≡ Menu

The Remarkable Kepler-11

Last June Centauri Dreams readers were excited about the release of Kepler results, but miffed that so much of the most interesting material was held back for later release. Now we have the release of these data, and the first thing I want to do is direct you to Greg Laughlin’s systemic site, where you can find a follow-up characterization flow chart to help work through systems of interest. Laughlin calls it a ‘template for the treasure map,’ and it’s available in full here.

What happens next? Laughlin (UC-Santa Cruz) notes the process: “Once the candidates hit the stands, there will be a rush to skim the cream, and a mobilization of follow-up observational campaigns to capitalize on the best opportunities in the data set.” He also reminds us that the brighter the parent star, the better the chances for delving deep into its exoplanetary mysteries. We’re in cream skimming time indeed, and we’ll have plenty to talk about in coming weeks. We’ll carry on tomorrow with thoughts on the 54 new planet candidates found in the habitable zone (five of these are near Earth-sized).

Planetary Bonanza Around Kepler-11

For today, though, I want to focus on the unusual case of Kepler-11, a system that contains no fewer than six planets orbiting a Sun-like star. Ponder this: Before the new discovery, we had determined both the size and mass for only three exoplanets smaller than Neptune. Now, in a single planetary system, we have added five more, all in tightly packed orbits around the primary. There is a sixth planet as well, though its mass remains undetermined.

Image: This diagram compares the orbits of the six planets in the Kepler-11 system with the orbits of Mercury and Venus. Credit: NASA/Tim Pyle.

The five inner planets in this system range from 2.3 to 13.5 times the mass of Earth. All have orbital periods of less than fifty days, meaning they would fit inside the orbit of Mercury within our own Solar System. The sixth planet has an orbital period of 118 days and, as noted above, we have no data on its mass. All six of these worlds have densities lower than the Earth’s. Jonathan Fortney (UC-Santa Cruz) led the work on the structure and composition of the planets:

“It looks like the inner two could be mostly water, with possibly a thin skin of hydrogen-helium gas on top, like mini-Neptunes,” Fortney said. “The ones farther out have densities less than water, which seems to indicate significant hydrogen-helium atmospheres. … These planets are pretty hot because of their close orbits, and the hotter it is the more gravity you need to keep the atmosphere. My students and I are still working on this, but our thoughts are that all these planets probably started with more massive hydrogen-helium atmospheres, and we see the remnants of those atmospheres on the ones farther out. The ones closer in have probably lost most of it.”

Finding six planets around the same star allows us to make advances in the burgeoning study of comparative planetology, and Kepler-11 seems to hold clues to planetary formation. Because stellar disks should lose their hydrogen and helium gas within about five million years, the assumption here is that the numerous small planets with hydrogen/helium atmospheres must have formed quickly. Moreover, their crowded location near the parent star suggests planetary migration, with at least some of the planets forming further out and spiraling inward over time.

New Techniques for Determining Mass

Let’s back out for a moment to the wider picture, because we’re seeing in this work a useful new way to figure out the masses of exoplanets. What Kepler does is to detect the periodic dips in brightness as planets pass in front of their host stars, the size of the dip allowing a determination of the planet’s radius. The planet’s orbital period can be found by measuring the time between transits. Finding a planet’s mass, in the case of Kepler-11, was done by analyzing variations in the orbital periods caused by the gravitational interactions between the planets.

Daniel Fabrycky, a postdoctoral fellow at UC-Santa Cruz, led this analysis:

“The timing of the transits is not perfectly periodic, and that is the signature of the planets gravitationally interacting. By developing a model of the orbital dynamics, we worked out the masses of the planets and verified that the system can be stable on long time scales of millions of years.”

Bear in mind that ground-based Doppler spectroscopy, which is normally used to confirm a transiting planet and determine its mass, could not be used here because of the small size of the planets involved and the distance of the star, some 2000 light years from Earth. Because Kepler is looking for small planets around relatively distant stars, orbital dynamics becomes the method of choice, and we can expect the same method to be used on many of Kepler’s discoveries. Fabrycky noted that the sixth Kepler-11 planet is too far separated from the other five to allow the orbital perturbation method to work. Thus we have no determination of its mass.

Six planets all orbiting in the same plane, showing the conservation of the disk pattern in the system from which they formed, is a scenario reflecting what we see in our own Solar System, though we have no other real comparisons between this system and ours. Kepler-11 is a fascinating find, one of many we’ll have the chance to talk about as Kepler results are sifted and the cream is skimmed. I’m out of the office most of the day, but I’ll have the citation for the Nature paper on this work up as soon as I can.

Now back — the citation is Lissauer et al., “A closely packed system of low-mass, low-density planets transiting Kepler-11,” Nature 470 (03 February 2011), pp. 53–58 (abstract).


Comments on this entry are closed.

  • kzb February 2, 2011, 14:33

    From NASA:

    “The findings increase the number of planet candidates
    identified by Kepler to-date to 1,235. Of these, 68 are approximately
    Earth-size; 288 are super-Earth-size; 662 are Neptune-size; 165 are
    the size of Jupiter and 19 are larger than Jupiter. ”

    I find this interesting, because the peak of the size distribution looks to be super-Earth to Neptune size. Possibly this is attributable to selection bias, in that larger planets are easier to detect. But if it is actually true, it’s a little unexpected to me, because Earth is the largest terrestrial planet in our own system.

    It’s good news though, because larger planets are more likely to hang onto their atmospheres and to remain geologically active over very long time periods. In the far future though, which colonists are going to want to live on a world with say 1.7g gravity ? :)

  • kurt9 February 2, 2011, 15:04

    The Kepler results appear to be in line with the predictions made several years ago, which were for 1,000 planets with 50 of them being Earth-sized. This suggests that the current theories about planets are correct.

    I see the Kepler-11 planets are thought to be more small Neptune like rather than terrestrial like ours. I thought these couldn’t form inside the “snow line”. Maybe they migrated in like the hot Jupiters and Neptunes found before. This planetary migration in towards the parent star seems to be common.

    What is the difference between a water world and a small Neptune, the size of the rocky core compared to the ocean and atmosphere? Perhaps these are more common than rocky planets like ours.

  • andy February 2, 2011, 15:19

    That is a pretty flat planetary system! A strong contrast to the lumbering superjovians at Upsilon Andromedae which are inclined by about 30 degrees, and even in our solar system Mercury has an inclination of over 6 degrees to the invariable plane.

    The fact that so many of these “super-Earths” are turning out to be “mini-Neptunes” is rather worrying for habitability prospects though. The only confirmed rocky exoplanets (CoRoT-7b and Kepler-10b) are close enough to their star to be remnant cores of Uranus-type worlds. Maybe once we start finding planetary systems with architectures more like the inner solar system this will change.

  • Zen Blade February 2, 2011, 16:15

    Cool news.
    I look forward to seeing more.
    Sorry I missed out on the DARPA thread of a few days ago.

  • jumpjack February 2, 2011, 17:54

    What about the 700 candidates planets of June 2010, to be confirmed today? Are 15 planets the only ones confirmed out of 700, or are data still being analyzed?!?

  • Istvan February 2, 2011, 19:11

    Absolutely fascinating from a planetary formation perspective!

    Unfortunately, Kepler findings aren’t going to help much in the quest for nearby targets that could further motivate societal interest and funding, given the implication that all Kepler targets are “relatively distant” stars. I am concerned we’ll need an interesting planetary detection within 10Ly, preferably around one of the nearest stars such as aCen or Barnard’s, to motivate technology projects and get people generally excited. “Interesting” doesn’t necessarily mean “habitable”, mind.

    Even the candidate jovian around eEridani (~10Ly) didn’t make too much of a splash. Its discovery was apparently announced in 2006, but I wasn’t aware of it until quite recently, and I count myself a pretty aware and attentive member of the common citizenry.

    It may unfortunately take a detection around aCen to light a fire and generate interest much beyond scientific circles.

  • kurt9 February 2, 2011, 19:17

    The data set suggests that the most common type of “terrestrial” planet is an ice or gas standard planet. We have only gas and ice giant planets in our solar system. So, until now, we did not know about ice standard and gas standard planets. Now we do, and they may turn out to be much more common than rocky planets.

    Does anyone know anything about the metallicity of the stars involved?

  • Thomas February 2, 2011, 20:58

    Still being analysed. Kepler is not an instant-results mission. Be patient.

    Kepler-11 is quoted as having a metallicity of [Fe/H] = 0.00.

  • Phil February 2, 2011, 21:00

    Its going to be funny if, 20 years down the road, it turns out that the most likely places to search for habitable zones will be on the exomoons of giants in the Hz. I note at least one or two Saturn sized giants discovered by RV last year were in the Hz of their stars.


  • ljk February 2, 2011, 21:04

    We are just scraping the tip of the cosmic iceberg when it comes to finding alien worlds, folks!

    Here may be another reason we have yet to hear or get a visit from any ETI: There are so many other places in the galaxy to visit.

  • Will February 2, 2011, 21:20

    Looking at the archive data sets…so the delay from data collection to public release actually increases as the mission continues? wtf? shouldn’t the pipeline get more efficient with time?

    Actual Start Time Actual End Time Release Date
    2009-05-13 00:15:49 2009-06-15 11:32:57 2010-06-15 00:00:00
    2009-06-20 00:25:09 2009-09-16 23:09:29 2011-02-02 06:00:00
    2009-09-18 17:19:58 2009-12-16 23:55:06 2012-06-18 21:43:41
    2009-12-19 21:03:56 2010-03-19 16:53:31 2012-06-18 21:43:41
    2010-03-20 23:47:15 2010-06-23 15:50:26 2013-06-18 21:43:41

  • mark February 2, 2011, 23:09

    I have a layman’s question: How “far away” from the Big Bang is Kepler 11 — i.e. in terms of time and in relation to the beginning of the universe how old is this system in comparison to ours. Are we the same age, older, younger?


  • William Truderung February 2, 2011, 23:55

    The most interesting multi-planet system I see in the data has to be KOI-730. It apparently contains the first known Trojan planet pair, with the four detected objects in an orbital period ration of 6:4:4:3.

    According to “Architecture and Dynamics of Kepler’s Candidate Multiple Transiting Planet Systems”, we find that “In Kepler data, the two co-orbital candidates began separated by ~118 degrees, with the trailing candidate reducing the gap at the rate of ~1 degree per month.”


  • torque_xtr February 3, 2011, 6:24

    I guess some of the planets are oxygen gas giants resulted from a neptunes with all their water photodissociated and hydrogen blown away… take it a little farther and that’s Bezpin! though it looks much more complicated after comparing their densities, escape velocities and possible thermosphere temperatures, it might be He+some O2 for outers…

  • Eric February 3, 2011, 11:46

    So many planets dynamically stable around a mid-size parent star… Hmm…. Maybe it’s astro-engineering.

    Actually, I’m only 99% kidding.

    How do you get so many largish planets so close into a G class star? It’ll be interesting to model how this scenario can unfold, and how likely, without a little help from little green men.

  • Adam February 3, 2011, 19:17

    Hi All

    Just adding to Eric’s suspicion of astro-engineering, I do wonder at the transit candidates twice the size of Jupiter.

  • Mike February 3, 2011, 20:11

    They must be little green heat loving men, Eric.

  • yeti101 February 4, 2011, 18:11

    Mark, according to the wiki page kepler 11 is 8 billion years old, thats 3 billion years older than us.

  • ljk July 3, 2012, 9:22

    Piano sonata in the key of Kepler-11

    By Phil Plait from his Bad Astronomy blog – July 1, 2012

    Via reddit (if you’re a redditor, go there and upboat!) I found a very interesting use of astronomical data in music. The composer [Update: Astronomer Alex Parker created this!] took the orbital information from the six-planet system called Kepler 11 and codified it into musical notes! From the YouTube notes.

    Here, I’ve taken each transit seen by the observatory and assigned a pitch and volume to it. The pitch (note) is determined by the planet’s distance from its star (closer=higher), and they are drawn from a minor 11 chord. The volume is determined by the size of the planet (larger=louder).

    The result is actually quite listenable!

    Full article here: