≡ Menu

COROT Finds a Small, Rocky World

The COROT mission’s 27-cm telescope has discovered the smallest exoplanet yet, with a diameter less than twice that of Earth. COROT-Exo-7b orbits a Sun-like star and highlights the ongoing space-based investigation into rocky worlds that is drawing ever closer to an Earth-mass object. This is the kind of work COROT was designed to do, flagging planetary transits across the face of a star from an orbital perch that allows long periods of uninterrupted observation and the chance to measure the size of the planets found. ESA’s Malcolm Fridlund discusses the significance of the find:

“This discovery is a very important step on the road to understanding the formation and evolution of our planet. For the first time, we have unambiguously detected a planet that is ‘rocky’ in the same sense as our own Earth. We now have to understand this object further to put it into context, and continue our search for smaller, more Earth-like objects with COROT.”

Finding a ‘super-Earth’ is one thing when you’re using radial velocity methods that peg its mass, but quite another when a transit allows a direct measurement of its size. COROT’s success reminds us that we’re at the dawn of the era of characterizing rocky worlds around other stars, an investigation that will tell us much about how common terrestrial planets are in an exoplanet population now dominated by gas giants.

What we know about COROT-Exo-7b is that it is a hot place indeed, orbiting its star in less than a day and close enough to it to boast temperatures between 1000 and 1500°C. Our online discussion began to develop overnight, after David Blank (James Cook University, AU) passed along the link to early data showing a world with 0.035 the mass of Jupiter and a radius 0.13 as large as that planet. That points to iron, and would make COROT-Exo-7b somewhat denser than Mercury.

This ESA news release speculates that the planet’s surface is covered in molten lava, or perhaps dominated by water vapor. In any case, its internal structure should provoke much further study. The paper on the new discovery is Léger et al., “Transiting exoplanets from the CoRoT space mission VII. COROT-Exo-7b: The first super-earth with radius characterized,” to be submitted to Astronomy & Astrophysics.

Addendum: Be sure to read systemic‘s take on COROT-Exo-7b, from which this:

Everything about CoRoT-7b reemphasizes the fact that planets are wont to turn up in every corner of parameter space to which observations are sensitive. In this case, a V=11 K0V star in the direction of the galactic anti-center displays 176 individual 1.5-hour 0.3 mmag photometric dips with a strict 0.854 day periodicity. These measurements suggest a 1.7 Earth-radii planet with a 20-hour year — a world that makes 51 Peg b look like Fargo North Dakota.

And this:

In any case, it’s a remarkable detection, and will be hugely influential as soon as the mass is confirmed. The planet is orbiting at only four stellar radii — with the star filling nearly a thousand square degrees of sky…

Comments on this entry are closed.

  • John Hunt February 4, 2009, 13:54

    This new exoplanet is, unfortunately, 390 light-years away which means that nobody’s going to suggest that it is a legit target for an interstellar mission that we should get started on right away. And herein is the rub. We are likely going to discover that our first Earth-class exoplanet with liquid water is many light-years away. It may be that there is not such an exoplanet within 20 light-years even all exoplanets to that distance been discovered. We may not be able to make the argument that there is a justifiable exoplanet target worthy of an interstellar mission if it must have liquid water and perhaps atmospheric evidence of primitive life. So, whereas an exoplanet in the “habitable zone” would be very exciting it probably won’t be our initial target for an interstellar mission.

    If we are going to make the argument, then we need to justify why we need to send a craft to a nearby exoplanet which doesn’t have liquid water and shows no evidence of bacterial life. Eventually when the cost of an interstellar mission becomes low enough and the speeds and the reliability of said missions are high enough then yes, we will be able to make that argument. But I am thinking that that point will probably not come this century, well after humanity is facing the consequences of our own cheap, self-replicating technology.

    So I suggest that we make the primary argument not based on the need for science but for the need to perpetuate humanity away from Sol system.

  • P February 4, 2009, 20:00


    That sounds a little pessamistic to me – we havent even found our first ‘earth analogue’ in a hZ orbit! There also must be quite a few (dozens?) low mass stars within 20 ly that we havent found yet and with the recent enthusiasm for the habitability of tidally locked planets around M dwarves I’m looking forward to the discovery of earth sized worlds within 20ly in the next 5 years.

    At least we shouldnt give up until Kepler [x-fingers] has done its work.


  • kurt9 February 4, 2009, 20:22

    Detection of exoplanets anywhere is useful because it provides the statistical background for deriving believable estimates of the presence of “Earth-like” worlds in the galaxy. Every discovery adds a data point to work with.

  • Colin Weaver February 4, 2009, 21:45

    I think it is way premature to declare the search for exoplanets within 20 years to be exhausted.

    We’re only picking up the low hanging fruit that stands out via the current methods, with much refinement of technique to come.

    If there were a system similar to our own around a star within 20 light years, how much would we detect with current methods?

  • Ronald February 5, 2009, 5:05


    we should distinguish between life-bearing planets (even if only micro-biological, which, I think, we cannot distinguish from higher life by spectroscopic analysis alone from this distance, anyway) and terraformable earthlike planets.

    The first category may indeed appear to be relatively scarce i.e. the nearest such planet may be relatively distant (many tens, even hundreds of ly?).

    However, the second category requires not much more than a roughly earth-mass planet (from about 1/3 to about 3 Me) in the habitable zone near a sunlike star, with liquid water, a transformable atmosphere (reasonable density and containing at least CO2 and N2), and maybe a few more requirements. This category is probably much more common and would be terraformable in the near future.

    COROT and Kepler are mainly statistical methods and will not give a systematic overview of nearby terrestrial planets and their characteristics. For such we will need missions like TPF, Darwin, maybe ELT and the like. And of course a future telescope in the sun’s gravitational focus at 550 AU (sigh, I was born too early).

  • Ronald February 5, 2009, 5:24

    Further to my previous post, on a rather optimistic note (but not unrealistic), I expect and dare bet, that small rocky planets (terrestrial in the broader sense), and even very promising ones, will turn up in the near future even around sunlike stars within a few tens of ly, such as 61 Virginis, Alpha Mensae, Beta Canum, 18 Scorpii, etc.

    Rationale behind this expectation are the positive correllation between metallicity and prevalence of planets, and the inverse correllation between planet size and abundance.

  • Ronald February 5, 2009, 11:10

    @Colin Weaver: I could not agree more.
    “If there were a system similar to our own around a star within 20 light years, how much would we detect with current methods?”
    Jupiter, with time and effort, but probably not even Saturn (yet).

  • John Hunt February 5, 2009, 14:23


    I, of course, hope that you are right that we’ll find rocky planets nearby and soon.

    When you look at the list of nearest stars and compare it to the list of known exoplanets the huge difference are the M spectral type stars. Of the 69 closest stars, 50 are M spectral type. But of the 71 confirmed exoplanets, only 3 are M type. If I understand this correctly it suggests that we are not likely to find planets around at least 72% of stars. In fact it seems there are only 9 G, F, & K stars out to 16 light-years.


    The four stars that you mentioned are, I believe, no closer than 27 light-years. Even at 0.1c this would be 270 years travel time which, in my opinion, exceeds what the Wait Equation would tolerate. Just knowing that small planets exist nearby would be great, but for an interstellar mission, waiting until the 270 year flight time is reduced might be problematic.

    we should distinguish between life-bearing planets…and terraformable earthlike planets.


I totally agree. I think that the motivation for the first true interstellar mission will be more engineering than science. We will be going with a plan on how we can transform that world rather than just what we can learn about it. What I would add is that I think that paraterraforming is far easier than terraforming resulting in a more liveable home in far less time.


I’m very intrigued by your mention of terraforming. I’ve got to imagine that the purpose would be to create a liveable home for humans. So how would you propose that a human civilization be established there?

  • andy February 5, 2009, 17:29

    If I understand this correctly it suggests that we are not likely to find planets around at least 72% of stars.

    There is a bias against finding planets around M dwarfs: the fact they are so faint makes it difficult to get radial velocities. The situation would be better in the infrared where M dwarfs appear brighter, but the technology for high precision infrared radial velocities is not as well developed as it is for visible light (such techniques would also be useful for finding planets around objects with even cooler spectral types than M, so there is work being done in this area). They therefore aren’t targetted as much as the G and K stars. The surveys that have been done suggest that M dwarfs are much less efficient at forming inner system jovian planets than sunlike stars, which are the easiest planets to detect. Gliese 876, the first M dwarf system found to have planets thus appears to be something of an anomaly.

  • Ronald February 5, 2009, 17:59


    as you are also indicating in the first subparagraph of your most recent post, I also believe that our best chances (by far) are with the solar type stars (late F, all G, early K). Though the statistics are still very slim (namely 1), it is probably no coincidence that of all star types, our own sun is a G star.

    2nd subpar.: to be honest, I was not thinking that practically and I did not consider any flight times, merely the presence of earthlike planets. I actually think that, from an interstellar travel perspective, 20, 30, 40 or 50 ly will not make a lot of difference, principally. When it comes to human survival and the challenge of bridging the gap, I think the issues are very similar for 20 – 50 ly: either we find the means to meet all those challenges or not.

    Terraforming: yes, the added advantages of terraforming are, besides more common occurrence, that an advanced civilization could that way arrange a planet, to some extent, according to its own liking and that there would be no ethical issues with regard to indigenous lifeforms.

    I am not sure that I understand your last question. Wouldn’t human civilization be a logical consequence of human settlement? Or do you refer to certain minimum requirements with regard to (initial) population size, diversity, cultural heritage, etc? Though also of importance, of course, I think the physical challenges of reaching and terraforming such a planet would be most daunting.

  • Stephen February 5, 2009, 23:04

    Is there any chance that COROT-Exo-7b is in fact a hot Jupiter, but the angle of it’s orbit grazes the star – that it’s not a transit but a partial transit? If so, what are the odds? And how could one tell?

    Is there a dimming when the planet is eclipsed by the star?

  • spaceman February 9, 2009, 8:41

    I read Greg’s thoughtful and perceptive post and I was struck by one thing. If, as radial velocity surveys seem to be indicating, approximately 30 percent of solar type stars have super-earths in 0 to 50 day orbits, then why did COROT only find one of these planets after a couple of years of observations?

  • MaDeR February 12, 2009, 17:11

    Stephen, it is possible to recognize apart hot jupiter from super-earth.

    But I noticed that there is no mass given for this world. Maybe it is uncertain.

    spaceman, they are restricted by follow-up observations with more traditional methods. CoRoT is said to have hundreds candidates!

  • spaceman February 15, 2009, 7:45


    Wow, I was not aware of the fact that COROT has so many candidates. However, are most of these hundreds super-Earths or hot-Jupiters?

  • ljk April 12, 2009, 16:40

    The changing phases of extrasolar planet CoRoT-1b

    Authors: Ignas A.G. Snellen, Ernst J.W. de Mooij, Simon Albrecht (Leiden Observatory, Leiden University)

    (Submitted on 8 Apr 2009)

    Abstract: Hot Jupiters are a class of extrasolar planet that orbit their parent stars at very short distances. Due to their close proximity, they are expected to be tidally locked, which can lead to a large temperature difference between their day and nightsides. Infrared observations of eclipsing systems have yielded dayside temperatures for a number of transiting planets. Furthermore the day-night contrast of the transiting extrasolar planet HD 189733b was mapped using infrared observations.

    It is expected that the contrast between the dayside and nightside of hot Jupiters is much higher at visual wavelengths as we move shortward of the peak emission, and could be further enhanced by reflected stellar light. Here we report on the analysis of optical photometric data of the transiting hot Jupiter CoRoT-1b, which cover 36 planetary orbits.

    The nightside hemisphere of the planet is consistent with being entirely black, with the dayside flux dominating the optical phase curve. This means that at optical wavelengths the planet’s phase variation is just as we see it for the interior planets in our own solar system. The data allow only for a small fraction of reflected light, corresponding to a geometric albedo <0.20.

    Comments: 18 pages of PDF, including Suppl. Info; Accepted by Nature (submitted 28 Jan 2009); Under Nature embargo until published

    Subjects: Earth and Planetary Astrophysics (astro-ph.EP)

    Cite as: arXiv:0904.1208v1 [astro-ph.EP]

    Submission history

    From: Ignas Snellen [view email]

    [v1] Wed, 8 Apr 2009 12:08:23 GMT (620kb)


  • ljk September 16, 2009, 12:01

    September 16, 2009

    Smallest Expoplanet Yet Has Rocky Surface

    Written by Nancy Atkinson

    More details are emerging on the extrasolar planet that was discovered by the CoRot satellite back in February. New information about this planet make it first in many respects: It is the smallest known exoplanet, it is the closest exoplanet yet to its host star, which also makes it the fastest; it orbits its star at a speed of more than 750,000 kilometers per hour.

    Plus, data reveal the presence of twin sister planet, another so-called super-Earth called CoRot-7c in this alien solar system. Was Obi-wan wise to conceal it?

    (Sorry, couldn’t resist the twin sister/Star Wars reference….)

    “This is science at its thrilling and amazing best,” says Didier Queloz, leader of the team that made the observations. “We did everything we could to learn what the object discovered by the CoRoT satellite looks like and we found a unique system.”

    Full article here: