How Common are Terrestrial, Habitable Planets Around Sun-Like Stars?

Article written: 27 Sep , 2011
Updated: 24 Dec , 2015


Once again news from the Kepler mission is making the rounds, this time with a research paper outlining a theory that Earth-like planets may be more common around class F, G and K stars than originally expected.

In the standard stellar classification scheme, these type of stars are similar or somewhat similar to our own Sun (which is a Class G star); Class F stars are hotter and brighter and Class K stars are cooler and dimmer. Given this range of stars, the habitable zones vary with different stars. Some habitable planets could orbit their host star at twice the distance Earth orbits our Sun or in the case of a dim star, less than Mercury’s orbit.

How does this recent research show that small, rocky, worlds may be more common that originally thought?

Dr. Wesley Traub, Chief Scientist with NASA’s Exoplanet Exploration Program outlines his theory in a recent paper submitted to the Astrophysical Journal.

A possible habitable world? Credit: NASA/JPL

Based on Traub’s calculations in his paper, he formulates that roughly one-third of class F, G, and K stars should have at least one terrestrial, habitable-zone planet. Traub bases his assertions on data from the first 136 days of Kepler’s mission.

Initially starting with 1,235 exoplanet candidates, Traub narrowed the list down to 159 exoplanets orbiting F class stars, 475 orbiting G class stars, and 325 orbiting K class stars – giving a total of 959 exoplanets in his model. For the purposes of Traub’s model, he defines terrestrial planets as those with a radius of between half and twice that of Earth. The mass ranges specified in the model work out to between one-tenth Earth’s mass and ten times Earth’s mass – basically objects ranging from Mars-sized to the theoretical super-Earth class.

The paper specifies three different ranges for the habitable zone: A “wide” habitable zone (HZ) from 0.72 to 2.00 AU, a more restrictive HZ from 0.80 to 1.80 AU, and a narrow/conservative HZ of 0.95 to 1.67 AU.

After working through the necessary math of his model, and coming up with a “power law” that gives a habitable zone to a star depending on its class and then working out how many planets ought to be at those distances, Traub estimated the frequency of terrestrial, habitable-zone planets around Sun-like (Classes F, G and K) stars at (34 ± 14)%.

He added that mid-size terrestrial planets are just as likely to be found around faint stars and bright ones, even though fewer small planets show up around faint stars. But that is likely because of the limits of our currently technology, where small planets are more difficult for Kepler to see, and it’s easier for Kepler to see planets that orbit closer to their stars.

Traub discussed how the quoted uncertainty is the formal error in projecting the numbers of short-period planets, and that the true uncertainty will remain unknown until Kepler observations of orbital periods in the 1,000-day range become available.

Check out our previous coverage of exoplanet detections using the Kepler data at:

If you’d like to read Traub’s paper and follow the math involved in his analysis, you can do so at:

Learn more about the Kepler mission at:

Source: arXiv:1109.4682v1 [astro-ph.EP]


46 Responses

  1. Anonymous says

    Kepler to study alpha centauri system? This must be one of its primary missions

    • Ray Fowler says

      Kepler was designed to stare endlessly at one patch of stars for the duration of its life. This allows it to tease an enormous amount of data from the targeted stars that have planetary transits (duration is very important for this).

      Even if it could be aimed at Alpha Centauri (it can’t, wrong hemisphere), we would lose all of the observing time for hundred of stars with transiting planets. And unless Alpha Centauri has planets (questionable for a binary system), they would still need to be aligned perfectly (fluke chance) for Kepler to get ANY data at all.

    • Torbjörn Larsson says

      Kepler’s primary mission was to establish mass statistics of exoplanets, so that it could be a confirmed target to look at nearby stars with later missions. It also provides data on how to best design these missions.

      Missions to look over the hundred or so nearest stars for planets have been described on the web, IIRC. However, Alpha Centauri has been scrutinized for a long time without any findings. I would think the likelihood it makes an interesting target is minute by now, but I don’t really know.

  2. Anonymous says

    30 %. it becomes higher and higher in each new paper. Why do they not just claim every star has a planet in its habitable zone.

    • Torbjörn Larsson says

      How could they do that, they lack the data to extrapolate better than they do? However, see my above comment on the bias the different methods have to work with.

  3. If we found one, say…15 light years away. Would there be enough will to really research it to make it a priority? Would we send off a small, fast probe right away so that we could hear back in 55 years? Has all this got a purpose? Is there any chance we could send a robotic midwife and a bunch of frozen eggs and sperm and given the right or “wrong” circumstances, put a colony in motion?

    • William Sparrow says

      I’m not sure I understand how we’d “hear back” in 55 years. AFAIK, the fastest probe currently travelling is Vogager 1 at 17.1 k/s relative to the Sun. That’s a far cry from C. I’m all for sending out more rovers, but until spacecraft propulsion technology advances to some reasonable percentage of C, I’m not sure it’s in the cards.

    • Anonymous says

      We couldn’t do any of those things, anytime soon. Remember,the fastest and one of the smallest probes we’ve ever launched, is still on its way to simply flyby Pluto…

      But it might spur work on larger space and ground based telescopes. We won’t be doing interstellar craft for some time, but given money (and sufficiently low launch costs), we *can* build light buckets as big as we want…and they’d serve more than that one purpose.

    • Anonymous says

      Before we start exploring systems light years away we should first start to look where to go to. Bigger telescopes might work but all have limits. To get some really fancy pictures we can use Sun’s gravity lens at some 550AU+. This would still require some tech & propulsion advancements but using this technique we could resolve even ground features of planets many light years away… now that would be cool 🙂

    • Anonymous says

      Before we start exploring systems light years away we should first start to look where to go to. Bigger telescopes might work but all have limits. To get some really fancy pictures we can use Sun’s gravity lens at some 550AU+. This would still require some tech & propulsion advancements but using this technique we could resolve even ground features of planets many light years away… now that would be cool 🙂

    • Torbjörn Larsson says

      These observations have many purposes. Besides the cultural effect of answering a really old question in many cultures, “are we alone”, there are scientific ramifications.

      One of the larger in my mind is that it seems possible that over time constrain conditions for abiogenesis. That would be helpful to narrow down possible pathways taken here and elsewhere. That goes to answer “how did we get here”.

      • Anonymous says

        Given sheer probability I feel we can say confidently we aren’t alone. But we are isolated given the huge distances between systems, which I find oddly perfect. Because I don’t know that we could ever engage another life form without setting an interstellar military buildup and plan in motion.

        What is it that we really seek from the knowledge of the existence of other life elsewhere? The abundance of varied mysterious life all around us on this planet alone might be enough wonder for one lifetime without messing with other solar systems.

  4. Anonymous says

    I don’t quite get the picture of possible habitable world. It looks like it might be tidally locked (it appears there are big differences between northern and southern hemispheres) and illuminated from the ‘top’. Also it appears it doesn’t have an atmosphere – but what are those ‘clouds’?

    The big thing in the middle might be some kind of ring system but what is the one above it (blown atmosphere?)? And also that foggy thing in top right corner – is it some kind of nebula? Dunno what was the artist’s intention to express here… any ideas?

    Apart from that It certainly looks very nice and ‘alien’ 🙂

    Oh, and of course – hurray for new statistical rockies! 😉

    • William Sparrow says

      The photo probably should have been tagged “Artist’s Conception” Or, were you kiddin’?

      • Anonymous says

        Yeah, I’m just curious what that conception is 🙂

        I could draw an elephant and tag it “possible habitable world”, so just want to know how and if such thing corresponds to a planet that can be found somewhere in the universe.

    • It’s just a artist’s rendering of a potentially habitable exoplanet that I thought would fit nicely with the article.

      • Anonymous says

        And it surely does! 🙂 (it’s one of the most interesting I’ve seen in exoplanets articles)

        I always like to have some “thought fun” looking at such pictures (the same as reading SF books and thinking if it is plausible) – it doesn’t take any entrainment away from it. And those clouds/rings made me curious.

  5. Anonymous says

    It’s frustrating that all the researchers use Kepler data from the first 136 days. It’s been in orbit for about 2 years. Why are we not looking at more of the data. Who is restricting a 500 day data set? the first several months identified 1,200 candidates. Certainly the candidate list is double that by now. Why not disclose the information? Are they going to wait until mission end at 3.5 years and then give us a few more months of data?

    • Anonymous says

      I can understand your frustration, however, I tend to step back and look at the numbers so far, it has a high potential to be profoundly exciting, to myself at least it already is.

      The area of sky Kepler is fixed on is miniscule, it is also missing a lot of systems whose transit planes are not aligned for Kepler to see candidates passing their host star.
      I think of Kepler as looking at planet earth but only being able to see the atoms on a single blade of grass, such is the comparison with Kepler’s view of our universe, if we can extrapolate what Kepler has seen so far to the rest of the universe I think profound will be an understatement.
      I for one am very grateful to be living at what seems to be a very special time in human history, not just for Kepler’s discoveries but for all the other potentially profound discoveries popping up all over the place.

    • HeadAroundU says

      Because some people work on it since 80s and they want to have them first.

      It’s around 1700 candidates now.

      Q3 data are out. We are still looking at Q2 on planethunters though.

    • Anonymous says

      Kepler’s design makes it more likely to find planets that are really close to their sun and really big. It was expected that it would find a large number of planets early on (the ones that transit their stars really quickly). The discovery of new planets becomes less likely for Kepler the longer it takes the planet to orbit their sun and the farther away they are from their sun. It definitely wont be able to find the same number of planets that it did in the first few months of the mission, although it will still find a few hundred more.

  6. Anonymous says

    While there may be 33% of star systems with habitable planets in the so called habitable zone, the question is how stable those orbits are. A planet may be gravitationally perturbed so it wanders in and out of that zone.


    • Anonymous says

      I think our universe is big enough and diverse enough to contain all variations of systems, there’s even room enough for multiples of all variations.

      • Anonymous says

        There may be many planets with life, but on a prokaryotic-like level. Mars might in fact be such a planet. I think it requires some stability in the orbital dynamics of the planet in order for there to be complex life. For instance, Mars may episodically cycle through comparatively warm periods and dry-ice temperature conditions. Currently it is in a dry-ice phase. Only bacteria-like life is robust enough to endure such extremes, and extreme changes. Mars has no large moon to stabilize its angular momentum from external perturbations, and its orbit gets perturbed more by Jupiter than Earth’s orbit. It is also small with a tenuous atmosphere. So life on Mars might be some sub-regolith ecosystem of single cells capable of making a living under these conditions. Mars might be a model for lots of planets out there.

        So the Gliese planet, or maybe a couple of them, may harbor life on a prokaryotic level. This mini-solar system has these planets gravitationally interacting rather strongly with each other. The orbits of these planets may shift around some. This will be the case for a G-class star with a terrestrial planet at around 1AU, but with a jovian planet at around 2AU. Such a planet might have a prokaryotic-like ecosystem, but if conditions episodically shift through extremes due to orbital perturbations that life may not evolve into the sort of complexity we have on Earth.

        Earth-like planets with complex life might be very exceptional with rather few of them in any galaxy.


      • Torbjörn Larsson says

        But if we go the bayesian route we can make any likelihood as small as we wish. It is good for just-so stories (or better, for making comparisons), but not for prediction.

        For example, if you are concerned about climate stability, it may work both ways. Too unstable climate, and we could have simple ecologies. Too stable climate, and again we could have simple ecologies. The truth is we don’t know.

        If we look for predictive models, I suggest that anyone concerned about the effects of orbit and rotation axis stability et cetera predicts Earth’s twin Venus first. Venus has a more circular orbit and a seemingly perfectly stable axis. Despite the presence of a near resonance Earth orbit neighborhood and “no large moon to stabilize its angular momentum from external perturbations”.

        Venus has also a dense atmosphere (too dense, in fact) protecting its surface from solar and cosmic radiation despite lacking a strong intrinsic magnetic field. A prediction is that it has lost its water to space, did not develop plate tectonics and therefore ended up having all its carbon in the atmosphere. But again, we don’t know for sure.

        Conversely, if we are concerned about tidal locked planets, Venus may again be a case study to predict, it rotates that slowly. Yet it has an even and stable climate over its whole surface.

        To sum up, I don’t see how we should be concerned about secondary effects when we are still trying to pin down first order habitability. And these secondary effects looks like we wouldn’t have to be overly concerned about them either at the outset, when we eventually get around to have to consider them.

      • Anonymous says

        Of course we do not have a deterministic model for the evolution of planets. Yet I suspect that planets have to have a measure of stability in their orbit, the proper energy flow from their parent star, and an ambient or average temperature comparable to Earth for there to be complex life.


      • Anonymous says

        It seems as though single-celled life will be more common regardless. They are the hardest things to kill on Earth and some of them require very little to survive. Although I am sure that there are other planets in the universe with complex life like ours, I would think that these planets would be very few and far between. Unfortunately, simple life is much harder to detect remotely and we will have to wait to see if our predictions are true. I would imagine that once we have the ability to look back on the data used in these predictions with the benefit of new discoveries, we will find that they are notoriously unreliable given the number of unknown or guestimated variables that are included in the math.

      • Anonymous says

        I tend to agree that this is plausible. Bacteria and prokaryotes are what really run this place we call Earth. Conditions have been sufficiently right for there to evolve euckaryotic associations of prokaryotes and then multicellular organisms. Yet this can be seen as an evolutionary trajectory that permits prokaryotes to acquire more solar energy. Plants acquire photo-energy and then leaves and other products are dropped with provide bacteria in the soil energy and materials. Animals are bacterial eco-systems, where for every somatic cell for that organism or animal there are 10 prokaryotic or bacterial cells. This includes us. So a planet with a complex biosphere is one which permits its simple celled organisms to increase their complexity in such a way as to increase the bio-energy flow on the planet.


      • Anonymous says

        I agree with what you’re saying. I just think that making predictions about what percentage of planets have life or complex life is questionable given that we only have one planet so far to study. I am guessing that it will be very hard to dectect a planet with life because many forms of life don’t even need sunlight or oxygen to survive (i.e. gold mine or hydrothermal vent type ecosystems). If life survives in a rock fissure somewhere deep underground on a nearby planet, how will we find it? I hope that I’m proved wrong and we find many planets with complex life. Then will come the frustration of not being able to study it directly.

  7. Anonymous says

    Our asteroid belt was a planet that orbited too close to jupiter in the past. I doubt binary stars under 1/2 light year apart could have stable solar systems, without their planets being bombarded by lots more comets then our solar system has, which condensed everything into planets, moons, and leaving only the possible oort cloud for far drifting cometary orbits. The best candidates for life are stars not near supernovas, and I think the earth has been lucky for a long time not to have received excessive gamma rays from SN’s. Radiation in the intergalactic medium when globular clusters containing old stars likely hop between galaxies, would be too intense and destroy life on planets, unless it was somehow shielded inside a dense nebulae. The big question is why haven’t we found evidence of life in outer space, are we too inconsequential to be contacted by anything of far greater intelligence and power?

    • Torbjörn Larsson says

      Our asteroid belt was a planet that orbited too close to jupiter in the past.

      According to astronomical data of their composition, and independently from planetary formation models such as the Nice model and later: no, it wasn’t. The asteroids and presumed planetoids like Ceres and perhaps Mars were disturbed and scattered by Jupiter migrating inwards, before they got the chance to aggregate or disperse in other ways.

      I doubt binary stars under 1/2 light year apart could have stable solar systems, without their planets being bombarded by lots more comets then our solar system has, which condensed everything into planets, moons, and leaving only the possible oort cloud for far drifting cometary orbits.

      You seem to mix two phases of solar system formation here, planetary aggregation from aggregating planetoids and later dispersal of non-planetary material. In our case the later was episodic, the Late Heavy Bombardment (LHB), probably triggered by giant planet migration (Jupiter, Saturn, Neptune). It is an open question how it happens in other systems, I think.

      Life on Earth survived after its aggregation was finished at the earliest, even if earlier abiogenetic attempts can’t be excluded. There are recent papers that shows abiogenesis unlikely in asteroids, comets and smaller planetoids however.

      When life arose isn’t clear. But models of the later LHB shows that even if it was an order of magnitude more intense in incoming mass/time, it likely couldn’t eradicate any existing cellular life. Prokaryotes procreate and disperse too rapidly and too deep into the crust for such sterilization events to finish them off.

      The best candidates for life are stars not near supernovas, and I think the earth has been lucky for a long time not to have received excessive gamma rays from SN’s.

      The solar system has gone some 20 times around the block (Milky Way rotation) and has dodged the bullet from different passing stars, as most stars have I think.

      The big question is why haven’t we found evidence of life in outer space,

      It isn’t “big” for abiogenesis, since the many possibilities makes such a weak question. Most anything could be the reason, besides frequency of habitability and abiogenesis.

      To sum up, we should probably be surprised if we didn’t observe inhabited planets in the next few decades.

      • Anonymous says

        I would be surprised if we did find an inhabited planet in the next few decades. It is incredibly hard to tell whether a planet has life when you can’t even directly observe the planet. Even the nearest potentially inhabited planets are incredibly far away from an observer’s point of view. How are we going to conclusively tell that a planet harbours life when the life is so small that you need a laboratory test to tell whether it’s there? Without boots on the ground of the new planet, I don’t think we’ll have an easy time finding such a planet, although time will tell if I’m right or not. We will find many candidate planets for detecting life, but are we really at a point in our scientific evolution where we could tell for sure?

  8. Gary Whitton says

    It would seem to me that the most suitable star for a habitable planet to orbit would be the one that offers the longest period for life to potentially take root. I have heard of stars somewhat smaller than our sun that last an incredibly long time…like the entire history of the known universe if not longer. Of course what I don’t know is how much the habitable zone shifts over the life cycle of the star.

    • Anonymous says

      I think about that a lot!

    • Torbjörn Larsson says

      Our type case, Earth, shows that life can arise in the first 1 Ga though.

      I think that can be severely constrained, but the observations and models that would give that are still weak and uncertain.

      However those who study abiogenesis thinks it would take on the order of tens of millions of years in the low end (optimists) and hundreds of millions in the high end (pessimists).

      What seems to take time is the development of an oxygenated atmosphere conducive for energetic multicellular life, aka “complex life”.

    • Anonymous says

      The reality is we just don’t know and until we have planets that have life to study, we are just making somewhat random guesses. It is possible, for example, that solar systems with longer lives have more time for hazards that can wipe life out completely. Life on Earth has nearly been exterminated a few times so far. Maybe it will be sometime in the not too distant future (geologically speaking).

  9. Anonymous says

    first in a paper they say that 1 tot 2 % has a earth sized planet in the habitable zone, and now someone claim it is 34 % with the same data. How can it be so different

    • HeadAroundU says

      Well, it’s only for F, G and K stars. Is the previous paper for all the stars?

      • Anonymous says

        no only for sunlike stars. That is F,G and K type stars, if i am correct. In that paper the are only looking at earth mass planet in the habitable zone. In this paper he look at all sizes of Terrestrial Planets, if i am right. It is still to soon to tell how much Habitable Planets are Around Sun-Like Stars. This is just speculation. Over 2 years they will have very different models about it

      • HeadAroundU says

        Radial velocity method can give us mass of a planet. Kepler doesn’t do it, with some exceptions, so that’s why it’s different. Or if you just strictly take 1 earth mass then no wonder it’s different.

        Lots of work must be done, but things get published and this is an interesting info for people.

  10. Anonymous says

    When looking for Earth-like planets in the habitable zone, it is important to realize that rock-centred moons can also exist in such a zone. In our solar system, the zone exists from the orbit of Venus to that of Mars. If a characteristic of the zone is the possibile existence of liquid water usually thought necessary to sustain life, then we must discuss Earth where life was discovered a while back, Mars, Europa, and one or two other moons including Encyladus. Our star provides the energy needed for liquid water on Earth and on very limited and rare occasions today on Mars where there was much more in the past, but other sources of energy have been responsible for liquid water on a few important moons of Jupiter and Saturn. What is the definition of “Habitable Zone?” Does it mean “possible habitat for higher life forms?” or, “possible habitat for any life form?” If the later, then the habitable zone is much larger in our own solar system and could easily be so elsewhere. Furhermore, we have evidence that more liquid water existed in our own solar system in the distant past then today. Just look at Mars, and the ice on many moons and dwarf planets beyond Mars. They can’t have always been frozen solid. If the habitable zone around our own star shifts over billions of years, couldn’t the same happen in other planetary systems? Doesn’t this change the search?

  11. Torbjörn Larsson says

    Keep in mind that HARPS finds at least twice as many planets than Kepler does: “The ongoing High Accuracy Radial velocity Planet Search (HARPS) has found that 30-50% of GK dwarfs in the solar neighborhood host planets with Mpl &lte; MNep in orbits of P  50 days. At first glance, this overall occurrence rate seems inconsistent with the planet frequency measured during Q0-Q2 of the Kepler Mission, whose 1,235 detected planetary candidates imply that ~ 15% of main sequence dwarfs harbor short-period planets with Rpl < 4 R?.

    A rigorous comparison between the two surveys is difficult, however, as they observe different stellar populations and measure different planetary properties.”

    [The paper goes on to how predict the difference, in an interesting manner.]

    Perhaps Traub can correct for such biased statistics, perhaps not. In any case the goal to find habitable terrestrials looks good!

  12. Anonymous says

    Habitable planets could receive DNA from outer space. scientists showed DNA self-assembles on carbon nanotubes. Buckyballs detected were believed created during supernovas, and probably graphenes were detected in the magellanic cloud. DNA could form in outer space this way, and seed most of the habitable planets for life. Man has the most complex DNA known on earth, of course more complex alien beings exist. Probably contact hasn’t happened, because of the great distances and light speed barrier. Yet, if our computers can detect and image alien planets, surely aliens living within 60 light years distance, could watch earth using complex electronics and model human life forms by studying the DNA. They could not save us from nuclear war destruction unless they’ve already sent ships in space to save us.

  13. Jérémy says

    Oh, that “habitable” zone…

    What about other factors like greenhouse effect and the proportion of greenhouse gases in the atmosphere – like carbon dioxide, water vapor or nitrous oxide – , the presence of moon(s) around the planet to stabilize it on its axis, or the magnetic field? It’s not really sufficient to judge on the distance between an exoplanet and its star. And what about, say Europa-like satellites which could perhaps be suitable for life, supposing that gravitational tides undergone due to the interaction with their planet would allow the existence of liquid water? And what about underground life? If you consider Mars in our solar system and not in the so-called “habitable zone”, the very discovery of such forms of life would be groundbreaking and show the limitations of so restricted a concept that is the “habitable’ zone .

    This “not-totally habitable zone with lots of aspects being ignored and/or dismissed” should really be deeply reconsidered and modified.

    • Jérémy
      It’s not really sufficient to judge on the distance between an exoplanet and its star.
      True. There are plenty of other factors that are potential determinants of an exoplanet’s habitability. The only one we can reliably measure right now is the distance between the planet and star. In just the right conditions we can get a glimpse of atmospheric gas composition, but I don’t think we have the ability to do this consistently.

      While we can’t rule out the possibility of exo-life living in Mars-like or Europa-like or Titan-like conditions around another star, we also can’t yet rule it IN since we haven’t actually found any other kind of life anywhere other than a planet in the habitable zone of a main-sequence G-class star.

      We have one data point.

      I would not be at all surprised if we find another data point or two within our solar system, and hope we do within my lifetime. It’s possible that there’s a hydrogen-eating prion in the methane lakes of Titan just waiting to be discovered. But until then, the search for habitable planets has to be defined by what we know to be habitable, rather than what we imagine could be habitable.

Comments are closed.