≡ Menu

HD 219134: A Nearby System with Multiple Transits

While we’ve all had our eyes fixed on TRAPPIST-1 (amid the still lingering excitement of the discovery of Proxima Centauri b), news about another stellar neighbor has caused only a faint stir. But what’s happening around HD 219134 (Gliese 892) is noteworthy, and it’s interesting to see that Michaël Gillon (University of Liège – Belgium) has had a hand in it. Gillon, after all, led the work on TRAPPIST-1’s two waves of exoplanet discoveries, culminating in the startling assemblage of seven Earth-sized worlds around the dim ultracool dwarf star.


HD 219134 is an orange K-class star (K3V) in the constellation Cassiopeia, and only about half the distance, at 21.25 light years, as TRAPPIST-1 (about 40 light years out). It was known before the recent Gillon et al. paper in Nature Astronomy that we had a super-Earth, HD 219134 b, in orbit here, which was soon joined by two more super-Earths, a gas giant and, a few months later, another two planets, making for a total of six.

This system characterization was radial velocity work using the HIRES Spectrograph at the Keck I telescope which you can track in A Six-Planet System Orbiting HD 219134 — Steven Vogt was lead author on that paper. I notice that Greg Laughlin, also at UC-Santa Cruz and a co-author on the Vogt paper, has singled out the latest work by Gillon and team on his systemic site. That’s because Gillon adds to the one already known transiting world (HD 219134 b) another transiting planet (HD 219134 c), which gives us the closest transiting exoplanet to Earth yet found.

Image: Exoplanet hunter Michaël Gillon (University of Liège, Belgium).

Both transits are Spitzer detections, and both planets have mass and radius estimates that make a rocky composition possible (4.74 ± 0.19 M and 1.602 ± 0.055 R for HD 219134 b, and 4.36 ± 0.22 M and 1.511 ± 0.047 R for HD 219134 c). Laughlin adds that this system “…very cleanly typifies the most common class of systems detected by the Kepler Mission — multiple-transiting collections of super-Earth sized worlds with orbital periods ranging from days to weeks. Upscaled versions, that is, of the Jovian planet-satellite members of our own solar system.”

Screenshot from 2017-03-10 08-46-47

Image: From the systemic site, the HD 219134 system plotted on a mass-period diagram of exoplanets found through the various methods of detection. Credit: Greg Laughlin.

We can expect to hear a good deal more about HD 219134 in upcoming space-based observations. Unfortunately, the K-class star is too bright for some, though not all, James Webb Space Telescope instruments. JWST may be able to detect atmospheric signatures for both transiting planets if they have extended atmospheres dominated by hydrogen. More compact atmospheres dominated by H2O or CO2 would demand precisions out of reach for JWST because of stellar brightness. The paper adds: “…the precisions are limited not by the photon noise but by the instrumental systematics.”

However, we may not be done with transits in this system. The detection of the HD 219134 c transit increases the probability that planets d and f also transit. From the paper:

Using the formalism of previous work, we compute posterior transit probabilities of 13.1% and 8.1% for planets f and d, respectively, significantly greater than their prior transit probabilities of 2.5 and 1.5%… A transit detection for one or both of these planets would increase further the importance of the system for comparative exoplanetology, and a search for their transits is thus highly desirable. Although such a transit search is probably out of reach of ground-based telescopes, it could be performed again by Spitzer, whose operations have been extended to end-2018, or by the space missions TESS [Transiting Exoplanet Survey Satellite] and CHEOPS [CHaracterising ExOPlanet Satellite], which are both due to launch in 2018.

Nearby transiting planets around a small star make HD 219134 a target we’ll be investigating for a long time to come. We can begin to put constraints on possible atmospheres for the two inner worlds in this system, and also tighten up our figures for planetary mass and radius while gaining valuable insights into the formation of super-Earths. You can see how the high-priority target catalog is growing for future observations, and we can expect our space-based assets to continue to add to it as TESS and CHEOPS come online.

The paper is Gillon et al., “Two massive rocky planets transiting a K-dwarf 6.5 parsecs away,” published online by Nature Astronomy 2 March 2017 (full text).


Comments on this entry are closed.

  • Ashley Baldwin March 10, 2017, 15:13

    I think the other point about this pivotal discovery is that it potentially mirrors the coplanar compact TRAPPIST-1 system where all 7 currently known planets transit . Only potentially on a larger scale . Even if the inner planets do in fact transit as with TRAPPIST-1 , it’s less likely that those further out from this much more luminous star do . However there is a yawning gap between planet e and planet g at 0.37 AU and 2.56 AU respectively that encompasses the whole “habitable zone ” for this stellar neighbour . I’m sure this will now be a target not just for transit spectroscopy like TESS and ChEOPs but also for the pantheon of next generation high resolution RV spectroscopes like ESPRESSO, EXPRES and NN-EXPLORE et al.

  • RobFlores March 10, 2017, 16:10

    Looks to me like there is room for another planet between the two
    with orbital periods of 46 d and 92 days. Do those two super Earth’s shepparding this orbital “grove” put a constrain on how large this hypothetical planet could be ? ( Maybe between .5 RE and RE .75, now that would be extra icing on the Cake. (now with Trappist1 data we can be pretty certain Earth sizes and smaller are not rare)

    • Ashley Baldwin March 10, 2017, 20:40

      Yes. And other methods not withstanding ,the quiescent nature of this stable old K star makes it an ideal target for observation with the ultra sensitive new RV spectrographs coming on line this year. These should allow precision measurements down to the necessary 10cm/s required to locate any neighbouring Earth mass planets . EXPRES in particular is bespoke with the sole purpose of conducting Debra Fischer’s “100 Earth project ” from the 4.3m Discovery telescope .

      • Ronald March 22, 2017, 4:06

        And I understood that ESPRESSO on the VLT will be able to do detections down to that level. BTW, you mentioned NN-EXPLORE, what kind of instrument is that, what will its capacities be with regard to RV?

  • DJ Kaplan March 10, 2017, 17:15

    That’s interesting. At this point, we are looking for gaps between giant planets where there might be earth-sized planets that we can’t detect yet.

    • Ashley Baldwin March 10, 2017, 20:28

      Titus -Bode eat your heart out. I reckon it will end up as the first 7 plus planetary system outside of our own or perhaps larger . It even has at least one gas giant already .

      A bigged up Super Earth TRAPPIST-1 co planar inner system and hopefully some Earth sized planets in and around 0.7 AU . Not absolutely the centre of the conservative hab zone but this is an old system at between 9 and 11 Giga years so tidal locking will have had plenty of time to extend outwards . The smaller mass of the K3 primary ( 0.8 Msun) should help mitigate to a degree though ( bearing in mind at 0.7 AU Venus is showing signs of synchronisation in the solar system at just 4.5 Giga years ) . Whether or not a terrestrial planet of Earth mass or even somewhat larger can maintain vulcanism, tectonics , secondary atmophere production and a meaningful magnetosphere for that duration is still a moot point . Time to find out.

      The system is near enough for direct imaging as well as transit photometry ( and possibly spectroscopy too as alluded in the article ) and RV spectroscopy . The slightly smaller contrast difference between the dimmer/smaller than sun star and close proximity of the system might also make it a high priority target for WFIRST , most especially the gas giant and any outer Super Earths . ( 1e9 plus contrast difference and at best 100 mas inner working angle ) . The WFIRST info does say it has a stretch goal of imaging a few nearby Super Earths . If JWST can indeed avoid sensor saturation then HD219134 is going to have everything thrown at it . An exoplanetologist’s dream. Two in a month !

      The EPIC and/or METIS instrumentation on the E-ELT are also likely to be employed too both for direct imaging as well as the emerging hybrid techniques of high dispersion coronagraphy ( described in an arxiv article earlier this week ) and high dispersion spectroscopy .

      Like TRAPPIST-1 we are going to hear ( and learn ) a lot more about this star and its planets over the next decade. I don’t think we have heard the last “embargoed ” news conference .

      I sense that things are progressing to the next level.

      • Ronald March 22, 2017, 4:02

        Question, also to Paul, based on his quote: “Unfortunately, the K-class star is too bright for some, though not all, JWST instruments (…) More compact atmospheres dominated by H2O or CO2 would demand precisions out of reach for JWST because of stellar brightness”.

        Does this mean that all roughly solar type (G, K) stars are too bright for spectral analysis by the JWST? And that the JWST is mainly suitable for (spectroscopy of) M dwarfs? That would be very regrettable. I presume that JWST could still be used for RV detections on solar type stars.

        • Paul Gilster March 22, 2017, 8:51

          Ronald, this is my understanding based on my reading of the Gillon paper. The star here, a K-class dwarf, is too bright to allow the fine-grained analysis of these atmospheric constituents through transmission spectroscopy.

  • Michael Fidler March 10, 2017, 22:16

    According to Wikipedia this system is 10 to 12.5 billion years old, so let’s say an advanced civilization developed there at sometime in the past. So the huge gap between e and g could be an area where a ring world could have been built or something similar, since the possibility that a large asteroid belt might have been there. I just throwing this out because we are finding so many systems near by, what would an artificial construct look like from our point of view. Would a ring world have long eclipses, would there be no RV from it. Could there be something like small Dyson spheres, like Gerard K. O’Neill space colonies. What would be the Lagrange points or stable orbits in a solar system like this be for artificial structures. This would be a fruitful area of conjecture since we need to theorize what may be found so if it shows up, there is some precedence for what the observation would look like. We could be looking at an artifact or something that is still active. So any ideas?

    • Ashley Baldwin March 11, 2017, 18:14

      I suspect something interesting will be found there , including the “habitable zone” for the system …..

      Some other very promising “mature” nearby systems that can soon be assessed again ,far more precisely , by the next generation of RV spectroscopes and imaging systems , not least Alpha Centauri , Lalande 21185 , Tau Ceti and 40 Eridani . All within 5 parsecs.

      ” TRAPPIST-1 “, fine . Meantime : HD 219134 /HR8832. I ask you ! For a star that is going to feature prominently in the next decade surely it warrants a proper name !? Any ideas ?

      • Ronald March 22, 2017, 9:56

        Talking about nearby ‘mature’ planetary systems of solar type stars, within about 10 parsecs/33 ly also: Epsilon Eridani, Sigma Draconis, 82 Eridani, 61 Virginis, HD 192310 (Gl 785), HD 102365.

        And over 30 other solar type stars for which planets have not been detected yet, possibly because they are still below detection limits of any present instrument.

        What is remarkable about several so far discovered multi-planet systems (around solar type stars), is that, as you mention in your comment above, they often show planetary ´gaps’ that encompass the habitable zone. In other words, in those cases no planets have yet been discovered in those gaps, but there could be small, terrestrial planets that are still below detection limit. A fascinating thought!

        Equally fascinating would be to know if and what predictive laws govern the size/orbit distribution of those planets, now that Titus Bode has been largely refuted as a universal law.

        Good examples of planetary gaps encompassing the HZ for solar type stars (other than the above HD 219134) are:
        – Tau Ceti: between e and f, 0.55 – 1.35 AU.
        – 55 Cancri: between f and d, 0.78 – 5.7 AU (mind: f is just inside and near the inner edge of the HZ, but is a gas giant of about 50 Me).
        – HD 40307: between f and g, 0.25 – 0.60 AU (mind: g is just inside and near the outer edge of the HZ, a gas dwarf of about 7 Me).

        In other cases, there is no planetary gap between planets, but a still empty HZ beyond the already discovered planets (which are usually gas dwarfs and Neptunes) in a compact system. The relevant question is here, whether small terrestrial planets can exist in the (Conservative) HZ outside the larger inner planets. Good examples of this are:
        – 82 Eridani: beyond planet d (5 Me) at 0.35 AU. Inner edge of CHZ at 0.8 AU.
        – Nu2 Lupi: beyond planet d (10 Me) at 0.41 AU. Inner edge of CHZ at 0.95 AU.
        – 61 Virginis: beyond planet d (24 Me) at 0.48 AU. Inner edge of CHZ at 0.85 AU.
        – HD 69830: beyond planet d (18 Me) at 0.63 AU. Inner edge of CHZ at 0.75 AU.

    • Harry R Ray March 12, 2017, 14:09

      Wikipedia ACTUALLY lists it as a SEVEN planet system, but there is a BIG CAVEAT here! The Wikipedia masses for b and c are much smaller than the ones listed above in the posting. ALSO: The seventh planet, h, has a semi-major axis very close to, and an orbital period not much smaller than planet e(i.e; 1842 days and 2.56AU for e and 2247 days and 3.11 AU for h). What’s EVEN MORE SURPRISING is that e’s orbit has an excentricity of 35% whereas h’s orbit is almost circular. How close do these two planets approach each other, AND, since e is MUCH LESS MASSIVE, could than h, could it have ONCE been a MOON of h?

    • Michael Fidler March 13, 2017, 23:57

      What about something like a ring moon? An example of why this may be one of the first major megalithic space projects would be a ring world around the earth at the geostationary orbital period. Right now what is the most valuable real estate off earth, the satellites in geostationary orbit! Building a ring world at this position would allow space elevators from earth to it, without having to stop since they orbit at the same speed as the earth revolves. With a circumference of a 163,000 mile it would only have to be 700 miles wide to have the same area as all of earths land mass ( above ocean level). This is well beyond the Roche limit of earth and with AI I could imagine a ring world similar to the science fiction novel by John Varley called Titan (1979).

      Now what would this look like from earth in another planetary system with transits? Probably very similar to a ring around the planet, but their would be some subtle differences that would show up in the light curve of the transit when the ring world entered and exited from in front of the star. There are several articles on Centauri-Dreams dealing with transits of rings and moons that may be of interest:
      Exomoons: A New Technique for Detection
      Searching for Exoplanet Rings
      Exomoons: A Data Search for the Orbital Sampling Effect and the Scatter Peak
      Exoplanetary Ring Systems and Their Uses
      Ren´e Heller

  • Gideon Marcus March 11, 2017, 11:52

    Thanks for this, Paul. TRAPPIST came up at my local sf convention, and I ended up fielding some questions about it. The next five years will be interesting, to say the least. Hard to believe that 30 years ago, extrasolar planets were purely theoretical…

    • Ashley Baldwin March 11, 2017, 18:02

      Another year of KEPLER, then GAIA, ChEOPs, TESS, SPECULOOS, ESPRESSO, EXPRES , JWST , VISIR/VLT , WFIRST, PLATO, ELTs , ARIEL? HDST?. The next twenty ain’t going to be so bad either…..

  • ljk March 13, 2017, 19:55

    Strange Loner Planet Gets Astronomers’ Attention

    Published: 13 March 2017

    by Matt Williams

    In the hunt for exoplanets, some rather strange discoveries have been made. Beyond our Solar System, astronomers have spotted gas giants and terrestrial planets that appear to be many orders of magnitude larger than what we are used to (aka. “Super-Jupiters” and “Super-Earths”). And in some cases, it has not been entirely clear what our instruments have been detecting.

    For instance, in some cases, astronomers have not been if an exoplanet candidate was a super-Jupiter or a brown dwarf. Not only do these substellar-mass stars fall into the same temperature range as massive gas giants, they also share many of their physical properties. Such was the conundrum addressed by international team of scientists who recently conduced a study of the object known as CFBDSIR 2149-0403.

    Located between 117 and 143 light years from Earth, this mysterious object is what’s called a “free-floating planetary mass object”. It was originally discovered in 2012 by a team of French and Canadian astronomers led by Dr. Phillipe Delorme of the University Grenoble Alpes using the Canada-France Brown Dwarfs Survey – a near infrared sky survey conducted using the Canada-France Hawaii Telescope at Mauna Kea.

    Full article here:


    To quote:

    We now reject our initial hypothesis that CFBDSIR 2149-0403 would be a member of the AB Doradus moving group,” said Delorme. “This removes the most robust age constraint we had. Though determining that certainly improved our knowledge of the object it also made it more difficult to study, by adding age as a free parameter.”

    As for what it is, they narrowed that down to one of two possibilities. Basically, it could be a planetary-mass object with a mass of between 2 and 13 Jupiters that is less than 500 million years in age, or a high metallicity brown dwarf that is between 2 and 40 Jupiter masses and two to three billion years of age. Ultimately, they acknowledge that this uncertainty is due to the fact that our theoretical understanding of cool, low-gravity, and metallicity-enhanced bodies is not be robust enough yet.

  • RobFlores March 14, 2017, 11:38

    Good grief what an old system.

    For these planets, Is there a marker for liquid core and active vulcanism that a can be detected in the atmosphere? (could we detect the effects of a weak magnetic field?)

    It would be useful data to test models on terrestrial planet core
    evolution, at least for S-T type planets

    12 billion years is a long time, even for S-T planet maybe Most
    of their atmosphere’s are depleted and only a few dozen MB are left?