≡ Menu

Exoplanet Prediction Confirmed

I’m late getting to this one, because I wanted to get Mike Gruntman’s paper on interstellar instrumentation finished. But for exoplanet enthusiasts like myself, the best news to come out of the recent American Astronomical Society meeting may have been the announcement of a new planet around the star HD 74156. So let’s talk about it, an interesting find because we haven’t had a new planet turn up just where predicted since Urbain Le Verrier and John C. Adams (independently) worked out the existence of Neptune by noting its effects on the motion of Uranus. Thus were calculations turned into observations and thence discovery.

Rory Barnes (University of Arizona) has been working on a theory that led to the HD 74156 discovery for some time. His computer simulations (begun with Thomas Quinn while both were at the University of Washington) on the stability of extrasolar planetary systems showed a key fact: All systems whose planets were close enough to affect each other gravitationally were on the edge of instability. All it would take would be a slight change in orbits to lead to disruptions. Astrophysicist Steven Soter explained the implications in an article last summer:

This remarkable result might seem surprising. But the prevalence of such marginally stable systems makes sense, Barnes and Quinn concluded, if planets form within unstable systems that become more stable by ejecting massive bodies. The investigators remarked, “As unsettling as it may be, it seems that a large fraction of planetary systems, including our own, lie dangerously close to instability.”

Mature planetary systems hold about as many planets as they can, spaced as closely as is consistent with stability. Think of an evolving planetary system as one that ejects disruptive elements before settling into its mature stage, a process of self-organization. If that hypothesis works, it becomes a useful predictor, as Barnes, Quinn and Sean Raymond (University of Colorado) saw. HD 74156’s two known planets were a test case, because the gap between them suggested the presence of a third. It took observations by Jacob Bean and team at the University of Texas to observe the system and make the actual discovery.

The question is, are the processes that proved so successful at HD 74156 likely to be universal? If so, we have a useful predictor of planets around other stars, helping us probe more deeply into systems we’ve already begun to study. Now comes word that another team has found a planet where the ‘packed planetary system’ theory suggested it would be around the star 55 Cancri. Thank you Debra Fischer and team.

55 Cancri f turns out to exist at the inner region of a large stable zone, suggesting the possibility of still further planets in this most interesting system. More on this from the 55 Cnc f discovery paper:

This fifth planet apparently resides in the previously identified gap between 0.24-5.8 AU, and it remains between 0.73 AU (periastron) and 0.84 AU (apastron), preventing orbit crossings with both the next inner planet, “c”, whose apastron is at 0.26 AU and the outer planet, “d”, whose periastron is at 5.5 AU, ensuring dynamical stability that is demonstrated numerically by N-body simulations. As the star’s luminosity is L = 0.60 L (from its effective temperature and radius), this fifth planet resides within the classical habitable zone. With a minimum mass of 45 MEarth, we speculate that it contains a substantial amount of hydrogen and helium, not unlike Saturn (M = 95 MEarth ) in the solar system.

The kind of calculations Barnes and team have employed also call for a planet in a stable region around HD 38529, although that one awaits confirmation. As we continue to explore the interesting hypothesis of ‘packed’ planetary systems, it begins to appear that efficiency is the watchword. If there is room for a planet to form without destabilizing gravitational effects, it does. And that tells us something about the ubiquity of celestial real estate.

The discovery paper is Bean et al., “Detection of a Third Planet in the HD 74156 System Using the Hobby-Eberly Telescope,” accepted by The Astrophysical Journal and available online. The 55 Cancri paper is Fischer et al., “Five Planets Orbiting 55 Cancri,” also accepted by The Astrophysical Journal and available here. Thanks to Dave Moore for helpful background links as this story developed, and to other Centauri Dreams readers who passed the story along and asked for comment.

Addendum: Be sure to check andy’s comment re a paper that may cast HD 74156 d into doubt.

Comments on this entry are closed.

  • andy January 15, 2008, 14:49

    Might be wise to be cautious on the HD 74156 discovery – according to this paper:

    It is interesting to consider planet candidates having orbital periods consistent with one year. There are about fifteen such planets. The masses of all of them but one are rather large (several Jupiter masses), producing large RV semi-amplitudes ~100 m/s. However, the planet HD74156 d announced recently at the base of HET observations (Bean et al. 2007) has relatively small mass and produces the RV oscillation of ~10 m/s only. Given similar amplitude of the annual periodicity in the HET data for the star 55 Cnc, the discovery of HD74156 d looks suspicious. More careful analysis shows that RV data for HD74156 from ELODIE (Naef et al. 2004) also show an annual variation of ~20 m/s, but in opposite phase. This inspires strong doubts about the existence of the planet HD74156 d. This planet candidate could quite be a false detection made due to annual errors in RV data from HET. In any case, its orbital parameters may be strongly distorted and are uncertain.

    (emphasis mine)

    Gliese 777 (HD 190360) might be another system which could host additional planets: there is a hot Neptune at 0.128 AU, and an eccentric superjupiter at 3.99 AU (periastron 2.55 AU) – there could conceivably be another planet in the gap.

  • Administrator January 15, 2008, 15:25

    Interesting! Thanks, andy, I’ve inserted an addendum in the original post.

  • Christopher L. Bennett January 15, 2008, 18:35

    Hmm… this theory sounds like sort of a modern-day Bode’s Law, but one based on physics rather than observation alone. Could this be an explanation for Bode’s Law? Can Bode’s Law be derived from it? I doubt it — I’d think that if the spacing suggested by the theory is dependent on the gravitational interaction of the planets, then the spacing of planets in our system probably has more to do with their respective masses, and thus it wouldn’t necessarily come out the same way in other systems. (And then there’s the issue of planetary migration altering the order of the planets after their formation.) Still, it’s an interesting question.

  • Ronald January 16, 2008, 3:30

    Fascinating indeed;

    I just wonder how this relates to metallicity and planetary abundance, since these two parameters are also indicated to be strongly correllated.

    Is it perhaps possible that the principle of being “filled to the brim” (or packed with planets) is mainly true beyond a certain metallicity threshold, i.e. for very high metallicity stars, whereas metal-poor stars may have much scarcer (and/or smaller) planets?

  • Starfleet commander January 16, 2008, 6:42

    Yes i was thinking it reminds me of Bode’s law as well. A guy on Astronomy today was waxing lyrical about it.

    Still I dont see why there would’nt be some sort of equilibria pattern to planet formation and available gravitational slots around stars. We see laws of nature acting on most things. Why not planetary systems?

    I think the coolest thing about this observation/prediction is that it may mean we can predict whether a star has a slot within the habitability range. Perhaps we can now estimate a whole planetary system once we find the first planet around that star.

    Has anyone tried taking Rory Barne’s theory and extrapolating it across all systems known to contain at least 1 planet? If his formulae is even mostly correct then it would save alot of time.

    Also i wondered if anyone knew what ratio of discovered extrasolar planets were found through planets passing across their sun (hence we see their elliptical plane in front of that star) compared to planets found because of a wobble in the star?

  • Ronald January 16, 2008, 12:40

    @Starfleet commander:

    for your last question, see The Extrasolar Planets Encyclopaedia, http://exoplanet.eu/, or http://www.obspm.fr/encycl/encycl.html

    With regard to the rest: I agree it would be a nice means for predicting and selecting prime targets, but all the same nothing beats scientific observation and discovery.

  • andy January 17, 2008, 6:05

    Another interesting point about the HD 74156 detection (assuming that the planet exists): HD 74156d provides an interesting contrast with our solar system’s asteroid belt: the asteroid belt of our solar system is stable for planets up to 5 Earth masses, but is a region where planet formation would be inhibited. Similarly, the gap between HD 74156b and c is a stable region where planet formation would be inhibited, except in this case it is actually occupied by a planet. (See Raymond 2008, “Terrestrial Planet Formation in Extra-Solar Planetary Systems”)

  • James M. Essig January 19, 2008, 1:27

    Hi Folks;

    The finding of all of these extrasolar planets never ceases to amaze me. We are gradually constucting a space map of our neck of the Milky Way Galaxy which will no doubt be useful for future manned interstellar missions including those in attempts to find and/or contact any existent ETI civilizations. The more we learn about extraterrestrial sources of material, the more we will have available for constucting extrasolar settlements and for use as fusion fuel for fusion powered interstellar space craft. I think this is really cool.


    Your Friend Jim

  • Administrator January 19, 2008, 16:13

    The mapping process is going to be fascinating to watch, particularly as we begin to fill in the terrestrial-size worlds. Right now we’re like early mapmakers sketching out the vague shape of coastlines. Exoplanet prediction methods may be valuable in helping us know where to look in particular planetary systems, but we’ll have to see how well they continue to be corroborated.

  • forrest noble January 20, 2008, 21:25

    This technology, and its continuous improvements, will probably provide continuous discoveries and entertainment for mankind for a seemly endless period of time. If we did find what seemed to be a perfect candidate it would take a long time to do anything about it.

    Maybe some of those ET’s in the same locations that we are looking are also looking back trying to locate us, not likely but an intriguing thought anyway.

    your friend forrest

  • andy January 21, 2008, 11:09

    Right now we’re like early mapmakers sketching out the vague shape of coastlines.

    That makes me wonder what the exoplanetary analogy of expecting Terra Australis when in fact all that exists is scattered islands would be…

  • ljk January 23, 2008, 15:54

    Thansk to the Bad Astronomer, I learned that a number of the
    presentations from the 211th meeting of the American Astronomical
    Society (AAS) held in Austin, Texas earlier this month are now
    online here:


  • ljk April 30, 2008, 11:43

    The Successful Prediction of the Extrasolar Planet HD 74156 d

    Authors: Rory Barnes (LPL, U. of Arizona), Krzysztof Gozdziewski (Torun Centre for Astronomy, N. Copernicus University), Sean N. Raymond (CASA, U. of Colorado)

    (Submitted on 28 Apr 2008)

    Abstract: Most of the first-discovered extrasolar multi-planet systems were found to lie close to dynamically unstable configurations. However a few observed multi-planet systems (e.g. HD 74156) did not show this trait. Those systems could share this property if they contain an additional planet in between those that are known. Previous investigations identified the properties of hypothetical planets that would place these systems near instability.

    The hypothetical planet in HD 74156 was expected to have a mass about equal to that of Saturn, a semi-major axis between 0.9 and 1.4 AU, and an eccentricity less than 0.2. HD 74156 d, a planet with a mass of 1.3 Saturn masses at 1.04 AU with an eccentricity of 0.25, was recently announced.

    We have reanalyzed all published data on this system in order to place tighter constraints on the properties of the new planet. We find two possible orbits for this planet, one close to that already identified and another (with a slightly better fit to the data) at ~0.89 AU.

    We also review the current status of other planet predictions, discuss the observed single planet systems, and suggest other systems which may contain planets in between those that are already known. The confirmation of the existence of HD 74156 d suggests that planet formation is an efficient process, and planetary systems should typically contain many planets.

    Comments: 11 pages, 2 figures, 1 table. Accepted for publication in ApJ Letters. A version with full resolution figures is available at this http URL

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0804.4496v1 [astro-ph]

    Submission history

    From: Rory Barnes [view email]

    [v1] Mon, 28 Apr 2008 22:24:42 GMT (90kb)


  • ljk November 7, 2008, 0:12

    55 Cancri: A Laboratory for Testing Numerous Conjectures about Planet Formation

    Authors: Dimitris M. Christodoulou, Demosthenes Kazanas

    (Submitted on 6 Nov 2008)

    Abstract: Five planets are presently believed to orbit the primary star of 55 Cnc, but there exists a large 5 AU gap in their distribution between the two outermost planets. This gap has attracted considerable interest because it may contain one or more lower–mass planets whose existence is not contradicted by long-term orbit stability analyses, in fact it is expected according to the “packed planetary systems” hypothesis and an empirical Titius–Bode relation recently proposed for 55 Cnc. Furthermore, the second largest planet is just the second farthest and it orbits very close to the star.

    Its orbit, the most circular of all, appears to be nearly but not quite commensurable with the orbit of the third planet, casting doubt that any migration or resonant capture of the inner planets has ever occurred and lending support to the idea of “in–situ” giant planet formation by the process of core accretion.

    All of the above ideas will be tested in the coming years in this natural laboratory as more observations will become available. This opportunity presents itself in conjunction with a physical model that relates the orbits of the observed planets to the structure of the original protoplanetary disk that harbored their formation at the early stages of protostellar collapse.

    Using only the 5 observed planets of 55 Cnc, this model predicts that the surface density profile of its protoplanetary disk varied with distance $R$ precisely as $\Sigma (R)\propto R^{-3/2}$, as was also found for the minimum–mass solar nebula.

    Despite this similarity, the disk of 55 Cnc was smaller, heavier, and less rotationally supported than the solar nebula, so this system represents a different mode of multi–planet formation compared to our own solar system.

    Comments: The physical model applied to the solar system in 0706.3205 (updated version 2) is now applied to the 5 planets of 55 Cancri. 5 tables, 2 figures

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0811.0868v1 [astro-ph]

    Submission history

    From: Dimitris Christodoulou [view email]

    [v1] Thu, 6 Nov 2008 04:55:38 GMT (47kb)