≡ Menu

Stars Passing Close to the Sun

Every time I mention stellar distances I’m forced to remind myself that the cosmos is anything but static. Barnard’s Star, for instance, is roughly six light years away, a red dwarf that was the target of the original Daedalus starship designers back in the 1970s. But that distance is changing. If we were a species with a longer lifetime, we could wait about eight thousand years, at which time Barnard’s Star would close to less than four light years. No star shows a larger proper motion relative to the Solar System than this one, which is approaching at about 140 kilometers per second.

The Alpha Centauri stars are the touchstone for close mission targets, but here again we could make our journey shorter with a little patience. In 28,000 years, having moved into the constellation Hydra, these stars will have closed to less than 3 light years from the Sun. Some time back, Erik Anderson discussed star motion in his highly readable Vistas of Many Worlds (Ashland Astronomy Studio, 2012), where I learned that the star Gliese 710, currently 64 light years out in the constellation Serpens, is headed squarely in our direction. Wait around for 1.3 million years or so and Gl 710 will push right through the Oort Cloud, with who knows what results in the inner system. A new paper considers these matters and tunes up the numbers on stellar encounters.

comet_outer_system

Image: Could a passing star dislodge comets from otherwise stable orbits so that they enter the inner system? Credit: NASA/JPL-Caltech).

A close pass from a star is bound to cause effects elsewhere in the Solar System, as Coryn Bailer-Jones (Max Planck Institute for Astronomy, Heidelberg) notes in his latest paper. Such an encounter can disrupt cometary orbits in the Oort, sending them into the inner system. Earth’s catalog of impact craters, which contains almost 200 known craters and doubtless should include many awaiting discovery, some of them beneath the oceans, is a reminder of what can happen. Nor should we forget that if we really drew the wild card, a close star turning supernova could have disastrous effects on surface life. So how many stars are problematic?

Bailer-Jones identifies the key candidates in this paper, assuming an Oort Cloud that extends to about 0.5 parsecs (1.6 light years), but he notes that a star passing even as close as several parsecs could produce significant cometary disruptions if the star were massive and slow enough. The author worked with 50,000 stars from the Hipparcos astrometric catalog in hopes of fine-tuning earlier studies of passing stars, but he notes that the search can’t be considered complete because radial velocities are not available for all stars and many are fainter than the Hipparcos work could detect. Further analysis will be needed using upcoming Gaia data.

But studying stars within a few tens of light years from the Solar System, Bailer-Jones finds forty that at some point were or will be within 6.4 light years of the Sun — the timeframe here extends from 20 million years in the past to 20 million years in the future. Fourteen stars, in fact, come within 3 light years of the Sun, with the closest encounter being with HIP 85605, which is currently about 16 light years away in the constellation of Hercules. The paper cites “…a 90% probability of [the star] coming between 0.04 and 0.20 pc” somewhere between 240,000 and 470,000 years from now, but Bailer Jones notes that this encounter has to be treated with caution because the astrometry may be incorrect. Future Gaia data should resolve this.

If HIP 85605 were to close to 0.04 parsecs of the Sun, it would be .13 light years out, or roughly 8200 AU, a close pass indeed. But one thing to keep in mind: Oort Cloud perturbation is not an unusual phenomenon, and the situation we are dealing with today is partially the result of encounters with stars that have occurred in the past. We have no data on the time between stellar encounters like these and the subsequent entry of comets into the inner system, making it all but impossible to link a specific passing star with a rise in the rate of Earth impacts. Bailer-Jones discusses all this on his website at the MPIA, where he notes the following:

A close encountering star is likely to perturb the Oort cloud sufficiently to increase the flux of comets entering the inner solar system. Let’s not forget, however, that this kind of perturbation is happening all the time due to the gravitational effect of the Galaxy as whole, and due to stars which [were] encountered even earlier. That is, there is a “background” of comets entering the inner solar system which we cannot necessarily associate with a particular stellar encounter. This is also because the time between an encounter and the time that comets enter the inner solar system could be many or even many tens of millions of years, much longer that than the typical time between close encounters.

Gl 710 is generally cited as the star making the closest encounter in previous studies, and Bailer-Jones sees a 90 percent probability that it passes within 0.10 to 0.44 parsecs, meaning an Oort Cloud passage in 1.3 million years. Looking into the past, the star gamma Microscopii, a G6 giant, encountered the Sun 3.8 million years ago, probably the most massive encounter within one parsec or less. Some encounters are recent: Tiny Van Maanen’s star, a white dwarf, passed near our Sun as recently as 15,000 years ago. While data from the Gaia mission will help us improve the parameters of this catalog of passing stars, Bailer-Jones believes the Gaia results will also make it possible to investigate the link between stellar encounters and impacts in a broad, statistical sense, helping us better understand the history of Earth impacts.

The paper is Bailer-Jones, “Close Encounters of the Stellar Kind,” accepted at Astronomy & Astrophysics (preprint).

tzf_img_post

Comments on this entry are closed.

  • David A. Czuba January 2, 2015, 15:40

    I’m curious. Since we can rewind to investigate past close encounters between our Sol and other star systems, what correlation there is with disturbances recorded in, say, the lunar impact record or mass extinctions on Earth. Is anyone able to find a study on this?

  • Alex Tolley January 2, 2015, 17:15

    As the 2 suns get close, won’t their respective Oort clouds result in swapping bodies? If the isotopic ratios differ between the bodies, this could provide a clue as to the numbers of close encounters.

    Elsewhere we have a suggestion that the moon might contain a record of comet impacts, so that we could use the isotopic ratios of these impact residues as a sample of the different objects in the Oort, indicating the possible number of previous close encounters, although not their precise timing.

    For very close encounters, the approaching star’s Oort cloud would enter the main solar system and presumably major planets like Jupiter would be having major gravitational effects on the Oort bodies, scattering them. Could that have a more immediate timing on impacts on inner system bodies like the moon?

    This is obviously a very distant project

  • Lionel January 2, 2015, 17:38

    Very surprising about visitoring stars closer than 6.4 light years – are there really just 40 within 4×10^7 years? (One every 1,000,000 years)

    The reason my surprise is that we seem to be at the beginning of a much busier 50,000 year window – see this wikicommons graph http://upload.wikimedia.org/wikipedia/commons/e/e9/Near-stars-past-future-en.svg

    Over just the next 50,000 years 5 star systems (and a larger no of stars) come within 5 light years, and most at their closest pass come nearly as close at 3ly

    So seems we are at an unusually busy period of perturbations – two orders of magnitude busier than the long term average suggested in the paper???

  • d.m.falk January 2, 2015, 19:58

    This could possibly offer a suggestion of an answer to the water problem of Comet 67P! It may, in fact, be a comet from a previous passing star system, which would make the Rosetta/Philae mission even more interesting: the first human-built craft to orbit and land, respectively, on an interstellar object! :)

    d.m.f.

  • Andrew Palfreyman January 2, 2015, 22:38

    Looking at stars that are receding directly from us would enable us to attempt a correlation with the geologic record. I am not aware of any such study.

  • Tom Baty January 2, 2015, 23:29

    This discussion makes me wonder about some of the free floating planets that are suppose to be out there unattached to a star. If they only approached within a large part of a light year..no problem. But if they were unobserved as non illuminated objects with no inherit light source and were Jupiter size or larger, could they be a problem?

  • Eniac January 3, 2015, 0:26

    @d.m.falk: Water problem? What happened was that the hypothesis of the cometary origin of Earth’s water was dealt a blow. This is a problem only for those who are irrationally attached to this hythesis, which was a bit shaky to begin with. For the rest of us it is new information. We learned something, and I would not call that a problem.

  • Eniac January 3, 2015, 0:35

    What I find fascinating about the realization that the galaxy is in flux is its consequence for the spread of starfaring civilizations. Such civilizations will not spread steadily with a single expanding front like bacteria in a dish, but fractally like milk in a coffee cup that is being stirred. After a few galactic revolutions, the combination of relative star motion (the stirring) and star hopping colonization (diffusion) will result in an evenly seeded galaxy, with unsettled systems becoming rare much more quickly than the model of steady expansion into a static galaxy would lead you to believe, erroneously.

  • Adam January 3, 2015, 1:15

    I wonder what an on-coming relativistic star from the Galactic Core would look like?

  • Mark Zambelli January 3, 2015, 6:39

    I’m wondering about the possible range in masses of extra-solar Oort clouds. Are all stars likely to have Oort clouds regardless of whether they host planets? With recent data showing the range in density of extra-zodiacal light in some young sytems can this correlate to the amount of mass available there to form comets (or does the increase in dust imply fewer surviving comets)?
    Perhaps our Oort cloud is a fair approximation for the ‘average’ but I can imagine extra-solar-Oort clouds that are larger in extent, density and mass than ours and also clouds that are sparsely populated; either through differences in initial conditions or as a function of age.
    It looks like a mighty task to try and unravel the details as they pertain to our bombardment history, especially when stellar encounters occur on quicker timescales than the journey from the Oort cloud to the inner system.

  • Anthony Mugan January 3, 2015, 8:38

    Imagine how the challenge of motivating (funding) deep space developments might be much easier for a civilisation with a very close stellar neighbour. Wide binary systems of broadly sun like stars ( longevity and stability) might be particularly interesting in this regard.
    In such a scenario the companion star would be very prominent and planets detected at a much earlier stage of technological development.

    From another perspective these close passes are important from a panspermia perspective.

  • Michael January 3, 2015, 14:57

    With the time that we have had going around the Galaxy it is a wonder that we have any Oort cloud left.

    @Adam January 3, 2015 at 1:15

    ‘I wonder what an on-coming relativistic star from the Galactic Core would look like?’

    The closer it gets to light speed the less we will see it in visible light, fast enough and we won’t see it at all.

  • ljk January 3, 2015, 19:43

    Here is the Centauri Dreams article on Erik Anderson’s book Vistas of Many Worlds from November of 2012:

    https://www.centauri-dreams.org/?p=25719

  • kamal ali January 4, 2015, 2:53

    “If HIP 85605 were to close to 0.04 parsecs of the Sun, it would be .13 light years out, or roughly 8200 AU”: can anyone tell me whether a ~8000 AU pass by 1 solar mass star (or even whatever HIP 85606 is) would have an effect on our planetary orbits? How many AU close would a 1 solar mass star need to get to to disturb the orbit of Neptune, for example? Any pointers to dynamical studies would be appreciated.

  • Lionel January 4, 2015, 6:51

    @Michael

    ‘With the time that we have had going around the Galaxy it is a wonder that we have any Oort cloud left.’

    I was having a similar thought – during the 4.6 billion odd years the solar system has been about there will have been ever so many close encounters.

    I’d like to know what the statistical models tell us about the number of sub-light year and sub 0.5 light year encounters over the lifetime of our solar system. Just pondering the geometry – presumably the frequency of these encounters is more or less the cube (or should it be the square?) of the closest pass distance, i.e. 1ly encounters are roughly 8 times less frequent than 2ly encounters and so on

  • steven rappolee January 4, 2015, 18:48

    Back on November 19th I was musing whether or not such encounters would lead to lower energy starship trajectories, Adam tells me most likely not.

    http://yellowdragonblog.com/2014/12/19/rappolee-interstellar-low-energy-transfer-trajectories/

    Bob Zubrin wrote that these close encounters spread life throughout the galaxy

  • Al Jackson January 4, 2015, 22:44

    There is a paper:

    Limits on the closest encounter with any other star since the formation of our solar system, Morris, Donald E.; O’Neill, Thomas G. , Astronomical Journal, vol. 96, Sept. 1988, p. 1127-1135

    Which discuses stellar encounters , over the approximate age of the solar system, encounters that produce changes in orbital inclination and eccentricity of the planets by amounts we don’t see. No object .1 Solar Mass has passed through the solar system in that time, and no object ~ 3 Jovian masses has passed within the Earth’s orbit.
    There are more sensitive tests involving the stability of the solar system done by Jacques Laskar. These simulations run for hundreds of millions of year with stellar encounters included , but I don’t know what the limits were. The mean motion resonances in the solar system are sensitive to perturbations.
    From Extrasolar planet and Kepler surveys there is an initial estimate that there are about 100,000 Nomad planets per main sequence star cruising the galaxy. I have not seen , yet, an estimate of the frequency of solar system encounters with these.
    In Jame’s Blish’s Cities in Flight there is an interstellar ‘nomad’ planet “He” which becomes important to the story plot. Blish wrote this way back in the 1950s!

  • Michael January 6, 2015, 4:00

    @kamal ali January 4, 2015 at 2:53

    “If HIP 85605 were to close to 0.04 parsecs of the Sun, it would be .13 light years out, or roughly 8200 AU”: can anyone tell me whether a ~8000 AU pass by 1 solar mass star (or even whatever HIP 85606 is) would have an effect on our planetary orbits? How many AU close would a 1 solar mass star need to get to to disturb the orbit of Neptune, for example? Any pointers to dynamical studies would be appreciated.’

    If we look at the effect of our Sun for example on the earth at 1AU it holds it in circular orbit. Now gravities effects fall to the square of distance so a 1 sol mass star at 8000 AU should have 1 / 64 000 000 of the effect. Little effect I would emagine even looking at Neptunes orbit of 40 AU which would be affected by the square of the ratio of the distance or 1 / 40 000 between the stars.

    @Lionel January 4, 2015 at 6:51

    ‘I’d like to know what the statistical models tell us about the number of sub-light year and sub 0.5 light year encounters over the lifetime of our solar system. Just pondering the geometry – presumably the frequency of these encounters is more or less the cube (or should it be the square?) of the closest pass distance, i.e. 1ly encounters are roughly 8 times less frequent than 2ly encounters and so on’

    I was thinking along the line of a 3 D effect as the Suns mass would have an effect of pulling stars towards it increasing the closeness of the encounter. I have had a quick online look at the stastical models of encounters and they appear to be quite complicated.

  • Eniac January 6, 2015, 14:56

    @Michael:

    If we look at the effect of our Sun for example on the earth at 1AU it holds it in circular orbit. Now gravities effects fall to the square of distance so a 1 sol mass star at 8000 AU should have 1 / 64 000 000 of the effect.

    I think it is even much less than that, because what perturbs an orbit is determined by the difference between the effects on the primary and the planet. In other words, it is tidal forces that count, and they decrease with the third power of distance.

  • andy January 6, 2015, 15:34

    Since we can rewind to investigate past close encounters between our Sol and other star systems, what correlation there is with disturbances recorded in, say, the lunar impact record or mass extinctions on Earth.

    Mass extinctions are probably useless for this, the most recent of the great mass extinctions is rather too far back to have any reasonable prospect of finding out what stars were in the neighbourhood at the time. Particularly when you consider we’re still finding red dwarf stars in pretty much our immediate neighbourhood! Furthermore it isn’t particularly clear that impacts have been the primary cause of any of the great mass extinctions: flood basalt eruptions appear to have a rather better track record in this regard, including the end-Cretaceous extinction.

  • Rob Henry January 7, 2015, 22:52

    Andy look where the continents are end Cretaceous.
    http://upload.wikimedia.org/wikipedia/commons/1/10/Blakey_65moll.jpg

    Yucatan is only 2000km from being antipodal to the Decan lava traps. Even worse, India’s location at this time is poorly known due to its much higher seed of travel compared to anything else, so that in some images I have seen it around 500km from antipodal at this time. Sure, that might be coincidences, but to my mind that is sufficiently unlikely to disallow the use of it as a counterexample to the meteorite impact theory (ie it can be more easily read as suporting it). Other evidence against looks weak to me at this time.

  • Michael January 8, 2015, 3:54

    @Eniac January 6, 2015 at 14:56

    ‘I think it is even much less than that, because what perturbs an orbit is determined by the difference between the effects on the primary and the planet. In other words, it is tidal forces that count, and they decrease with the third power of distance.’

    I was thinking along the lines of lowering light levels and hence temperature changes on the planets surfaces which goes to the square of the distance. It is so small as not to worry about it, it would be the comets/asteriods that would be scattered that would be the greatest danger.

    But thinking a little deeper it could bode well for later water addition to parched worlds as each close encounter would force some water bearing comets from ‘their’ Oort clouds to potentially impact dry inner worlds or even bring frozen life to them.

  • andy January 8, 2015, 15:18

    @Rob Henry. The idea that the Chicxulub impact triggered the Deccan traps seems very unlikely to me. See for example this Princeton news release about modelling the effects of such an impact. In particular, previous models which used a spherical Earth model substantially overestimated the likely effects at the antipodes due to these models showing convergence of the shock waves to a single point, which would not happen on the real non-spherical Earth.

    I’m not saying that the Chicxulub impact would have had no effect, but the case for it being the cause of the end-Cretaceous extinction does seem somewhat weaker than it did a decade or so ago.

  • Michael January 9, 2015, 5:25

    @Rob Henry.

    ‘The idea that the Chicxulub impact triggered the Deccan traps seems very unlikely to me. See for example this Princeton news release about modelling the effects of such an impact. In particular, previous models which used a spherical Earth model substantially overestimated the likely effects at the antipodes due to these models showing convergence of the shock waves to a single point, which would not happen on the real non-spherical Earth.’

    I was looking at the impact animation but I cant see if they took into account the rotation of the earth which would tend to offset the waves and consequently the antipodal point. I can only presume that the impact angle played no part as well.

    My thoughts are that even though the waves would tended to arrive at the antipodal point a sizeable concentration of wave passing through a mantle plume in the region could allow an eruption earlier by disturbing it, squeezing and fracturing the rock around and above it. There was certainly mantle plumes in the region.

  • Rob Henry January 9, 2015, 17:53

    andy, there is much that still disturbs me in trying to eliminate that possibility. Firstly, I must disclose a longstanding annoyance with geologist’s modeling. They never seem to fully explain or derive them from first principles, compelling me to reverse engineer every time I want understand. Anyhow, I have come back to this very problem several times, starting as a student before the iridium anomaly in the K-T boundary was found. Even though biochemistry was my major, I discussed the issue with my lecturer who specialised in seismology and, even at that time, he was sure the antipodal focusing would be insufficient due to those anomalies in Earth’s shape, but there is one image I could never shake…

    Imagine that Earth was perfectly elastic, and its shape and the impact was radially symmetrical. In that case a meteorite mass clump of earth would rise from the ground and hurtle out at impact speed. As we reduce elasticity and symmetry we are still going to get a slight lifting of the ground at the antipodes… Our next poorly modeled problem is that geology texts imply that eruptions can not be caused by crushing, but are rather the result of sudden release of pressure. This made me wonder if a massive pressure lowering over a huge volume for a few seconds would do the trick, and perhaps one day I will find the answer in a paper, but I’m no longer holding my breath for it.

  • Rob Henry January 9, 2015, 19:47

    Above, I think I put my concerns quite poorly, as if I expected a sudden explosive eruption. My concern was the positive feed back loop of erupted material removing material from the area above and lowering pressure. Modeling such unstable systems is so hard you tend to use the data available from events in modern times, and that impact would be obviously far outside those conditions.