≡ Menu

Globular Clusters: Home to Intelligent Life?

I can think of few things as spectacular as a globular cluster. Messier 5 is a stunning example in Serpens. With a radius of some 200 light years, M5 shines by the light of half a million stars, and at 13 billion years old, it’s one of the older globular clusters associated with our galaxy. Clusters like these orbit the galactic core, stunning chandeliers of light packed tightly with stars. The Milky Way has 158 known globulars, while M31, the Andromeda galaxy, boasts as many as 500. Giant elliptical galaxies like M87 can have thousands.


Image: The globular cluster Messier 5, consisting of hundreds of thousands of stars. Credit: ESA/Hubble & NASA. Via Wikimedia Commons.

Given the age of globular clusters (an average of ten billion years), it’s a natural assumption that planets within them are going to be rare. We would expect their stars to contain few of the heavy elements demanded by planets, since elements like iron and silicon are created by earlier stellar generations. And indeed, only one globular cluster planet has been found, an apparent gas giant that accompanies a white dwarf, both orbiting a pulsar in the globular cluster M4.

A team led by David Weldrake (Max-Planck-Institut für Astronomie) reported on its large ground-based search for ‘hot Jupiters’ in the globular clusters 47 Tucanae and Omega Centauri in 2006, finding no candidates. So should we give up on the notion of planets in such environments? Rosanne Di Stefano (Harvard-Smithsonian Center for Astrophysics) thinks that would be premature. She presented the research at a press conference at the ongoing meeting of the American Astronomical Society, which this year occurs in Kissimmee, Florida.

Working with Alak Ray (Tata Institute of Fundamental Research, Mumbai), Di Stefano points out that we have found low-metallicity stars known to have planets elsewhere in the galaxy. And while we can couple the occurrence of Jupiter-class planets with stars that have higher metallicity, there seems to be no such correlation when we’re dealing with smaller, rocky worlds like the Earth. Perhaps, then, we shouldn’t dismiss planets in globular clusters.

Di Stefano also argues that while stellar distances are small within a globular cluster, this may not work against the possibility of habitable planets, and may actually be a benefit for any civilizations that do emerge there. Most of the stars in these ancient clusters are red dwarfs with lifetimes in the trillions of years. The habitable zone around such small stars is close in, making potentially habitable worlds relatively safe from disastrous stellar interactions.

So let’s imagine the possibility of life evolving in ten billion year old globulars. In our comparatively sparse region of the Milky Way, the nearest star is presently some 4.2 light years away, a staggering 40 trillion kilometers. Within a large globular cluster, the nearest star could be twenty times closer. Imagine a star at roughly 15000 AU or somewhat less, about the same distance from our Sun as Proxima Centauri is from Centauri A and B. Then open out the view: Imagine fully 10,000 stars closer to us than Alpha Centauri, the night sky ablaze.

I would see this as a powerful inducement to developing interstellar technologies for any civilization that happened to emerge, and Di Stefano agrees, according to this CfA news release. The ‘globular cluster opportunity’ that she identifies would play off the fact that broadcasting messages to nearby stars would presume round-trip travel times not a lot longer than it took to get letters from the United States to Europe in the 18th Century by sail.

As to this site’s own idée fixe, Stefano sees further good news:

“Interstellar travel would take less time too. The Voyager probes are 100 billion miles from Earth, or one-tenth as far as it would take to reach the closest star if we lived in a globular cluster. That means sending an interstellar probe is something a civilization at our technological level could do in a globular cluster.”

That does give me pause — imagine one of our Voyagers already a tenth of the way to another star. If civilizations can develop inside globular clusters, they could be just what Di Stefano says, “the first place in which intelligent life is identified in our galaxy.” The problem for our current exoplanet methods is that even the closest globular cluster is a long way from us — both M4 and NGC 6397 are approximately 7200 light years out. Gravitational microlensing may produce some planetary finds, and perhaps new transit efforts could be effective, though we would be working on the outskirts of the cluster and small rocky worlds would be a tough catch.

Even so, it’s a breathtaking prospect, because my imagination is fired by the possibility of intelligent life looking out on a sky packed with stars from the center of a globular cluster. I also take heart from the fact that there is much we don’t know about such clusters. M5, for example, dates back close to the beginning of our universe, yet we know that along with its ancient stars, it also has a population of the young blue stars called ‘blue stragglers.’ Explaining their formation should help us better understand the dynamics of this dazzling environment.


Comments on this entry are closed.

  • Terry Moseley January 6, 2016, 11:39

    Although most stars in Globulars are red dwarfs, many are much bigger, even giants, otherwise we wouldn’t see them as individual stars even in small telescopes. That means that many have already gone through the SN phase, enriching their environment with metals, so planets should not be that rare.
    That’s the good news. The bad news for SETI is that with so many close stars in a GC there would be fairly frequent nearby SN explosions, with nasty consequences for any advanced life!
    Also, with average star separations so small, there would be many more close passes, disrupting planetary orbits, even flinging some out of their star’s orbit altogether. Even close-in planets in the dwarf star’s HZ, while still probably gravitationally bound, would suffer orbital changes, and would be subject to frequent bombardment by comets from the star’s Oort cloud, assuming that most stars have one.
    Primitive life might well survive, but it could be too hazardous an environment for advanced life to evolve and survive.

  • RobFlores January 6, 2016, 12:09

    All those M stars increase the chances of a planet/moon conductive to
    evolution of life forms.
    With such large time scales and stability there is the possibility of low metabolic life forms evolving complexity. And if there is a paucity of
    elements beyond C,H,O,N it is very probable that those limits will
    be reflected in the pace of evolution, at least at the single cell level.
    On the other hand, excessive stability is not necessarily conductive
    to increasing complexity. A stable ecology with all niches filled means
    less opportunity for speciation, even at the single celled level.
    With starts pressed together in a cluster, maybe such a region of space is more conductive to interstellar seeding of micro organisms from one source. If we find many habitable planets/Moons, maybe we will find the same single celled have the same lineage, on many of them.

  • Michael January 6, 2016, 12:46

    Not just stars close by but black holes, pulsars and white dwarfs the means to see to the edge of the Universe. If we did get a lensing probe to our Sun’s focal point they would be a nice place to stake out. As for there been so few planets detected it could more to do with the distance than planet forming processes.

  • Adrian Ratnapala January 6, 2016, 13:04

    @TerryMosley while the average insterstellar distance in a globular cluster is very small, what is the average distance to a supernova candiate? I.e. how many big stars do GCs have, and often do they explode? Are the numbers worse than for our own neigbourhood?

    As for orbital variations and the Oort cloud, we are getting into unknown-unknowns territory. We don’t even know if _we_ have an Oort cloud. In a neighbourhood were interstellar distances are of the order of the cloude size, who knows what happens?

    Similarly, suppose the orbit of GC planets vary significantly on million-year timescales, will that acutally harm evolution? Consider our own ice ages, which come and go much faster than that. Human intelligence mangaged to evolve exactly during the period when the Earth’s climate started to oscillate hugely (https://en.wikipedia.org/wiki/Quaternary_glaciation). Who can say what would happen on other planets?

  • David Bernard January 6, 2016, 14:01

    Well it will be interesting if there are Chirpsithra out their!

  • Paul Gilster January 6, 2016, 14:55

    For those interested in the Chirpsithra, see Larry Niven’s The Draco Tavern (Tor, 2006) for background.

  • Alex Tolley January 6, 2016, 15:48

    David Hardy’s painting of alien life in a globular cluster.

    From Challenge of the Stars written by Patrick Moore, illustrated by Hardy.

    With such close proximity, wouldn’t the stars move quite perceptibly over a relatively short time, sufficiently so that a generation of star maps would show these movements and therefore indicate distance more easily than on Earth? Alien astronomers might therefore have a sense of stellar distances quite early on compared to us.

    While we are thinking perhaps 40 years to alpha Centauri, and ETI probe might do a flight to the nearest star in just 5 years, which is very doable as a mission time for us, with data returning with just a 6 month delay.

  • FrankH January 6, 2016, 15:51

    I take issue with this sentence:
    “we have found low-metallicity stars known to have planets elsewhere in the galaxy”

    I don’t think that’s a given. Until you have either unequivocal transit data, or direct imaging, these planets are putative or possible detection.
    For example, the planets around Kapteyn’s star, which is pretty close (so a good S/N ratio) may not exist – the planet signals may just be part of misinterpreted the star’s activity:


    @RobFlores – I doubt we’ll be finding many moons around red dwarf planets – at least not in the HZ of the average, faint M dwarf. The Hill sphere for a planet that close to its host star is going to be tiny.

  • Ashley Baldwin January 6, 2016, 16:22

    I think common opinion is that given the tight packed nature of globular clusters the increases the likelihood of multiple stellar mergers leading to rare but big “blue stragglers” . In terms of metallicity , we know that for stars in conventional areas that the gases clouds of formation are seeded by regular supernova explosions leading to increasing metallicity in successive generations of stars . In globular clusters blue stragglers may be rare but given the compact nature of the stars any supernova would distribute heavy elements over an area that had a far greater stellar density than areas of conventional concentration so more stars would be effected . Thus even a small number of blue stragglers would be very much more effective at distributing their elements than normal core collapse type II supernovae. Sowing by a few tractors rather than by many hands so to speak . Certainly on fertile ground . .Their elements would thus “dope” pre-existing stars rather than “seed” gaseous star forming clouds , but the end result would be the same or similar. Stars with higher levels of metallicity thanks to supernovae.

  • Jim Vassilakos January 6, 2016, 16:24

    Yeah, my guess is that there would be more debris (comets, asteroids, etc.) flying around. Oort clouds would likely overlap quite a bit. There’d probably be lots of rogue planets drifting between the stars. The upshot is that a habitable planet would be subject to more “cosmic events”, including large objects falling from the sky. I’m not sure there would be 10000 times as many, but if there were, then instead of expecting a global extinction-sized asteroid once every 250 millions years or so (probably a conservative estimate), you might expect one once every 25000 years, which would be pretty awful for any sort of animal life larger than, say, a rat. Hence, there might not be a sufficient window for the evolution of the large brains presumably required for the development of sentience & global civilization. Not saying that it couldn’t happen… just that the odds might be stacked very heavily against such outcomes, although the sheer number of star systems available for various lifeforms to evolve might overwhelm the negatives. Maybe most of the species that end up surviving would have evolved in largely subterranean environments. Bunnies & Burrows, anyone? https://upload.wikimedia.org/wikipedia/en/3/35/Bunnies_and_burrows.jpg

  • Hiro January 6, 2016, 17:51

    The outer region of globular cluster is okay I think, at least the distribution of star is lower than the inner region hence nasty events are minimized. I’ll read Asimov’s Night Fall again….

  • Frank Smith January 6, 2016, 18:55

    Gotta disagree with paper’s authors on this one. A supernova within a couple of light years would make the survival of advanced life unlikely. The advanced age of globular clusters would make it statistically likely that there have been numerous supernovas during their history.

  • Neil S January 6, 2016, 22:41

    Thanks to your column I came across this post about the interior lives of GC’s.

    Apparently several things could happen to any planet of a GC star that would be hard on complex life. It could be ripped from its star by a passing star and be tossed about in the cold; its star could become part of a binary which then becomes closer and closer together as its energy is transferred to other passing stars; it could pass close to a pulsar or its star could get thrown into the very dense core of the cluster.

    Way too much excitement!

  • ljk January 7, 2016, 1:16

    Forget supernovae in a globular star cluster, all you need are enough moons to eclipse the suns of a world where the inhabitants have only known daylight just long enough to cause a darkness that will bring down their civilization in a flash – literally:


    And here is the full story:


    While it is nice to see modern professional astronomers finally stepping it up a bit when it comes to alien intelligences – why didn’t we just have a few such fellows get everyone all stirred up about a possible Dyson Swarm just under 1,500 light years distant – I wonder if the astronomer suggesting aliens in a globular star cluster might have considered that they do not have to be native to this ball of suns. Rather, advanced ETI might be dwelling there in the equivalent capacity of a mining colony. If they are sophisticated enough, they could also attempt to extend the lifetimes of those stars and even stave off supernovae so they can get all the resources they can.

  • kzb January 7, 2016, 9:45

    There was a paper some years ago about the possibility of a delocalised planet population in GCs. Whilst the negative result for 47-Tuc seems to argue against that, it was only sensitive to large gas giants which are expected to be rare given the low metallicity environment. On the other hand, any planets formed in wider orbits are quite likely to be disrupted by close stellar encounters and find themselves without any star to orbit.
    So it’s still a possibility that there are a large number of delocalised small planets in these clusters.

  • Harry R Ray January 7, 2016, 11:18

    A rocky planet HAS been discovered orbiting a VERY anemic star, adding to the credibility of this claim! If this COULD happen, a NEW avenue of research for AMETEUR SETI wound be CONTINUOUS monitering of a targeted globular cluster for GRAVITATIONALLY LENSED BENFORD BEACONS! If SEVERAL candidate signals are detectes in any one cluster over a period of years, a more sensitive space radio telescope could then be launched to stare(a la Kepler) at that cluster ONLY to possibly pick up other ET signals such as gravitationally lensed lightsail leakage. Could Dr Shuch(ore anyone ELSE at ARGUS) QUANTIFY the chances of success and report back to this website via either comment or post?

  • Nelson Bridwell January 7, 2016, 15:40

    Color me skeptical:
    (1) The relative lack of elements (C, N, O, P, S) that are essential for biological life.
    (2) The effects of nearby supernovas.
    (3) The effects of nearby GRBs.
    (4) If these are the core remnants of dwarf galaxies, then presumably they contain very large black holes, which also doesn’t sound especially promising.

  • andy January 7, 2016, 16:32

    As far as I’ve been able to figure, it should be possible for a star in a globular cluster to avoid passing through the core (where the potential for disruptive encounters is highest) for relatively long periods of time. Supernovae might not be all that big a problem if you can survive everything else: with one rapid episode of star formation the massive stars wouldn’t be much of a problem after the first few tens of millions of years. Type Ia explosions might still be a hazard though – not sure if there are any predictions on the rate of such events in globular clusters, and the eruption of V838 Monocerotis indicates that stellar mergers (the kind that might be responsible for forming blue stragglers) can be quite spectacular also.

    Question is what kind of planets could form in a globular cluster: presumably the environment would have been extremely hostile, with large numbers of O-type stars in a small volume of space, probably the Arches cluster might give some idea of the type of environment. This might have resulted in evaporation of discs around many of the stars. So far there is only one known globular cluster planet: a giant planet orbiting the millisecond pulsar/white dwarf binary PSR B1620-26. Whether it formed as a planet or as a sub-brown dwarf is an interesting question. Kapteyn’s star has been suggested as a planetary system orbiting a star that was once part of Omega Centauri, but the association with Omega Centauri has been called into question (as has one of the claimed planets), and Omega Centauri appears to be a remnant dwarf galaxy rather than a true globular cluster.

    Another question is whether life would be expected at the current epoch: globular cluster stars are old, giving more than enough time for planetary cooling to take its toll on geological activity and the magnetic dynamos. (I tend to be a pessimist as regards the lifetime of technological civilisations and the feasibility of interstellar travel, even for the relatively close distances of stars within a globular cluster).

  • Geoffrey Hillend January 7, 2016, 17:25

    @Ljk you didn’t read that article carefully enough. The half day long eclipse is a rare planetary alignment and would not be physical threat to an Earth like planet on in a globular cluster on a long term basis. It was the psychological shock of the eclipse that is being stressed.

  • Sanford Zane Meschkow January 7, 2016, 21:57

    Someone should do a study on how the hazards for solar systems and planets might vary in various parts of a globular cluster over time. Perhaps stable conditions that would allow life forms to evolve were almost impossible early in the cluster’s history, what with supernovas, rogue planets, comets, etc., but things may have calmed down by now. Would some planets be subjected to a small but constant drizzle of heavier elements produced by many red giants and distant supernovas to the point that after billions of years planets too small to have plate tectonics would grow and become more geologically active and more Earthlike?

  • ljk January 8, 2016, 10:26

    Geoffrey Hillend said on January 7, 2016 at 17:25:

    “@Ljk you didn’t read that article carefully enough. The half day long eclipse is a rare planetary alignment and would not be physical threat to an Earth like planet on in a globular cluster on a long term basis. It was the psychological shock of the eclipse that is being stressed.”

    I was being a bit facetious. I love Asimov’s story and know it quite well. I was putting out the idea of perhaps living in a globular cluster would have some rather different effects on the psychology of any intelligent beings there, though Lagash’s case was a rather intense one with six suns drowning out the rest of the Universe most of the time.

    And for those who could see all those stars in the cluster and then discovering how relatively close they are, might that compel such inhabitants to want to reach for them?

    On a related note, the giant globular star cluster Omega Centauri is very likely a remnant of a former galaxy that encountered the Milky Way long ago and was largely absorbed. So if OC was once a galaxy, does this mean it has an even better chance for containing suns – and life?


    How many SETI efforts have been aimed at GSCs? And yes I am very well aware of the Arecibo Message sent to Messier 13 in 1974.


  • Stan Erickson January 8, 2016, 16:35

    I would guess the biggest problem is stellar encounters. With stars so close, these should be common events in the history of any solar system. Expect that planetary orbits would be disrupted, certainly thrown out of any ‘habitable’ zone, made elliptical or even cast out of the solar system. I’m too lazy to compute the probability of these happening, but there should be some good rule-of-thumb to estimate it.

  • Hiro January 8, 2016, 17:53

    Instead of focusing the cluster center, how about the narrow region around 2/3 to 4/5 from the core? The distance between stars should be far enough for primitive life to evolves but I’m not sure about the distribution of heavier elements in this region hence another pure speculation.

  • Nelson Bridwell January 8, 2016, 19:15

    If I were looking for technological life rather than mutating slime, I suspect that it might be more promising to look for places that would be attractive to non-biological life forms.

    Perhaps “far away from the madding crowd” of galaxies. Perhaps places where there are vast quantities of unconsumed hydrogen that can continue to generate needed energy long after the very last white dwarf has gone dark.

    Someplace that is not too hot. Maybe a balmy ~ 3 Kelvin, which could be great for superconductivity and minimize thermal wear.

    And maybe distributed over a wide enough extent that any localized damage from particles, cosmic rays, and rogue black holes cannot take down the entire fabric of civilization…

  • Nelson Bridwell January 8, 2016, 19:35
  • ljk January 11, 2016, 9:45

    The late Robert Bradbury also supported GSC as good places for Artilects to go:


    GSCs can work if you look outside the usual paradigm box of aliens that are similar to us; Think of artificial intelligences unbounded by our biological limitations. Ones that truly know how to use interstellar resources, especially ones in relatively close proximity to each other. Vast amounts of energy from so many suns together.

    A very relevant CD article on the topic from 2014: