≡ Menu

A Super-Earth in Transit (and a SETI Digression)

We’ve been finding planets using radial velocity methods — analyzing the gravitational effects of planets around their stars — since the mid-1990s, and the Kepler mission has brought the transit method to the fore, looking at the lightcurves of stars when planets pass in front of them as seen from Earth. Now we have new information about a transiting planet, 55 Cancri e, in a multiple planet system, information that has been developed by reanalyzing earlier radial velocity data. The new techniques were applied by Rebekah Dawson (Harvard-Smithsonian Center for Astrophysics), working in tandem with Daniel Fabrycky (UC-Santa Cruz) to predict the orbit of 55 Cancri e.

Earlier radial velocity data on the planet had suggested a tight orbit of 2.8 days, but the new analysis pegged the orbit at something less than 18 hours. The proximity to the central star meant that the chances of seeing a transit were higher than thought (the probability moved from 13 percent to 33 percent), leading to a space-based observation by the Canadian MOST satellite (Microvariability & Oscillations of STars). The result: 55 Cancri e is found to transit its star every 17 hours and 41 minutes, validating Fabrycky and Dawson’s method.

Image: This close-up of the constellation Cancer shows the location of 55 Cancri (circled in red). Its larger component, 55 Cancri A, hosts a planetary system that includes the hottest, densest super-Earth currently known: 55 Cancri e. Credit: Harvard-Smithsonian Center for Astrophysics / Created with Voyager 4, copyright Carina Software.

We now learn that 55 Cancri e is even more of a hell-hole than we suspected, reaching temperatures of 2700 degrees Celsius. The G-class central star, 55 Cancri A, is the closest known star to the Sun with transiting planets. It weighs in at roughly one solar mass, and is visible with the naked eye some 42 light years from Earth. Its system in interesting in many ways, with five known planets and a small red dwarf at approximately 1000 AU. The outermost planets are evidently Jupiter-class giants, but 55 Cancri e is a super-Earth now revealed to be of about eight Earth masses, with a diameter some 60 percent larger than our planet.

We’re talking about a planet almost as dense as lead. 55 Cancri e now joins the ranks of super-Earths with measured masses and radii, which gives us the opportunity to compare the properties of this class of world. The star here is the brightest of all known super-Earth hosts (2 magnitudes brighter than any other transit host, according to the paper on this work). But the transit depth — the fractional change in brightness as the planet transits the star — is among the smallest known. That makes follow-up observations trickier than expected. From the paper:

The shallow depth makes certain follow-up observations challenging despite the abundance of photons. To resolve the transit ingress and egress, and thereby improve estimates of the planet’s orbital inclination and absolute dimensions, it will be necessary to improve the signal-to-noise ratio in the phased light curve. This will require observations of more transits, or the use of larger-aperture telescopes. More daunting, but within the realm of possibility, is the detection of occultations and orbital phase variations that might betray the presence of an atmosphere…

But I also want to point to the paper’s conclusion, which notes “…there is some pleasure in being able to point to a naked-eye star and know the mass and radius of one of its planets.” True enough! And the new technique may help us refine the orbits of other planets discovered through radial velocity methods, Dawson noting “We’ve cleared up some confusion in the systems we studied, and we believe we’ve provided a way to avoid future confusion.”

The paper is Winn et al., “A Super-Earth Transiting a Naked-Eye Star,” submitted to Astrophysical Journal Letters (preprint).

A harmless digression: Ponder a habitable planet in a binary system like 55 Cancri A. Would the presence of a red dwarf practically on its doorstep, a star that could have planets of its own, give more of an impetus to any intelligent beings there to develop deep space technologies than would be the case around solitary stars like our Sun? I think that it would. In fact, I think widely spaced binaries are excellent SETI candidates for this very reason, offering a provocative sky with a nearby target, an interstellar mission nowhere near as daunting as we face in reaching Alpha Centauri. Goals count, and when they are tangible and near at hand, technologies may follow. Societies like this may be more space-minded than we are, and potentially more detectable.


Comments on this entry are closed.

  • Tom Baty May 2, 2011, 13:33

    Interesting article…wait and see if this method will work on other targets. Your “digression” I found intreging as well, never thought about that before and it is an excellant observation from the SETI point of view.


  • Alex Tolley May 2, 2011, 13:40

    “Would the presence of a red dwarf practically on its doorstep, a star that could have planets of its own, give more of an impetus to any intelligent beings there to develop deep space technologies than would be the case around solitary stars like our Sun? I think that it would. In fact, I think widely spaced binaries are excellent SETI candidates for this very reason”

    The target hypothesis is probably valid, as arguably our own moon was a stimulus. But bear in mind a sun within 1000 AU might also have a rather nasty effect on the local “Oort cloud”, increasing the bombardment of any hypothetical planet potentially inhabited by macro life forms. I’d like to see calculations that show that either this could not happen, or conversely, that the cloud would be swept fairly quickly reducing its effect by the time macro life emerged.

    For Dan Dare fans, the planet Cryptos in the constellation of the three Suns of Los (alpha Centauri?) , was the setting for the adventure involving the Crypts and the Phants. Both races had their home world around a different sun and had spaceflight.

  • Paul Gilster May 2, 2011, 14:15

    Good point re bombardment, Alex, as even Proxima Centauri (at roughly 10,000 AU) is possibly a factor in any cometary cloud around Centauri A and B, though obviously it would be much less of one than a dwarf at 1000 AU.

  • andy May 2, 2011, 15:44

    There have been a few other cases of confusion with alias periods in the exoplanetary catalogue, perhaps the most famous is GJ 581d, originally thought to have an 83-day orbit, later determined to be a yearly alias of the true 67-day period.

    Regarding perturbations from the red dwarf star on a hypothetical Oort cloud, I have to wonder whether an Oort cloud would actually form in the first place in such a binary. Nevertheless 55 Cnc A is not a particularly good candidate for habitable planets: it contains a giant planet 55 Cnc f at least half the mass of Saturn right in the habitable zone. If the typical scaling laws for satellite systems apply to the 55 Cnc A planets, it is unlikely that any of its moons are massive enough to maintain habitable conditions. Not sure if the system parameters are known well enough to make any statements about the stability of the Trojan points against perturbations from the other planets.

  • Steven Rappolee May 2, 2011, 18:47

    and that old sci-fi fav, the planetary system with two habitable worlds in its Goldilocks zone!
    even if the second world was less then ideal to the world with the thinking beings it might be a powerful force to the world with the thinking beings to explore that neighbor!
    But given the gravitational dance of worlds, what would be the chances of two such worlds? none around a red dwarf.
    possible around a sun like star? a larger star, yes but,
    that sun would have a shorter lifespan.

  • Paul Gilster May 2, 2011, 21:43

    Steven, the idea isn’t two planets in the habitable zone of the same star. It’s one planet each around both of the stars in a widely spaced binary.

  • Phil May 2, 2011, 23:41

    I note that 55 Cancri B might itself be a double. But has there been any RV or transit work done on B, or has it literally and figuratively been outshone by its now more famous cosmic partner?

    Red dwarf lovers want to know! :)


  • spaceman May 3, 2011, 0:27


    Interesting. Although I had heard of the 55 Cancri system, I did not know there is a red dwarf companion at 1,000 A.U.

    I like your digression, as I have had the same thoughts regarding widely separated binary systems as well. Almost certainly there exists within our galaxy, let alone the Universe as a whole, a fair number of the widely seperated binary systems each with it own habitable planet. I think a binary star at, say, 500 A.U. with its own system of planets would provide an irresistable target for exploration. The propulsion technologies enabling such a voyage would indeed pave the way for multi-light year interstellar travel!

  • Enzo May 3, 2011, 1:56

    Interestingly, 55 Cancri is another “compact” solar system around a G star, similar to Kepler-11. That is, with the exception of 55 Cancri d, all the planets are within Venus’ orbit.
    This article on the Systemic blog :

    Seem to find a trend in Kepler and our outer planets’ moons data that points in the direction of an ‘uncommon” solar system with planets so far apart.
    If “compact” systems are common, habitable earths will not be.

  • andy May 3, 2011, 2:22

    So is this planet’s density high or low? MOST says high, Spitzer says low. I’m somewhat sceptical that such a low-density planet could survive in such a strongly-irradiated environment though…

  • Bynaus May 3, 2011, 3:55

    Today, another paper on the transit of 55 Cnc e was put on arxiv.org.


    The transit was observed with the Spitzer space telescope, and interestingly, they find a larger radius of about 2 Earth radii…

    Re the bombardment problem, isn’t the Oort cloud in the solar system thought to start much more farther out than 1000 AU? On the other hand, the habitable zone of 55 Cnc A is occupied by a gas giant. Ok, it might have moons, but its a low mass giant, so it would have to have captured its Mars-to-Earth-sized moon during migration, which is not very likely. But its a very interesting system nevertheless!

  • kzb May 3, 2011, 7:54

    @ Enzo
    I think observational bias is something to keep very much in mind. The close-in planets are more likely to be detected. I’ve not read the link you posted yet though.

  • Tobias Holbrook May 3, 2011, 9:05

    The habitable zone giant might not be able to have habitable moons – but what about planets at it’s lagrange points? A Mars sized body could possibly survive in either L4 or L5…

  • ljk May 3, 2011, 9:11

    Remember that gas giant planets may themselves harbor life. See here:


  • andy May 3, 2011, 14:34

    But has there been any RV or transit work done on B, or has it literally and figuratively been outshone by its now more famous cosmic partner?

    As far as I’m aware, 55 Cnc B is too dim in the optical for radial velocity measurements to be particularly feasible. Hopefully the various instruments that are being developed to measure radial velocities in the infrared will be able to survey stars like 55 Cnc B in future.

  • Enzo May 3, 2011, 20:08


    I hope in observational bias too but the article in the link says :
    “It will take more time to determine whether the solar system is really all that weird, but with each passing month’s accumulation of fresh exoplanets, our eight-planet set-up manages to seem slightly less ordinary. Jupiter, for example, induces a 12 m/s velocity half-amplitude, and the high-precision radial velocity surveys have been operating for long enough so that if true-Jupiter analogs were the rule, then we’d perhaps be hearing of more of them being detected.”

    In other words, we are not finding cold Jupiters at 5 AU as much as we’d like for our solar system not to be weird. And 12 m/s surveys have been going on for quite a while now.

  • Rob Henry May 3, 2011, 21:00

    I am wondering which is the greatest boon to interstellar travel, a permanent close companion star, or the temporary close approach of a passing star? I calculate that even at our distance from the galactic centre, at least one star should approach within 30,000 au of us every ten million years.

    Option B for our interest in such systems comes if we believe that any ETI inquisitive enough to me interested in METI must be self destructive on timescale of a few thousand years. In that case we can believe that each member of the binary continually restocks the other after Armageddon, so dramatically boosting L for the system as a whole.

  • kzb May 4, 2011, 7:43

    @ Enzo
    Yes I’ve read the article now.
    I think it is still a fact though, that Earth-mass planets in Earth-like orbits are below the detection limit of the current Kepler data?

  • Anthony Mugan May 4, 2011, 11:00

    @kzb – You are right that at the moment the released Kepler data is over too short a time period to pick up true earth analgues. If I understand it correctly Kepler is capable of detecting earth analogues, but needs more time than the currently released data to detect planets with orbital periods of around 1 year. I think they need ideally at least four transits – it should get there if all continues to work well. On the more general points around how ‘typical’ the solar system is I think we need to be very careful around sample bias in the current data – although systems do seem to be coming in a wide variety of configurations

  • andy May 4, 2011, 15:29

    Interestingly, 55 Cancri is another “compact” solar system around a G star, similar to Kepler-11. That is, with the exception of 55 Cancri d, all the planets are within Venus’ orbit.

    From stability analysis you could get two planets with similar masses to Saturn in the gap between planets f and d. Such planets would have escaped detection so far. Certainly the masses are much higher, and the orbits more widely-spaced than those in the ultra-compact super-Earth systems (HD 40307, Kepler-11).

    As for the Jupiter-analogues, a few have started showing up: taking a look at the Extrasolar Planets Encyclopaedia, 55 Cnc d, HD 134987c, HD 99492c, HD 13931b, Mu Ara e, Ups And e and HD 154345b are examples of long-period giant planets in low-eccentricity orbits. Then again, the majority of these examples reside in multi-planet systems that look very unlike our own solar system.

  • Ronald May 9, 2011, 6:55

    With ref. to the harmless digression: this would probably be even more so in case of (wide) solartype binaries, such as (beside of course Alpha Centauri) Zeta Reticuli and 16 Cygni. Even if the other component did not possess any life-bearing planets but one or more terrestrial planets the incentive for terraforming might be enormous.

  • andy May 9, 2011, 14:47

    Ronald: would be nice to get some evidence of S-type planets around more than one star in the same system. To my knowledge the companions of planet-hosting stars that have been surveyed have been disappointingly RV-stable (including 16 Cygni A), and the few substellar objects in such orbits are more like brown dwarfs than true planets.

  • Ronald May 10, 2011, 4:58

    andy: are you implying that in planet-hosting binary systems (in most cases) only one of the component stars will have planets and the other component star will be bare?
    That would be quite significant with regard to planet formation modeling. But for (very) wide binaries I don’t see a good reason for this.
    Maybe the reason for 16 Cygni A being planetless is its great brightness (1.7 * solar luminosity) having blown away (inner) parts of the primordial dust disk?
    Interestingly 16 Cygni B has a super-Jupiter of at least 1.7 (possibly 14) Mj at 1.66 AU, the very outskirts of its HZ, which may have a large moon, a ‘moon of Endor’ or Pandora like situation.
    I am very curious to know what kind of planets Zeta 1 and 2 Reticuli have, if any.
    (Wide) binaries consisting of two solar type stars are (very) rare.

  • andy May 10, 2011, 15:40

    Ronald: I’m not entirely sure. In the majority of cases, the star which is not a known planet-host is too dim for RV measurements, in others it may be that only close-orbiting, massive planets can be ruled out around the secondary. The lack of double S-type planetary systems may be a matter of small number statistics and the low incidence of gas giant planets in general and hot Jupiters in particular.

    On the other hand, there are definite cases of binary systems with detectable dust around only one of the stars, which may indeed imply that there are a fair few systems out there which only formed planets around one star. Since we have examples of S-type planets around both primary and secondary stars, it does not look like there is a hard rule as to which of the two stars will end up with planets in such a case.

    As to 16 Cygni, it is in fact triple: there is a red dwarf star in a wide orbit around 16 Cyg A. On that point, all planets so far known in triple star systems (triples being composed of a binary and a distant tertiary) are orbiting the tertiary, rather than either of the stars in the binary.

  • Ronald May 11, 2011, 4:20

    andy: “On that point, all planets so far known in triple star systems (triples being composed of a binary and a distant tertiary) are orbiting the tertiary, rather than either of the stars in the binary.”

    Ouch! If consistently true, that does not bode well for Alpha Centauri.
    But then again, it doesn’t always seem to be so: there is a super-Jupiter around 16 Cygni B (if not a brown dwarf).

  • andy May 11, 2011, 13:23

    Ronald: 16 Cygni does follow this rule: the system structure is (AC)-B. Stars A and C form the binary, with star B (the planet-host) being the tertiary.

  • Ronald May 12, 2011, 6:56

    andy, thanks, I stand corrected.

    I wonder whether there is a good explanation known for this. If it is always the (distant) tertiary star having the planets, it may primarily be determined by distance, indicating that the distance between the binary pair is simply too close for a stable planetary system to develop and persist. Which also does not bode well for many (close) binaries.

  • ljk July 21, 2011, 17:48

    Nearby “earth-like” planet: not so much

    There’s some chatter on the web right now over a new scientific paper about a nearby exoplanet, and what I’m seeing are people speculating that it might be earth-like. Technology Review even titled their article “Astronomers Discover Habitable ExoEarth Orbiting Binary Star”.

    The problem with that is that the planet’s not terribly earth-like, and it may not be habitable*.

    So what’s the deal? I read the journal article (PDF), and this really is a good story, just not the one I’m seeing the chatter about.

    55 Cancri is a nearby binary star at a distance of about 40 light years. One star is a dinky red dwarf, and the other is a fairly Sun-like star, though somewhat smaller and cooler. It’s also much older, roughly 10 billion years old, more than twice the age of the Sun. It’s actually at the point where it’s starting to evolve into a red giant, and is called a sub-giant.

    Back in 2007 it was announced that at least five planets orbit the bigger of the two stars (called 55 Cancri A; confusingly the red dwarf is 55 Cancri B (note the capital letter), while the planets are called b-f (lower case)). They range in mass from 0.026 to 3.84 times that of Jupiter (8.3 to 1200 times the mass of the Earth). 55 Cancri e is the lowest mass of these, but is extremely dense and hot, so not at all earth-like.

    55 Cancri f is the interesting planet, though. The astronomers in question observed the star using an interferometer, allowing extremely precise measurements of the star’s size, which in turn yielded very accurate numbers for its temperature and mass. All these together can be used to figure out its “habitable zone”, the region around it where an orbiting planet would have liquid water on its surface.

    Now right away, I’ll say that finding the HZ (as we in the know call it) is not really straightforward. For example, a planet that has a thick atmosphere can be farther from its star and still have water due to the greenhouse effect; in fact, without air the average surface temperature of the Earth would be below freezing! And the greenhouse effect depends on what’s in the atmosphere, its density, and so on. So I am wary of any declarations of planets being habitable based on this alone.

    Full article here:


  • ljk January 14, 2012, 2:18

    Re-thinking an Alien World

    Jan. 13, 2012: Forty light years from Earth, a rocky world named “55 Cancri e” circles perilously close to a stellar inferno. Completing one orbit in only 18 hours, the alien planet is 26 times closer to its parent star than Mercury is to the Sun.

    If Earth were in the same position, the soil beneath our feet would heat up to about 3200 F. Researchers have long thought that 55 Cancri e must be a wasteland of parched rock.

    Now they’re thinking again. New observations by NASA’s Spitzer Space Telescope suggest that 55 Cancri e may be wetter and weirder than anyone imagined.

    Full article here:


  • ljk May 9, 2012, 17:11

    Light From a ‘SuperEarth’ Detected for the First Time

    by Nancy Atkinson on May 8, 2012

    The star 55 Cancri has been a source of joy and firsts for planet hunters. Not only was it one of the first known stars to host an extrasolar planet, but now the light from one of its five known planets has been detected directly with the Spitzer Space Telescope, the first time a ‘smaller’ exoplanet’s light has been detected directly.

    Planet “e” is a super-Earth, about twice as big and eight times as massive as Earth. Scientists say that while the planet is not habitable, the detection is a historic step toward the eventual search for signs of life on other planets.

    “Spitzer has amazed us yet again,” said Bill Danchi, Spitzer program scientist. “The spacecraft is pioneering the study of atmospheres of distant planets and paving the way for NASA’s upcoming James Webb Space Telescope to apply a similar technique on potentially habitable planets.”

    Full article here: