≡ Menu

Detecting Centauri Planets

What are the chances that we’ll find habitable planets around Alpha Centauri A and B? Centauri Dreams has long kept an eye on the work of Greg Laughlin (UC-Santa Cruz) and colleagues, who have been working on the Alpha Centauri question with ever more interesting results. Following their work on Greg’s systemic site has been fascinating, and for those who would like to be quickly brought up to speed, it’s useful to know that Laughlin has made their recent paper summarizing these findings available online. Anyone serious about the study of these closest stars to Earth will want to download and read these promising results.

Laughlin’s group simulated the formation of planetary systems around Centauri B, beginning with a disk populated with 400 to 900 lunar-mass protoplanets, then following its development over 200 million years. To say the results are encouraging would be an understatement. All the simulations lead to multiple-planet systems, with at least one planet of Earth mass or slightly larger within the tantalizing range of 0.5 to 1.5 AU. I’m going to poach Greg’s diagram from the latest systemic article by way of summarizing these findings below; please consider this a pointer as well to add systemic to your regular reading list.

Alpha Centauri A and B simulations

Image: Findings from Laughlin’s team on the kinds of planetary systems we should expect around Alpha Centauri B. The team includes UC-SC graduate student Javiera Guedes, Erica Davis, Eugenio Rivera, Elisa Quintana and Debra Fischer. Credit: Greg Laughlin/systemic.

Note the somewhat larger planetary sizes of the larger planet in each run as compared to the Earth (in the top row). Laughlin notes that Centauri B’s higher metallicity should lead to more massive terrestial planets, as the simulation confirms. The question now arises, with orbits of terrestrial worlds now believed to be theoretically possible around both Centauri A and B, and with simulations showing the possibility of a rocky world in the habitable zone of Centauri B, what can we do to get a confirmed detection? It would be one of the most significant finds in exoplanetary science, one that would fire the imagination and immeasurably boost the case for close-up studies of these stars and their possibilities for life.

We’ve detected almost three hundred exoplanets, some of them (via microlensing) thousands of light years from home. Why haven’t we been able to track down planets around these closest of all stars? The answer lies in the limitations of our radial velocity methods, but as Laughlin and team show, the right kind of survey may be able to surmount them.

But first, what work has been done on Centauri planets? In fact, we have excellent observations of these stars dating back over 150 years because of their proximity and sky-dominating visibility, with radial velocity data for both Centauri A and B tabulated since 1904. A 1999 study was able to determine, for example, that the stars are not orbited by any planet with a mass above ten Jupiter masses. Two years later, a new analysis placed tighter limits on Centauri planets: No planets larger than 2.5 Jupiter masses for Centauri A and 3.5 Jupiter masses for Centauri B could exist.

Laughlin’s team now takes the mass of any planet in a circular orbit within 1 to 3 AU down to somewhere in the neighborhood of 0.3 Jupiter masses around Centauri A and 0.5 Jupiter masses around Centauri B. No gas giants around the Centauri stars! And with consistent findings that Earth-mass planets can form within 2.5 AU of these host stars and remain stable for billions of years, the outlook for small planets is quite promising. Bear in mind, too, that 20 percent of all planets discovered thus far have been found in multiple systems, with three around binaries with an orbital separation similar to that between Centauri A and B.

Finding small planets is no easy task, because it’s much easier to identify a large planet with radial velocity methods; the signature of an Earth-mass planet is vanishingly small. But Laughlin’s team believes it is achievable. From the paper:

A successful detection of terrestrial planets orbiting α Cen B can be made within a few years and with the modest investment of resources required to mount a dedicated radial-velocity campaign with a 1-meter class telescope and high-resolution spectrograph. The plan requires three things to go right. First, the terrestrial planets need to have formed, and they need to have maintained dynamical stability over the past 5 Gyr. Second, the radial velocity technique needs to be pushed (via unprecedentedly high cadence) to a degree where planets inducing radial velocity half-amplitudes of order cm s−1 can be discerned. Third, the parent star must have a negligible degree of red noise on the ultra-low frequency range occupied by the terrestrial planets.

That latter is worth commenting on. While both of the primary Centauri stars are thought to be likely candidates for planets, Centauri B is the easier to investigate because compared to its companion star, Centauri B is exceptionally quiet in terms of the kind of oscillations and stellar activity that can produce ‘noise’ in radial velocity data. Studies of Centauri B in the X-ray spectrum indicate low chromospheric activity of the sort that would be associated with only weak stellar flares. If that assumption holds true, we may be able to push radial velocity methods down to the exceedingly fine scales needed to detect planets of this size.

The key word is ‘dedicated,’ the authors assuming the need to access a telescope capable of continuous observations over lengthy periods. It takes time to accumulate radial velocity data, particularly when working at this level of detail, but a detection of one or more Centauri planets within a few years, possibly within the habitable zone, makes this an exciting prospect. The Centauri stars thus become not just valuable targets in and of themselves, but also serve as a testbed for pushing radial velocity techniques to new levels. Imagine the magnitude of this discovery:

Our simulations and those of Quintana et al. (2007) show that for disks with small inclinations relative to the binary orbit, large terrestrial planets tend to form in or near the habitable zone of α Cen B. Thus, it is possible we may detect a habitable terrestrial planet around at least one of our nearest stellar neighbors.

The paper is Guedes et al., “Formation and Detectability of Terrestrial Planets around α Centauri B,” accepted by The Astrophysical Journal. Now available at the arXiv site.

Comments on this entry are closed.

  • Edg Duveyoung February 25, 2008, 12:14

    If there is an Earth class planet around the closest star to us, I think it would enliven the imaginations of folks as much as the canals on Mars did.

    Folks don’t know how vast the distance is to Centauri, nor do they know how much time and money would be needed to build a starship to get there, so most people will just put two and two together and think that any planet in a habitable zone must have life. Voila, canals, and probably a dozen Hollywood productions in the first year.

    Pray for that planet being discovered! It would loosen the wallets of politicians who want to be a latter day Kennedy with a “moon mission” to focus American Pride upon.

    Or pray for a creature to walk past Oppy.


  • Tim February 25, 2008, 13:11

    Hi: Am I reading your figure incorrectly or have you reversed the positions of the “E” and “V” at the top of the figure?

  • Administrator February 25, 2008, 13:23

    Looks fine to me, Tim. I used the same figure that Greg Laughlin did in his post; reading from the left across the top, you should see M, V, E, M — assume the Sun at the far left of the chart and it works out.

  • Hans Bausewein February 25, 2008, 15:01

    Are there no gravitational lense candidates here?

    By combining the parallax due to earth’s orbit and the orbits of A and B there must be a quite large area being covered in the long term.

    If a background star just misses our line of sight we can maybe even use a solar system probe to look.

    Well, maybe still too far away. A few arcsecond or 20 ? Still not very much :(

  • stargazerdude22 February 26, 2008, 5:27

    Detection of an Earth mass planet in the habitable zone around either of the two would indeed fire up the public’s imagination.

  • Edg Duveyoung February 26, 2008, 8:33

    Because Centauri is so “close” and parallax is “huge,” we have to move a telescope a long distance in order to get another “local” star lined up for the 550 telescope, but what about the “fact” that stars are so numerous that “along the way to one’s next observational stance, many other stars get — willy nilly — lined up for observation?”

    The quantity of stars in any given direction may not be so “dense” that Obler’s paradox must be wrestled with, but it sure seems that there’s always more stars found with each advance in telescopic magnification. So, if I’m right, then the time spent traveling to a certain point could be well utilized on other observations of objects more distant or objects of lesser interest, but solid astronomy could be done.

    With the promised magnification of 550 AU, maybe the question to ask is how to block out all the photons of the stars that are “exactly” lined up with whatever target one is pointing at and somehow filter out the possible gravitational lensing that such a “star line into infinity,” might present.

    Question: how do we know that the entirety of the universe is “mostly gravitation lens free?” Maybe every observation is an incredible “mess” of light from many sources unknown such that it is impossible to sleuth out the particulars. The undiscovered-yet math to handle the n-body problem might be laughably easier to comphrehend than the math needed to differentiate a photon arriving from ten light years out from another arriving from 10 billion light years out….given that some photons will be “tired light” and redder while another may be from a nearby, in fact red, star.

    Okay, now my head hurts.

    PS: “It would indeed be remarkable if Nature fortified herself against further advances in knowledge behind the analytical difficulties of the many-body problem.”
    —Max Born, 1960

  • James M. Essig February 26, 2008, 9:11

    Hi Folks;

    Detection of an Earth massed planet in the Centauri system would be awesome. It would be a Centauri Dream… come true. I could imagine a crash course huge effort to build fusion rockets that could get us to perhaps 1/3 to 2/5 C with maximal Isp of 3,000,000. Electrodynamic breaking could be accomplished for fueled wieght reduction, such as reverse magsail, and/or other electrodynamic means of breaking.

    Can you imagine some of our children or grandchildren having kids that would actually set boots on one of these planets. Man, I think we could launch the first mission by 2040 at the latest and get there in 10 years ship time, perhaps even 10 years Earth time. Upon receiving an RF signal that the craft had arrived 4 years later including audiovideo clips of the mission having been sucessfully accomplished, such would make all of us humans stand a little taller. If it was primarilly say a joint U.S./NASA, ESA, Russian, and say Japanese mission, I think even the likes of Osama Bin Laden would have to rethink their entire focus and say, “Job Well Done, Congrats!”.

    It is kind of erie in a way that we could end up being the ETs landing on this planet 4 lightyears away if it exists, even during our lifetime. At some level or another, regardless of our political, religious, and ethnic backgrounds, regardless of our individual personality profiles, occupation, vocations, marital status, etc., all of us would welcome such a crash program with open arms if a Federally mandated funding opportuinity backed by U.N. support was developed for this purpose.

    If we had sucessfullly recieved word that the mission had sucessfully touched down on any terrestrial Centauri system planets, say by 2050, I hope to still be living by then and although I would be 88 years old, I would have to invite all of us here at Tau Zero to an outdoor Turkey Smoker party. I might be walking with a cane, but I think I would find the wherewithall to start the grill and prepare one heck of a party.



  • ljk February 26, 2008, 10:39

    Detecting Planets around Compact Binaries with Gravitational Wave Detectors in Space

    Authors: Naoki Seto

    (Submitted on 22 Feb 2008)

    Abstract: I propose a method to detect planets around compact binaries that are strong sources of gravitational radiation. This approach is to measure gravitational-wave phase modulations induced by the planets, and its prospect is studied with a Fisher matrix analysis.

    I find that, using the Laser Interferometer Space Antenna (LISA), planets can be searched for around ~3000 Galactic double white dwarfs with detection limit ~4 M_J (M_J~ 2×10^{30}g: the Jupiter mass).

    With its follow-on missions, planets with mass ~1 M_J might be detected around double neutron stars even at cosmological distances z~1. In this manner, gravitational wave observation has potential to make interesting contributions to extra-solar planetary science.

    Comments: 4 pages, to appear in ApJL

    Subjects: Astrophysics (astro-ph); General Relativity and Quantum Cosmology (gr-qc)

    Cite as: arXiv:0802.3411v1 [astro-ph]


  • Ronald February 26, 2008, 11:19

    @James Essig: yes, I share the (Centauri) dream! It would be awesome indeed (I will be 88 in 2050, turning 89 by the end of that year, but what the heck, my grandparents lived even much longer, into their high ’90s, and with biomedical progress…).

    I read in the referenced article, that “the planet (i.e. a terrestrial in an Alph Cen B system, Ronald) is detectable even if the radial velocity precision is 3 m/s”, if stellar back-ground noise is minimal. That sounds very promising and within reach.

    Of course the Alph Cen B (continuously) Habitable Zone (CHZ) is nearer to the star that the earthlike situation pictured in the figure above, I guesstimate from about 0.55 – 0.7 AU, possibly up to about 0.9 AU. This makes runs 3 and 4 the best. Run 3 is particularly interesting, because it has a second terrestrial on the outskirts of the habitable zone.

    I wonder which instruments foreseen, if any, preferably even ground-based (such as Keck interferometer, GMT, VLT, ELT) or space-based (SIM, TPF, Kepler) could directly image and spectroscopically analyze such a planet.

  • Ronald February 26, 2008, 13:15

    In the past I had been rather annoyed about the close binary state of Alpha Centauri, fearing it might spoil the fun for any terrestrial planets, but it might actually appear that the two sister stars serve a function to each other’s planetary systems more or less similar to that of Jupiter (provider of cometary water, meteorite vacuum cleaner).

    And I can’t help musing a little bit, about the intelligent inhabitants of Alpha Centauri B b, wondering whether a terrestrial planet orbiting the nearby G2V star can ever have any higher lifeforms, without the blessings and protection of a binary star…

  • yeti February 26, 2008, 14:30

    of course a terrstrial planet in the centauri system would be amazing but im sure another paper which studied the disks around binary systems found good results for tight binaries and also those greater than 50AU.

    Alpha centauri A&B binaries are 11-35AU apart. The worst candidates for planets according to the report. Planets are still possible tho

  • Chester February 26, 2008, 15:55

    This is over a decade old, but it may still be accurate:


  • James M. Essig February 26, 2008, 22:24

    Hi Ronald;

    Thanks for the enthusiastic response to my previous above posting. I am glad to see that we are essentially the same age. My Dad’s father lived to be about 86 and would have lived longer I am sure if he has taken better care of his health. I would like to reach atleast into the high 90s +, a prospect which I think is doable for both of us given medical advancements and anti-aging research.

    Medical based life span enhancement is going to make a lot of difference for very long flight duration to further away star systems even with using only mildy relativistic ship velocities. Once we get the propulsion systems worked out, I see that we would be able to send craft all over the Galaxy perhaps even if only in multigeneration ships with greatly lengthened human generation lifespan. I am looking forward out of curiosity as to what the funding opportuinities will be over the next couple of U.S. executive administrations once we hopefully wind down the spending on the war in Iraq. I hope that with the phase out of operations over there and proposed troop draw-downs, we can finely really get back to more ambitious manned space flight funding.



  • george scaglione February 27, 2008, 9:09

    jim,if we find an earth like planet in the centauri system it would indeed be a dream come true! why people might be interested in trying to go there!! which in turn would be one heck of a shot in the arm for the space program no matter how you look at it !! ps my own father died young as a result of completely neglecting his health.sadly things could have been alot different. anyway thanks and i hope i hear from you soon. george

  • forrest noble February 27, 2008, 22:43

    Hey Jim, George and everybody,

    The Alpha Centauri system would seemingly have lots of planetary possibilities as suggested. Probably plenty of moons also to provide resources for space cities/ colonies.

    Don’t see us going there in this century however.

    your friend forrest

  • ljk February 29, 2008, 7:09

    Planetary Systems in Binaries. I. Dynamical Classification

    Authors: Genya Takeda, Ryosuke Kita, Frederic A. Rasio (Northwestern Univ.)

    (Submitted on 27 Feb 2008)

    Abstract: Many recent observational studies have concluded that planetary systems commonly exist in multiple-star systems. At least ~20% of the known extrasolar planetary systems are associated with one or more stellar companions. The orbits of stellar binaries hosting planetary systems are typically wider than 100 AU and often highly inclined with respect to the planetary orbits. The effect of secular perturbations from such an inclined binary orbit on a coupled system of planets, however, is little understood theoretically.

    In this paper we investigate various dynamical classes of double-planet systems in binaries through numerical integrations and we provide an analytic framework based on secular perturbation theories. Differential nodal precession of the planets is the key property that separates two distinct dynamical classes of multiple planets in binaries: (1) dynamically-rigid systems in which the orbital planes of planets precess in concert as if they were embedded in a rigid disk, and (2) weakly-coupled systems in which the mutual inclination angle between initially coplanar planets grows to large values on secular timescales. In the latter case, the quadrupole perturbation from the outer planet induces additional Kozai cycles and causes the orbital eccentricity of the inner planet to oscillate with large amplitudes.

    The cyclic angular momentum transfer from a stellar companion propagating inward through planets can significantly alter the orbital properties of the inner planet on shorter timescales. This perturbation propagation mechanism may offer important constraints on the presence of additional planets in known single-planet systems in binaries.

    Comments: 14 pages, 15 figures, submitted to ApJ

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0802.4088v1 [astro-ph]

    Submission history

    From: Genya Takeda [view email]

    [v1] Wed, 27 Feb 2008 21:15:52 GMT (1414kb)


  • Tom March 3, 2008, 4:53

    I remember tthere was a study similar to this one a few years back which came to the conclusion that planets would almost certainly not form in binary systems with average orbital separations less than or equal to 50 A.U. Now, however, we know of a couple of instances in which the conclusion of this pessimistic skeptical paper is demolished. Likewise, I predict that many planets will be found in close binary systems and the Ryosuke paper will turn out to be, well, pessimistic and skeptical….

  • Administrator March 3, 2008, 9:41

    Tom, I for one am getting more optimistic than ever about planets in relatively close binary systems. The work emerging on the Centauri stars is particularly exciting in that it may not be long before we’re able to confirm terrestrial planets around Centauri B, a sign that similar systems are also likely to have them.

  • ljk March 7, 2008, 15:36

    Alpha Centauri should harbor detectable, Earth-like planets, according
    to new study by UC Santa Cruz astronomers

    For Immediate Release

    SANTA CRUZ, CA–A rocky planet similar to Earth may be orbiting one of
    our nearest stellar neighbors and could be detected using existing
    techniques, according to a new study led by astronomers at the
    University of California, Santa Cruz.

    The closest stars to our Sun are in the three-star system called Alpha
    Centauri, a popular destination for interstellar travel in works of
    science fiction. UCSC graduate student Javiera Guedes used computer
    simulations of planet formation to show that terrestrial planets are
    likely to have formed around the star Alpha Centauri B and to be
    orbiting in the “habitable zone” where liquid water can exist on the
    planet’s surface. The researchers then showed that such planets could
    be observed using a dedicated telescope.

    “If they exist, we can observe them,” said Guedes, who is the first
    author of a paper describing the new findings. The paper has been
    accepted for publication by the Astrophysical Journal.

    Coauthor Gregory Laughlin, professor of astronomy and astrophysics at
    UCSC, said a number of factors converge to make Alpha Centauri B an
    excellent candidate for finding terrestrial planets. The Doppler
    detection method, which has revealed the majority of the 228 known
    extrasolar planets, measures shifts in the light from a star to detect
    the tiny wobble induced by the gravitational tug of an orbiting
    planet. Factors that favor the use of this technique for Alpha
    Centauri B include the brightness of the star and its position in the
    sky, which gives it a long period of observability each year from the
    Southern Hemisphere, Laughlin said.

    Detecting small, rocky planets the size of Earth is challenging,
    however, because they induce a relatively small wobble in their host
    stars. According to Laughlin, five years of observations using a
    dedicated telescope would be needed to detect an Earth-like planet
    around Alpha Centauri B.

    Coauthor Debra Fischer of San Francisco State University is leading an
    observational program to intensively monitor Alpha Centauri A and B
    using the 1.5-meter telescope at the Cerro Tololo Inter-American
    Observatory in Chile. The researchers hope to detect real planets
    similar to the ones that emerged in the computer simulations.

    “I think the planets are there, and it’s worth a try to have a look,”
    Laughlin said.

    To study planet formation around Alpha Centauri B, the team ran
    repeated computer simulations, evolving the system for the equivalent
    of 200 million years each time. Because of variations in the initial
    conditions, each simulation led to the formation of a different
    planetary system. In every case, however, a system of multiple planets
    evolved with at least one planet about the size of Earth. In many
    cases, the simulated planets had orbits lying within the habitable
    zone of the star.

    In addition to Guedes, Laughlin, and Fischer, the authors of the paper
    include UCSC postdoctoral researcher Eugenio Rivera and graduate
    student Erica Davis, and Elisa Quintana of the SETI Institute. This
    research was supported by NASA and the National Science Foundation.


    Note to reporters: You may contact Guedes at 1-831-459-5722 or
    javiera@ucolick.org; Laughlin at 1-831-459-3208 or
    laughlin@ucolick.org; and Fischer at 1-415-338-1697 or

  • DaMatriX March 8, 2008, 17:56

    The gravitational dynamics of the Alpha Centauri system may allow the existence of a planetary system, but this may not have been the case in te past. For example, Alpha Centauri A and B may have been born much closer to each other and has been drifting away from each other in a later period of time. This may have prohibited planetary formation (?)

  • Ron S March 10, 2008, 20:36

    I had wondered about the same point DaMatriX raises. We know from detailed simulations of our own solar system that orbital eccentricities of the planets can in some cases vary widely even over periods of only several million years. The Guedes paper states that the present orbits of Cen A & B are kept static through the simulation. For the short simulation period constant orbital parameters make sense but it does not tell us anything about long-term stability. With the tight bands on planet formation is this system, I would think that any reduction in the minimum separation of the stars could very well disrupt the system, and even lead to ejection of one or more planets.

    Unfortunately a reliable simulation of system evolution over its 5~6 billion year history may not be possible since any planets would also perturb the orbits of the stars.

    A question for anyone who might know. I did a quick calculation and found that an Earth equivalent at 1 AU distance from its Cen A or B primary would be ~24 magnitude (Earth albedo, 1/2 illuminated) with an apparent maximum separation of 0.76 arcseconds. With masking technology we have today is there any likelihood of a direct detection of such a planet (Cen A or B stellar disk diameter ~7 mas)? I’m guessing the answer is ‘no’ since I haven’t heard it being proposes, but I wonder how close to being possible it would be.

  • Ronald March 11, 2008, 4:56

    Hey DaMatrix (for others: we met recently on a Dutch discussion forum, where I pointed him to this website, a little bit of ‘evangelization’ from my part).

    Can anybody (or DaMatrix himself) elaborate in this? It seems important enough, even crucial. I assume that Laughlin and his team (who have dome a lot of research and modelling on AC) have considered this, but what is really known? What is the normal tendency with regard to orbital development among binaries?

  • Ronald March 11, 2008, 7:01

    Administrator: I don’t know very well where to put this, but I recently noted that there has been a very interesting update on Porto de Mello’s work on ‘Astrobiologically Interesting Stars within 10 parsecs of the Sun’ that you mentioned in 2005;

    It is “A Detailed Catalogue of Astrobiologically Interesting Stars within 20 parsecs of the Sun”, also by Gustavo F. Porto de Mello, 2007. I could only find the abstract and would be interested in more.

  • Administrator March 11, 2008, 7:20

    Ronald, let me look into this one (the de Mello work) and see what I can find. I’ll also check on orbital development in binaries; have never heard of this objection re the planet formation scenario around the Centauri stars before.

  • Ronald March 11, 2008, 12:46

    Ref. to DaMatrix issue: well, we already know that giant planets can have inward migration in the course of their (early) existence, as a result of primordial dust cloud drag etc.

    Could there be something similar for binary stars? When I Googled on combination of ‘orbital decay’ and ‘binary stars’, a lot came up. Unfortunately, I have no time for it myself now, but just a hint.

  • Ronald March 11, 2008, 12:57

    After I sent my previous post concerning orbital decay of binaries, I realized that this is exactly the opposite of the issue raised by DaMatrix, and would pose less of a problem for planets, as long as the two stars are still far enough apart. (It could pose a future problem though, if the orbit keeps decaying).
    Is the opposite, as suggested by DaMatrix, a common phenomenon among binaries: ‘orbital drift’, and to what extent?

  • Administrator March 11, 2008, 15:17

    Re orbital decay around binaries like Centauri A and B, this is from the Guedes, Laughlin et al. paper recently released; note that they are talking about gigayear scales, which seems to preclude overt concern over drastic changes:

    “Several studies of α Cen A and B show that terrestrial planet formation is possible around both stars despite their strong binary interaction (Quintana et al. 2002, 2006, 2007). Results to date consistently indicate that planetary systems with one or more Earth-mass planets can form within 2.5 AU from the host stars and remain stable for gigayear scales. Numerical simulations and stability analyzes of planetesimal disks indicate that material is stable within 3 AU from A/B, as long as the inclination of the disk with respect to the binary is 60◦ (Quintana et al. 2002; Wiegert & Holman 1997). In essence, with regard to the formation process, the companion star plays the perturbative role that the gas giants in our solar system are believed to have played during the formation phases of the Sun’s terrestrial planets. The perturbations allow for the accretion of a large number of planetary embryos…”

  • Ron S March 11, 2008, 20:15

    Paul, the issue that interests me is the stability of the binary pair’s mutual orbits, and that was the point of my earlier comment. An increase of orbital eccentricity would reduce the radius from the primaries within which planet orbits are stable. I believe Ronald raised the same question, although I am not only concerned with orbital decay, rather with long-scale periodicities of orbital eccentricities. If the primaries’ orbits were only decaying, the planets would have formed when the primaries were further separated than they are today.

  • Administrator March 11, 2008, 21:15

    So you’re saying, as I read this Ron, that these long-term periodicities would affect planetary stability, and by extension that the stable orbits that Laughlin’s team and others see around Centauri A and B would only be stable throughout their range assuming a lack of these periodicities. Am I reading this right? If so, I’m not aware of any work explicitly addressing this in recent times but of course the older literature may have something, and I’m also wondering whether Elisa Quintana and Jack Lissauer have maybe looked at it in their recent series on binaries. I’ll check into the latter possibility, which I may have missed.

  • Ron S March 11, 2008, 22:40

    Yes, that’s it exactly, Paul. I am ignorant of long-duration simulations of binary orbits (I certainly don’t work in the field!) but I do know that solar/stellar system models invariably show these types of periodicities. Variations in eccentricities can be quite large, and (while not in our own solar system, apparently) can lead to body ejections. Knowing of that work I thought it a reasonable question to ask of binaries.

    Perhaps I read the paper incorrectly, and they also included the evolution of the primaries’ orbits rather than keep them constant, although that isn’t my reading. From the start of section 3:

    “Our N-body simulations take place in the wide binary regime, with Cen B as the central star and Cen A orbiting with binary semi-major axis aAB = 23.4 AU and eccentricity eAB = 0.52 (Pourbaix et al. 2002). We adopt the stellar masses to be MA = 1.105M⊙ and MB = 0.934M⊙ (Pourbaix et al. 2002) and radii RA = 1.224R⊙ and RB = 0.862R⊙ (Kervella et al. 2003). Due to its low mass and large distance from the AB pair, Proxima Cen is neglected in our simulations.”

    In section 2 they give the initial conditions of the simulation (number of planetoids, inclinations, eccentricities) and that the run is 200 My.

    As I read it they did not test the long-term evolution of the primaries orbits.

  • andy March 12, 2008, 4:01

    There are several things which would change the orbit of Alpha Centauri A/B: mass loss from the stars would tend to increase the semimajor axis, tidal forces would tend to decrease it – however the former is not particularly significant in the main sequence phase, the latter is insignificant for stars as widely spaced as Alpha Centauri A/B. Treating the two stars as a two body problem means you shouldn’t get eccentricity oscillations.

    On the other hand if Proxima is bound and on an orbit highly inclined to the A/B pair, it could drive Kozai oscillations which could cause the eccentricity of the A/B pair to reach high values.

  • ljk March 12, 2008, 8:24

    The Daily Galaxy’s take on exoplanets for AC:


  • ljk March 12, 2008, 14:00

    How to Get to Alpha Centauri

    By Charles Q. Choi

    Special to SPACE.com

    posted: 12 March 2008 – 06:17 am ET

    If the nearest star system, Alpha Centauri, does harbor rocky
    planets similar to Earth as new findings suggest, there exist a
    host of ways to get us there, in theory.

    Sending a person to Alpha Centauri within a human lifetime
    wouldn’t be easy. Alpha Centauri is 4.37 light-years away —
    more than 25.6 trillion miles, or more than 276,000 times the
    distance from the Earth to the sun.

    “Interstellar travel is extremely hard,” said science fiction author
    and NASA physicist Geoffrey Landis.

    But the lure has never been stronger. Scientists last week said
    the Alpha Centauri system has the ingredients for an Earth-like
    planet, and they think they can spot it

    Full article here:


  • ljk April 28, 2008, 12:43

    The Alpha Centauri Binary System: Atmospheric Parameters and Element Abundances

    Authors: G. F. Porto de Mello, W. Lyra, G. R. Keller

    (Submitted on 23 Apr 2008)

    Abstract: The Alpha Centauri binary system, owing to its binarity, proximity and brightness, is a fundamental calibrating object for the theory of stellar structure and evolution. This role, however, is hindered by a considerable disagreement in the published analyses of its atmospheric parameters and abundances.

    We report a detailed spectroscopic analysis of both components of the Alpha Centauri binary system, differentially with respect to the Sun, based on high quality spectra (R = 35 000, S/N > 1000). The atmospheric parameters of the system are found to be Teff = 5820 K, [Fe/H] = +0.24, log g = 4.34 and xi = 1.46 km/s, for Alpha Cen A, and Teff = 5240 K, [Fe/H] = +0.25, log g = 4.44 and xi = 1.28 km/s for Alpha Cen B.

    The parameters were derived from the simultaneous excitation & ionization equilibria of the equivalent widths of Fe I and Fe II lines, by fitting theoretical profiles to the Halpha line and from photometric calibrations, good agreement being reached between the criteria for both stars. We derived the abundances of Na, Mg, Si, Ca, Sc, Ti, V, Cr, Mn, Co, Ni, Cu, Y and Ba, concluding that the abundance pattern of the system is solar but for significant Na, Mn and Ni excesses, and a deficit of Ba.

    An analysis of the position of the two stars in up-to-date theoretical evolutionary diagrams yields masses and ages in good agreement with the dynamical and seismological data. Its abundance pattern can be deemed normal in the context of recent data of metal-rich stars.

    Comments: Astronomy & Astrophysics submitted

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0804.3712v1 [astro-ph]

    Submission history

    From: Gustavo Porto de Mello [view email]

    [v1] Wed, 23 Apr 2008 16:37:05 GMT (168kb)


  • ljk June 5, 2008, 8:16

    Modelling the excitation of acoustic modes in Alpha Cen A

    Authors: R. Samadi, K. Belkacem, M.J. Goupil, M.-A. Dupret, F. Kupka

    (Submitted on 4 Jun 2008)

    Abstract: We infer from different seismic observations the energy supplied per unit of time by turbulent convection to the acoustic modes of Alpha Cen A (HD 128620), a star which is similar but not identical to the Sun. The inferred rates of energy supplied to the modes (i.e. mode excitation rates) are found to be significantly larger than in the Sun.

    They are compared with those computed with an excitation model that includes two sources of driving, the Reynolds stress contribution and the advection of entropy fluctuations. The model also uses a closure model, the Closure Model with Plumes (CMP hereafter), that takes the asymmetry between the up- and down-flows (i.e. the granules and plumes, respectively) into account. Different prescriptions for the eddy-time correlation function are also confronted to observational data.

    Calculations based on a Gaussian eddy-time correlation underestimate excitation rates compared with the values derived from observations for Alpha Cen A. On the other hand, calculations based on a Lorentzian eddy-time correlation lie within the observational error bars. This confirms results obtained in the solar case. With respect to the helioseismic data, those obtained for Alpha Cen A constitute an additional support for our model of excitation. We show that mode masses must be computed taking turbulent pressure into account.

    Finally, we emphasize the need for more accurate seismic measurements in order to discriminate, in the case of Alpha Cen A, between the CMP closure model and the quasi-Normal Approximation as well as to confirm or not the need to include the excitation by the entropy fluctuations.

    Comments: 9 pages, 8 B&W figures, accepted for publication in Astronomy & Astrophysics

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0806.0725v1 [astro-ph]

    Submission history

    From: Reza Samadi Dr [view email]

    [v1] Wed, 4 Jun 2008 14:38:41 GMT (57kb)


  • ljk June 5, 2008, 8:17

    Planet formation in Alpha Centauri A revisited: not so accretion-friendly after all

    Authors: Philippe Thebault, Francesco Marzari, Hans Scholl

    (Submitted on 4 Jun 2008)

    Abstract: We numerically explore planet formation around alpha Cen A by focusing on the crucial planetesimals-to-embryos phase. Our code computes the relative velocity distribution, and thus the accretion vs. fragmentation trend, of planetesimal populations having any given size distribution. This is a critical aspect of planet formation in binaries since the pericenter alignment of planetesimal orbits due to the gravitational perturbations of the companion star and to gas friction strongly depends on size.

    We find that, for the nominal case of a MMSN gas disc, the region beyond 0.5AU from the primary is hostile to planetesimal accretion. In this area, impact velocities between different-size bodies are increased, by the differential orbital phasing, to values too high to allow mutual accretion. For any realistic size distribution for the planetesimal population, this accretion-inhibiting effect is the dominant collision outcome and the accretion process is halted. Results are robust with respect to the profile and density of the gas disc: except for an unrealistic almost gas-free case, the inner accretion safe area never extends beyond 0.75AU.

    We conclude that planet formation is very difficult in the terrestrial region around alpha Cen A, unless it started from fast-formed very large (>30km) planetesimals. Notwithstanding these unlikely initial conditions, the only possible explanation for the presence of planets around 1 AU from the star would be the hypothetical outward migration of planets formed closer to the star or a different orbital configuration in the binary’s early history.

    Our conclusions differ from those of several studies focusing on the later embryos-to-planets stage, confirming that the planetesimals-to-embryos phase is more affected by binary perturbations.

    Comments: accepted for publication in MNRAS (Note: abstract truncated. Full abstract in the pdf file)

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0806.0819v1 [astro-ph]

    Submission history

    From: Philippe Thebault [view email]

    [v1] Wed, 4 Jun 2008 16:57:04 GMT (94kb)