such low populations that everyone knew and respected everyone (and thus exactly the opposite to your claim). All others would be very suspicious of these micro-colonies, but they could exist if all their interactions with others were very heavily monitored.

  • Rob Flores April 27, 2015, 23:18

    Rob Henry, are you so sure about that?
    A small population trusting each other?

    I draw you to a contra-factual, ALASKA (albeit an imperfect example)
    There Is a lot of literature and pop culture pointing to the fact that in the
    recent past Alaskans where human habitation was sparse, had a very cloistered mindset. Not a spirit of trusting your fellow man or even caring about them.
    Maybe a better comparison is a Polynesian colonial expansion, but note
    they were of the same race and creed, will space colonies be multi-cultural?

  • Larry Kennedy April 28, 2015, 0:02

    @Rob Henry MAD on an individual basis. What a terrible thought.

  • Wojciech J April 28, 2015, 12:21

    We all should be wary of making assumptions based on theoretical outcomes. Until the said planets are observed directly and in full, we might find out that the theory doesn’t match reality.Time will tell.
    I am with Hop David re colonization and spaceships. As I stated before, once you have technology allowing you to travel to other stars to colonize their planets, you no longer have to. You already have technology allowing you to create sustainable artificial environments with far less cost intensive measures and far less time intensive measures.
    If a planet has a bioshpere it would be for us too much time consuming and wasteful to adopt either it or our species for colonization. It is much better as living natural laboratory to study.
    Exception from this would be natural “habitable dead worlds” where it would be relatively easy to terraform the planets.But I would guess this would be pretty rare.
    Gregory Benford-thank you for reminding me of the much anticipated novel by Kim Stanley Robinson. I almost forgot that it will be published in a month or so :)

  • Ronald May 6, 2015, 18:13

    I am looking forward to similar analyses of other solar type stars with planets in the solar neighborhood, such as 82 Eridani, 61 Virginis, Nu 2 Lupi.
    And even more so to the time that we have so many reliable data that we can really do modeling of planetary systems based on stellar type and elemental abundances.

  • Jenny May 7, 2015, 1:06

    These planets are huge, but what if either of these two words have Mars-sized moons? On Tau Ceti F life could have existed under frozen, cracking ice with thermal heating caused by tidal heating with other moons, Life that could then move to the surface when the habitable zone moved over the planet. In the case of Tau Ceti E, life could have retreated into the rocky layers of the planet or floated in the upper atmosphere (remember that on Venus there is an Earth-pressure environment high in the clouds.)

  • andrew July 31, 2015, 6:36

    @Hop David. “reach a level where we can even think about traveling to another star system.” We’ve had the technology and the ability to travel to another star system for over 55 years. Its called a nuclear pulse rocket. The idea is pretty simple, very powerful, passed initial testing and quite practical with today’s technology. We could easily reach another star system with one. Sadly NASA has blocked scientific progress and its monopoly on progress is only now beginning to break. NASA crushed Project Orion, the nuclear pulse rocket project, and chose a chemical rocket program run by former nazi scientists.

    In addition there are other crafts that we could make or at least think of today that could reach another star system. Such as a solar sail driven by a laser, some type of ion drive propelled by nuclear reactors, etc. A very promising idea is micro-sized spacecraft propelled by a solar sail. Really, just any craft that is based on an energy other than chemical rockets or batteries which are both very inefficient (though a nuclear battery is certainly better than the others we’ve used, such as the one used in Curiosity rover).

    I think the only practical spacecraft that we will be able to use anytime soon though is a nuclear fission based craft. Whether than be by a nuclear reactor or an external shaped nuclear charge explosion. We just have to have the will and the ability to get past the bureaucratic red tape.