≡ Menu

Dyson Sphere Hunt Using Kepler Data

The idea of the multiverse — an infinite number of universes co-existing with our own — has a philosophical and mathematical appeal, at least if you’re a follower of string theory. Indeed, there are those who would argue there could be as many as 10500 universes, each with its own particular characteristics, most probably inimical to the development of life. But I have to say that I’m far more interested in the universe that is demonstrably here, our own, and thus the news that Geoff Marcy has received a grant to look for Dyson spheres catches my eye more than news of a similar grant to physicist Raphael Bousso to probe multiverse theory.

Not that I have anything against Dr. Bousso (UC-Berkeley) and his work, and if he eventually does find a way to make predictions of multiverse theory that can be tested, I’m all for it. But I think the new grants, given to the researchers in a series called New Frontiers in Astronomy & Cosmology International Grants (funded through the UK’s Templeton Foundation), were wisely balanced between the practical (observational astronomy) and the highly theoretical. Marcy will use his to probe the continually swelling Kepler datastream looking for distinctive signatures.

Instead of planets, though, Marcy has more unusual targets, the vast structures Freeman Dyson hypothesized over fifty years ago that could ring or completely enclose their parent star. Such structures, the work of a Kardashev Type II civilization — one capable of drawing on the entire energy output of its star — would power the most power-hungry society and offer up reserves of energy that would support its continuing expansion into the cosmos, if it so chose. Marcy’s plan is to look at a thousand Kepler systems for telltale evidence of such structures by examining changes in light levels around the parent star.

Image: A Dyson sphere under construction. Credit: Steve Bowers.

Interestingly, the grant of $200,000 goes beyond the Dyson sphere search to look into possible laser traffic among extraterrestrial civilizations. Says Marcy:

“Technological civilizations may communicate with their space probes located throughout the galaxy by using laser beams, either in visible light or infrared light. Laser light is detectable from other civilizations because the power is concentrated into a narrow beam and the light is all at one specific color or frequency. The lasers outshine the host star at the color of the laser.”

The topic of Dyson spheres calls Richard Carrigan to mind. The retired Fermilab physicist has studied data from the Infrared Astronomical Satellite (IRAS) to identify objects that radiate waste heat in ways that imply a star completely enclosed by a Dyson sphere. This is unconventional SETI in that it presumes no beacons deliberately announcing themselves to the cosmos, but instead looks for signs of civilization that are the natural consequences of physics.

Carrigan has estimated that a star like the Sun, if enclosed with a shell at the radius of the Earth, would re-radiate its energies at approximately 300 Kelvin. Marcy will turn some of the thinking behind what Carrigan calls ‘cosmic archaeology’ toward stellar systems we now know to have planets, thanks to the work of Kepler. Ultimately, Carrigan’s ‘archaeology’ could extend to planetary atmospheres possibly marked by industrial activity, or perhaps forms of large-scale engineering other than Dyson spheres that may be acquired through astronomical surveys and remain waiting in our data to be discovered. All this reminds us once again how the model for SETI is changing.

For more, see two Richard Carrigan papers, the first being “IRAS-based Whole-Sky Upper Limit on Dyson Spheres,” Journal of Astrophysics 698 (2009), pp. 2075-2086 (preprint), and “Starry Messages: Searching for Signatures of Interstellar Archaeology,” JBIS 63 (2010), p. 90 (preprint). Also see James Annis, “Placing a limit on star-fed Kardashev type III civilisations,” JBIS 52, pp.33-36 (1999). A recent Centauri Dreams story on all this is Interstellar Archaeology on the Galactic Scale but see also Searching for Dyson Spheres and Toward an Interstellar Archaeology.


Comments on this entry are closed.

  • Edg Duveyoung October 11, 2012, 9:20

    How many years do we estimate must pass before we could get to Type II at our present rate? 1,000 years? 5,000?

    Yet, even today’s science is talking about ripping space itself apart so as to cherry-pick what we want from the virtual particles.

    Isn’t it more likely that we figure out how to get free energy from a vacuum long before we have the engineering heft to do a Cosmic Bucky Dome?

    Would there be a detectable signature of that civilization-big magnitude of quantum hacking?

  • Eric October 11, 2012, 10:48
  • Greg Parris October 11, 2012, 12:19

    This is good stuff. Collecting the power from our star on a hyper-industrial scale makes lots of sense. Based on our current physics, this is the most cost effective method for harvesting 348 X 10E26 watts (K2 status). You can’t do this sort of thing on the earth’s surface because there is no adequate way to dissipate enough heat and there isn’t enough energy content here anyway (i.e. fusion of deuterium in the earth’s oceans). No, you either have to build a solar scale reactor and bring the fusable material to it or you harvest the power from our existing solar scale open core reactor that already has its own fuel. Hey! Let’s do the latter!

    If you want to do high energy density things your going to need high energy density fuel (i.e. antimatter) and if your going to do lots of this you need lots of power (i.e. Status > K1). Assuming no breakthrough physics, other folks in our galactic neighborhood are going to come to similar conclusions and some of them will build hyper-industrial energy harvesting of stars.

    I think searching for Dyson objects is a good thing to do. Bravo!

  • James M Essig October 11, 2012, 12:20

    Hi Paul;

    I share your interest and fascination with the one universe that we know of. For all we know, our universe may extend forever in space and forward in time. There is a lot of talk about speculative concepts for inter-universe travel within a multiverse, however, perhaps from a philosophical standpoint, the best way for we humans to travel in relation to our own universe is to stay within 4-D Einsteinian spacetime. One can argue from philosophical grounds that this space time is the most natural for us to consider.

    I have been considering prospects for extremely cloaked spacecraft traveling at extreme gamma factors, even ensemble gamma factors, and how such spacecraft being thermodynamically almost completely seperated from 4-D spacetime may in effect not even be traveling in 4-D spacetime in the most proper sense, but rather instead be traveling in a margin of 4-D spacetime or the boundary between such space time and no space-time at all. I mention this concept in my second recently published book.

    The prospect that our universe may be infinite in spatial extent, but which continues to expand forever is quite exotic enough to relish in. Considering the possibility of future phase changes, inflationary events, and symmetry breaking events is quite profound. How many such future events might happen remains to be seen, but may indeed be limited to arbitrary undefined infinite numbers where the potential infinite numbers are uncertain and therefore may continue to grow for ever.

    My above speculations are indeed extreme, but perhaps if we can intuit and mathematicalize a multiverse composed of infinite numbers of universes, such speculations may seem more plausible.

    Back to the primary subject at hand, looking for Dyson Spheres and ring worlds is a fascinating prospect. Our Sun as a 4 x (10 EXP 26) wattt power source puts out about 10 EXP 14 times the power by Humanity does by industry on planet Earth. Viable small end red dwarfs average about 10 EXP 22.5 watts but will be around for trillions of years.

    I have an appreciation for solar energy as I worked in the inventive field of light weight in situ resource harnessing equipment. My brother John and I were experimenting with a 6 foot diameter inflatable reflector we manufactured by hand at very low cost where John would aim the reflector during near sunset at a log I was holding near the focal point which was about 36 feet away from the device. People driving by would notice the log smoking out of no then bursting into flames.

    Needless to say, although I am an affectionato for nuclear and sub-nuclear energy applications for manned starship travel, other folks instead fear nuclear and sub-nuclear physics research and development. However, such folks can also have the starship cake as well. Simply consider the collection of vaste stellar light sources for industry and special and/or general relativistic space travel.

    I would think that stellar power would be too big of an energy source not to be utilized by advanced ETI civilizations. We should definately continue looking for Dyson Spheres and ring worlds.

  • Greg Parris October 11, 2012, 12:37

    Question: There have been a number of discussions about Dyson spheres and Dyson swarms in the past. Does anyone know the difference in detection signature between the two? Also, what would be the likelihood that the Dyson object radius would be materially different than 1AU. If the purpose of the structure is to collect power, what other radii might be optimal and what would be the difference in detection signature? Thanks.

  • JoJo The Rat October 11, 2012, 15:24

    A detection seems unlikely given the small sampling size and criteria that aren’t necessarily relevant (e.g., being a Sun-like star and having planets). Such things might be useful for finding planetary civilizations, but wouldn’t a Kardashev II civilization build a sphere around the best kind of star for a sphere, not for planet dwelling? In any case, it’s a very small sampling, and not likely to produce any hits. I like that they’re looking though.

  • Steve Bowers October 11, 2012, 15:28

    I’d put the swarm or shell as close to the star as convenient, so that the power density would be maximised. But that would cause the waste heat emissions to occur over a smaller surface area; this would make them hotter and more detectable. If instead you want to reduce the visibility of the swarm or shell, you make it as big as convenient, so that the waste heat is cooler.

    Swarms would resemble a dust cloud from a distance, although probably more regular in shape. A shell, on the other hand, would be a different proposition. it could take the form of a statite bubble, supported by light pressure, or a suprastellar shell as described by the late Paul Birch. The suprastellar shell would require fantastic amounts of matter to build, unless you left substantial holes in the structure. Statite bubbles would necessarily be very lightweight indeed; about as massive as a single large asteroid.

  • Christopher Phoenix October 11, 2012, 17:02

    It is quite likely that spacefaring civilizations will use vast arrays of solar cells to gather vast amounts of energy from sunlight- possibly even up to enclosing an entire star in a massive swarm that blocks much of the light from the star. I think it is a great idea to hunt for alien civilizations by looking for the byproducts of such advanced technology, not deliberate beacons, since we don’t have to make any assumptions about whether, why, and how aliens attempt to communicate with other civilizations.

    I have a question, though- is Geoff Marcy only looking for the signatures of a solid sphere encircling a star at one AU, or a Dyson swarm? Freeman Dyson originally suggested that an advanced civilization will build a massive swarm of solar power satellites to gather energy. The solid “Dyson sphere” appeared in SF after Dyson’s original proposal, but no existing building material is strong enough to construct one, so it is likely to remain fictional. Also, what if the swarm is actually located much closer to the star so that the satellites have access to more intense sunlight? Would this change the signature much?

  • NS October 11, 2012, 21:22

    Kepler data is not restricted to sun-like stars or stars with planets, so it should be possible to check a variety of stellar types. I don’t see what criteria Marcy is using though.

  • James Jason Wentworth October 11, 2012, 21:28

    This leads me to wonder: Might a statite partial–*hemispherical*, that is–Dyson sphere (or such an array of “Dyson Dot” statites) be used to convert a stellar system into a gigantic, light pressure-propelled “starship?” Particularly for inhabitants of a stellar system whose sun is a red dwarf star, the required amount of material for the statite(s) would not be unreasonable, and “taking one’s solar system along” would neatly avoid all of the social and resource recycling problems of multi-generational starships and other such powered space arks. Also:

    If the system’s inhabitants were (or discovered how to become) very long-lived or perhaps even immortal, their main problem would just be patience–“Are we there yet? Are we there yet?” :-) The upcoming Dyson Sphere search might turn up such “Starship Sun” ‘vessels’ if any exist, particularly if they were heading in our direction, with their infrared-emitting statite hemispheres facing in our general direction. Something with a high proper motion like Barnard’s Star, yet appearing much cooler to us despite being of stellar size, might be a candidate “Starship Sun.”

  • Paul October 11, 2012, 22:34

    Finding a human styled “Dyson Sphere” in the limited Kepler data which have not even been able to live up to their stated mission objective, finding an earth sized planet in the habitable zone of it’s host star.
    Talk about a needle in a galactic haystack….lots of luck with that.

  • Eniac October 11, 2012, 22:50

    The assumption that rigid Dyson spheres require impossible materials warrants a second look. There is some math here: http://www.nada.kth.se/~asa/dysonFAQ.html#STRENGTH which appears to indicate that a sphere around our sun would require a material about 10-100 times stronger than the strongest we know (diamond, I suppose). This is already not as bad as one might fear. Furthermore, this required strength is proportional to the mass of the star and inversely proportional to the radius of the sphere. So, by chosing a lighter star and a larger sphere, we might just be able to pull it off.

    Are there any light, but bright stars out there? Is suppose those would be the ones burning out the fastest, so a compromise between feasibility and longevity seems in order….

  • Eniac October 11, 2012, 23:06

    Actually, the above reference estimates the pressure as that of the base of a tower of less than 100,000 km height in Earth normal gravity, which would amount to 3.5 * 10^12 Pascal, given the density of diamond of 3,500 kg/m^3. The compressive strength of diamond is estimated here: http://adsabs.harvard.edu/abs/1979JAP….50.2763N as 4 * 10^11 Pascal.

    This would mean that we have just a little bit less than one order of magnitude to go before we can build that rigid Dyson sphere. Theoretically speaking.

  • Eniac October 11, 2012, 23:19

    Since constructing a Dyson sphere is quite a large project, it would seem that there should be quite a few of these half-completed ones out there, like the one depicted in the beautiful image of the original post. If they are not completely stationary, you might expect them to alternately obscure and not obscure the star, which should give rise to a very distinct signature in the Kepler light curve of that star. Was this considered by Geoff Marcy or others?

    PS: The link in my previous post did not post well. Here is an alternative address: http://jap.aip.org/resource/1/japiau/v50/i4/p2763_s1

  • Eniac October 11, 2012, 23:32

    Incidentally, I would also like to note at this opportunity that the tensile strength of Diamond, while not exactly known, is estimated as “could be as high as 90–225 GPa ” at Wikipedia (http://en.wikipedia.org/wiki/Material_properties_of_diamond). At the higher end of the range, that would make it sufficiently strong for a space elevator, which some like to think of as the ideal tool to make those first baby steps into space. So, perhaps the coming “Diamond Age” is going to make a big difference in how we get into space, among other things.

    I can’t wait for those diamond water bottles, by the way, the ultimate in low-waste packaging. And, of course, diamond Saran wrap.

  • ljk October 12, 2012, 9:19

    Forget Wimpy Plans and NIMBYs, Let’s Solve the Energy Crisis by Blowing Up Mercury

    Posted by Derek_Mead on Wednesday, Apr 04, 2012

    With all the squabbling about oil killing us all, climate change screwing with polar bears, nuclear plants falling apart, solar panels sucking on a cloudy day, and wind turbines scything through migratory birds with a gory violence best explained by an Omega Crom song, there’s a big point that all the complainers in the energy debate are ignoring: These days, we are being huge wimps.

    Millennia ago us humans were building the pyramids and the Coliseum; last century Edward Teller was talking about expanding the Panama Canal with nuclear bombs. Yet this century we’re piddling along, arguing about whether wind turbines are too noisy or whether solar arrays look ugly. Where are the big ideas?

    Thankfully, George Dvorsky of Sentient Developments isn’t being a child. He’s thinking big to solve our energy crisis. His solution? Build a Dyson sphere, a massive solar array that would envelop the Sun itself. Such a system would provide us with so much damn energy we wouldn’t even know what to do.

    Imagine: You could leave on all the lights in the house all the time! Run the microwave and toaster without power surges! (Plus there’s that whole thing about eliminating dirty energy, nuclear disasters, and wars over energy resources.)

    As Dvorsky argues, a Dyson sphere would transform us from a wimpy, embarrassing, joke-in-the-universe Type 1 Kardashev civilization into an energy-secure, boomboxes-blasting-all-the-time Type 2 world. We’d finally be masters of our domain, as it were.

    Full article here:


  • ljk October 12, 2012, 10:09

    James Jason Wentworth said on October 11, 2012 at 21:28:

    This leads me to wonder: Might a statite partial–*hemispherical*, that is–Dyson sphere (or such an array of “Dyson Dot” statites) be used to convert a stellar system into a gigantic, light pressure-propelled “starship?”

    LJK replies:

    The thought of Dyson Shells/Swarms moving through the galaxy is an amazing one, to be sure. Nothing like taking a virtual solar system of resources with you. You probably wouldn’t be bothered by much, either, expect maybe by another mobile Dyson Shell.

    Fritz Zwicky had a similar idea about using our Sol system as a megastarship with the ultimate fusion reactor, our yellow dwarf star, providing the propulsion:


    If Dyson Shells themselves cannot be moved around, they would certainly be able to provide the collected energy to push solar sail missions to other star systems. The beams might also make formidable interstellar scale weapons:


    What we really need to do, though, is get past the original notion of a Dyson Sphere as a huge enclosure for organic beings like us. Just as a typical interstellar vessel would be much more efficient unmanned and controlled by an Artilect (AI, Artificial Intellect), so Dyson Shells would probably make more sense to be constructed by and for Artilects.

    The late Robert Bradbury, who I credit for doing much to push us past the usual paradigms when it comes to the evolution of intelligence, artificial minds, and SETI among other concepts, thought of Dyson Shells as Matroishka Brains, nested shells dedicated to these powerful artificial intelligences.


    Orion’s Arm has also played with this idea:


  • ljk October 12, 2012, 10:28

    Paul said on October 11, 2012 at 22:34:

    “Finding a human styled “Dyson Sphere” in the limited Kepler data which have not even been able to live up to their stated mission objective, finding an earth sized planet in the habitable zone of it’s host star. Talk about a needle in a galactic haystack….lots of luck with that.”

    I have been advocating this here and elsewhere since at least 2008, but for those who want to search for Kardashev Type II and III civilizations should take a closer look at spiral galaxy NGC 5907. It appears to have an abundance of red dwarf stars above the norm for this type of stellar island. Or at least that is what the professionals are saying.

    See here for more information:


    Now, can we get some professional astronomers with the right equipment to study NGC 5907 and other similar galaxies properly, or will they balk at such an idea?

  • Greg Parris October 12, 2012, 10:49

    Has there been any consideration given to the ecological impact to the local system by the building of a Dyson swarm? This would most certainly make any planets outside the structure’s radius uninhabitable and prevent any terraforming opportunities. It may make more sense to build your power plant in a neighboring system and not spoil the natural ecology of where you live.

  • Matt McIrvin October 12, 2012, 14:16

    “Isn’t it more likely that we figure out how to get free energy from a vacuum long before we have the engineering heft to do a Cosmic Bucky Dome?”

    Not if it isn’t possible.

  • Rob Henry October 12, 2012, 16:37

    Eniac, note that all those Dyson sphere calculations are for shells that are both rigid and static, I always wondered if there exists an answer was more dynamic and complicated, yet still feasible and formed a perfect sphere.

    Also note that the strength of diamond is not sufficient for a space elevator in your calculations before we realise that it is possible to taper it. Actually if elevators are the way to the future, and our descendants insist on planting their feet on solid ground, Mars (or perhaps even Ceres) should end up being the central hub for our civilisation, due to accessibility factors.

  • Raimo Kangasniemi October 12, 2012, 16:38

    Christopher Phoenix, Marcy’s group is searching for Dyson swarms. They consider actual Dyson spheres to be impractical. Basically they are looking for transit signals from Dyson swarms.

  • Christopher Phoenix October 12, 2012, 18:28


    Sure, but where do you suppose to find this material that is 10-100 times stronger than any known material (and in such quantities that you can build a solid sphere around the whole sun)? The strength of materials depends on the forces that hold the atoms together. Short of finding some wonderful new material held together by forces equal to or greater than the nuclear binding force, there is no super-material we can use to construct a Dyson Sphere. Good luck with that. Sure, you can say “Maybe future people/aliens will be able to do this!!”, but at a certain point you are just clicking your heels and saying you believe in magic.

    Additionally, there are the problems with constructing the Dyson Sphere (half a sphere is unstable). One can imagine bringing a number of previously free orbiting components (i.e. twenty spherical triangles) together in a dramatic capping process- but tremendous precision would be required. But, forgetting that for a moment, where are we going to get the quantity building materials for this sphere? Dyson estimated that there is enough mass in our solar system to build a shell at least 3 meter thick, but this is an overstatement as most of the material in our solar system is hydrogen and helium, and not usable as building materials so far as we know. We could fuse them into heavier elements, but if we can fusion elements on that scale, who needs a Dyson sphere? Don’t forget we still need those super-materials, which calls for complete conversion of most of the normal matter in our solar system to as-yet undefined super-materials- i.e. magic. As one person said, if you can build a Dyson Sphere you don’t need it.

    A simple Dyson Swarm is much more likely, and much more difficult to detect. Megastructures of all kinds are lots of fun to imagine, but solid spheres around stars are rather implausible!! I was always partial to planetary haloes myself. A satellite halo consists of countless artificial satellites orbiting a planet in a ring so dense, it is visible from space. We might build a densely populated ring of orbital habitats, for instance. The more extreme version is a solid, unified ring-like structure that completely encircles a planet.

    It is amusing to imagine a heavily armed satellite halo acting as a literal “defensive line” around a planet- say, a ring of heavily armed battle stations with powerful lasers to destroy any alien attackers. Maybe with some kamikaze “space mines” to fill the gaps. This begs the question of why the alien motherships don’t just approach our planet from another vector that avoids the heavily armed halo- but perhaps we can build many orbital haloes, each at a different altitude and orbit, to cover most of the entire planet from any approach. General Patton would not approve, however. “Fixed fortifications are monuments to man’s stupidity”- George S. Patton on the Siegfried Line.

  • Brett Bellmore October 12, 2012, 19:00

    Nice art, but pictures of “partially completed” Dyson spheres always give me a chuckle. Barring some extreme form of unobtainium, any Dyson sphere is going to be constructed so as to be in dynamic equilibrium everywhere at all times, and so will exhibit radial symmetry. Likely you’d start with a ring of solar power satellites around the star’s equator, and start building towards the poles in both directions. As you departed the plane of the equator, light pressure would have to be used to compensate for off axial forces.

    As closure approached 100%, the interior would fill with reflected sunlight, like a balloon inflated with radiation, and light pressure would start increasing. The rotation would be gradually braked, perhaps ballast added to compensate, and you’d end up with the entire star tiled with radiation levitated power stations edge to edge, standing still. The whole shell barely thicker than a children’s balloon, in all likelihood.

    A partially completed sphere would doubtless have some interesting effects on the stellar spectrum, as the light leaking out the gaps would have undergone multiple reflections, and the star would be heating due to being surrounded by a bubble of high intensity light. There’d probably be spotlights transmitting simulated sunlight to the remaining planets, at power levels calculated to put them all in the habitable zone.

    Wouldn’t look anything like the images artists usually generate.

  • Eniac October 12, 2012, 22:35

    James Jason Wentworth:

    This leads me to wonder: Might a statite partial–*hemispherical*, that is–Dyson sphere (or such an array of “Dyson Dot” statites) be used to convert a stellar system into a gigantic, light pressure-propelled “starship?”

    This is an interesting idea, but such contraptions could never achieve appreciable velocities. A star uses up only about a percent of its mass over billions of years, which by conservation of momentum implies that it would take billions of years to accelerate to 1/100 th of the “exhaust velocity”. Assuming that the dominant output of a star is its light, that would be 0.01 c. It would thus take millions of years to accelerate to 0.00001 c, or 3 km/s. It would require enormous patience to get anywhere this way. Not that there really is anywhere to go without the risk of destroying one’s system upon arrival.

  • NS October 12, 2012, 23:47

    I think it highly unlikely that Dyson spheres or any alien technology exists, and less likely that we could find them if they did. I will applaud if someone proves me wrong. Marcy and anyone else who attempts it deserves our support.

  • IB October 13, 2012, 10:30

    The usefulness of the Kepler datastream continues to astound. First a cornucopia of exoplanets, now a search for xenoarchaeology. Best wishes to Dr Marcy.

    Success would not just be the greatest discovery in human history but also a wonderful, uplifting message to us about our own potential. I’m curious about the future of SETI-related searches and their effect on our thinking about about our own future. What if we continue to find nothing? How many more years of negative results before we begin to seriously consider the possibility that a Great Filter is ahead of us? And how long will it before we begin to pass through it? Could the continued absence of Contact despite intensive searching for it begin to have a cultural effect?

  • A. A. Jackson October 14, 2012, 8:00

    So what if there is never evidence for ‘Dyson Artifacts’ or Kardashev civilizations?
    On the face of it this would tell me that advanced civilizations are more subtle than we imagine. Once again that we can not even predict what our future as a complex technological civilization will be.
    Philip Morrison was not a fan of Malthusian scenarios. He expressed these views in print and once I heard him speak off hand about it at a lecture he gave back in 1979 near San Francisco. He considered the ideas interesting, but too rooted in anthropomorphic thinking and stuck in a kind of human historical timeline binding that he felt lacked imagination. He said more than once too much deduction was made from ‘pure thought’.
    It’s not that we should not think about such things , but caution about hubris on this subject should be one’s operational mode, since alas, we might be cornered into some closed box thinking.

  • Rob Henry October 14, 2012, 18:53

    Eniac, I think your reply to James Jason Wentworth might be misleading, in that you are clearly giving the delta V available for mid course corrections here not potential for delta V if we are examining time spans greater than about ten thousand years.

    OK, I admit that use of repeated close encounters with other stars would strip away the outer planets, but, my guess would be, that such manoeuvres could enable such a civilisation to pass close enough to desirable stars that dozens of new interstellar colonies could be started per million years with no more effort required that would be required than for Earth to establish one on Mars.

  • jkittle October 14, 2012, 22:46

    hey – any lost heat is wasted energy so one might expect a pretty low temperature signature. It occurs to me that there are far fewer brown dwarfs observed than was expected a few years ago when WISE was launched. Maybe they are encapsulated in Dyson style energy capture devices. it must be easier to build them around smaller objects… and there is still PLENTY of energy from a deuterium – burning substellar objects.

  • Abelard Lindsey October 15, 2012, 1:16

    Sure, but where do you suppose to find this material that is 10-100 times stronger than any known material (and in such quantities that you can build a solid sphere around the whole sun)?

    AB matter?


  • Eniac October 15, 2012, 7:09

    @Chris: You are right, of course. I just wanted to point out that unlike most of these far-out megastructures, this one is not completely absurd, physically. If you cannot find the extra order of magnitude in the material (which is likely, as you say), you can adjust the size and the stellar mass to make it possible. I am also intrigued by Brett’s suggestion that a highly reflective surface would give rise to greatly increased light levels once the sphere is closed. This could help, both through the use of light pressure and through an increase in the shell’s radius while maintaining habitable light levels.

    I imagine starting with a very thin shell which floats on the light and is gradually thickened as light pressure increases in the later stages, and then further thickened when the shell is complete and can take compressive load. This would completely address the “incomplete shell” issue.

    If, say a reflectivity of 99% could be achieved, light level upon closure would be increased 100 fold. To maintain an Earth-normal light level at the surface, you need to increase the radius tenfold, which makes diamond plausible as a material right there. If the outer surface is just as reflective as the inner one, the equilibrium temperature would still be 300K, despite the much larger surface area.

    Such a 100-fold increase in size would, of course, make the sourcing of building materials much more difficult. From the Dyson FAQ (http://www.nada.kth.se/~asa/dysonFAQ.html), which you also seem to be referencing (without attribution), it appears that we might find enough carbon for at most a few centimeters, given the 100-fold enlarged shell.

  • Eric October 15, 2012, 10:00

    Note that non-natural candidates have already been spotted in the Kepler data:

  • ljk October 15, 2012, 15:01

    Dr. Hugo de Garis, who is usually ahead of the game when it comes to these matters, has also been promoting femtotech as the technology of choice of the really advanced ETI:


    If nothing else, this and Dyson Shells/Swarms do help to get our minds thinking outside the box when it comes to SETI, which is a good thing because certain relevant ideas and technology that were science fiction when Frank Drake fired up Project Ozma in 1960 are now reality.

    They also keep us from ignoring search realms such as pulsed laser beams, which the Radio SETI faction did for over thirty years even though laser inventor Charles Townes was promoting Optical SETI at the same time as radio and Freeman Dyson’s were coming on the scene. Heck, even Albert Einstein was promoting a form of Optical SETI with the presumed inhabitants of Mars back in 1937!

    The reasons for such self-imposed limitations? Well, besides the fact that a lot of SETI folks were genuinely invested in and supportive of radio waves as the choice of interstellar and intergalactic communications, they also rejected Optical SETI because some of them honestly thought and felt that since it was rather beyond human capabilities that ETI would not be using it either! In reality even sophisticated amateurs have had the ability to detect interstellar laser pulses since at least the early 1970s.

    Read the history here:


    I am sure most would agree that we need to conduct SETI using as many legitimate scientific fields and tools as possible if we want to even have a chance to find other minds out there, in light of the fact that we won’t be exploring the galaxy directly any time soon.

  • johnq October 15, 2012, 15:24

    It appears that the consensus, which I subscribe to, is that a Dyson Swarm makes far more physical sense than a material Dyson Sphere. If so, and I am sure I will hear about it if I error, then it also makes sense to consider how such a collection of objects might come into being — to perform an economic analysis of it. By that I mean the following: how to assign property rights to individual parts of the swarm and their orbits, being able to buy and/or sell both, so we can have a market, one that generates knowledge/capital leading to profit and ultimately customer satisfaction. And loses as well, when things don’t work out in the customer satisfaction department. In other words, the whole creative-destruction thing that is crucial to capitalism.

    And there is the question of collective goods, generation and distribution. The matters of enforcement, broken contracts, receivership — law in other words. I raise these concerns because the alternative, government tax and spend, which it what it always comes down to, raises formidable difficulties not to mention the whole dreary business of what institutions or who will be making the decisions (Hint: It won’t be you or me). The good news is that there is a theory devoted to government decision making and the distribution of collective\public goods (“Public Choice” theory). And this fifty year-old theory complements and builds upon the over a couple of centuries of ever sophisticated market, including free market, analysis. I believe progress could actually be made in addressing these questions now on the economics of a Dyson Swarm.

    All too often interstellar studies, inspiring as they are, reduce themselves to the physics of the business, instead of looking at the questions, to put it bluntly, of how the heck all this star traveling is going to be paid for. I believe payment is a crucial part of the design (it was a big deal for Apollo). I have long thought that one has to have a solar economy up and running (at least a small Dyson Swarm) before true interstellar voyages with real starships (not just probes) can commence. Everyone dwells on the energy required as a crucial factor in the design of the starships, but what about the cost and all that cost entails in a market? That is, how might starships fit into the economy of the time? These trade-offs in design are going to have to be worked out in some fashion. In brief, physics is necessary but it is not sufficient and it would be interesting since we are going to be thinking so much about the former, to be also thinking along the latter (i.e. economic) lines.

    Sadly, there is not much overlap between the two camps. The vast majority of the interstellar community has no interest in economics; moreover, most economists work taxation studies, debt analysis and the like, all very mundane. Few are generalists. Some of them are quite good at the specific subjects that is their life’s work and I respect them for what they do. Most, however, wouldn’t come near a project like this.

    In reading some of the ideas for space institutes and the like, it would be nice if there could be a found a place for economic thinking but getting the two groups to talk to one another, let alone date, is going to be difficult. It is not impossible though.

    Note: Physicist Laurant Nottale has been doing some amazing work on fractal space-time and has developed (among other things) a Schrodinger-like equation for solar systems. In doing so he has discovered a possible quantum of velocity. It could be fascinating if someone applied his work to a Dyson Swarm.

  • Christopher Phoenix October 15, 2012, 20:15

    @Eniac Still, the Dyson Sphere (in its classic SF form, at least) is far, far less plausible than a simple Dyson Swarm- which we could start building in the near future with the construction of the first space-based solar power stations. I did reference the Dyson Sphere FAQ document- I should have mentioned that.

    Assuming your light-pressure idea works, what use do you imagine this sphere fulfilling? Also, what about the hazards posed by meteoroids and comets? Solar flares? If you use a smaller, M-class red dwarf star, you may have to worry about “super flares” much larger than we experience on Earth.

    Another of my favorite megastructures is the laser stars. Artificial stellar lasers can tap into the power of a star, directing a small portion of its output into a direction beam toward a space station or planet to sustain a biosphere or perhaps to propel an interstellar light-sail. Apparently, they will work on binary stars or even dead stars like white dwarfs, unlike Dyson Spheres.

    Laser stars also sound just like something out of E.E. “Doc” Smith’s space fiction, and while the link claims that laser star are limited as weapons, I would not want to find myself on the receiving end of one. That said, it would probably take the beam several minutes to reach an approaching starship, during which time it could attempt to dodge the oncoming laser beam…

  • James Jason Wentworth October 15, 2012, 20:18

    That is what I always liked and admired about Arthur C. Clarke–he was not only a technological visionary, but he also looked in depth into how space hardware and space activities (including lunar colonies) could be of commercial importance. His 1968 non-fiction book “The Promise of Space” is an excellent example of his visionary-*and*-entrepreneurial thinking. Even in his 1957 book about Project Vanguard (“The Making of a Moon,” which had to be hurriedly revised when the first artificial satellite turned out to have been launched by the Soviet Union), he predicted that artificial satellites would soon have great commercial value, particularly for communications.

  • Eniac October 15, 2012, 21:01


    Eniac, I think your reply to James Jason Wentworth might be misleading, in that you are clearly giving the delta V available for mid course corrections here not potential for delta V if we are examining time spans greater than about ten thousand years.

    I am not sure what you are getting at here. From my consideration above it becomes quite clear that it would take millions of years to accelerate to 3 km/s. In 10,000 years, we would barely have started to move towards the nearest star (we’d be going at a rate of 30 m/s), and certainly not reached it. Not even close. Not even far, really….

  • Eniac October 15, 2012, 21:07

    Still, the Dyson Sphere (in its classic SF form, at least) is far, far less plausible than a simple Dyson Swarm

    Yes, of course. However, it is far, far more interesting from a megastructure point of view….

  • ljk October 16, 2012, 9:26

    James Jason Wentworth said on October 15, 2012 at 20:18:

    “Even in his [Clarke’s] 1957 book about Project Vanguard (“The Making of a Moon,” which had to be hurriedly revised when the first artificial satellite turned out to have been launched by the Soviet Union), he predicted that artificial satellites would soon have great commercial value, particularly for communications.”

    I am sorry to say that over 55 years after the launch of Sputnik 1, comsats are still about the only real moneymakers when it comes to space services.

    Hopefully the Planetary Resources company and others will start changing this situation for the better, because otherwise we will not expand into or colonize space and the public will continue to whine about how much the government space programs cost and fail to see their benefits.


    Besides, somebody’s got to get the Sol Dyson Swarm going.

  • Ron S October 16, 2012, 12:56

    Eniac: “However, it is far, far more interesting from a megastructure point of view….”

    Big toys for big boys?

    It always seems to me that a Dyson [insert random polyhedral shape(s) here] is like cavemen sitting around a campfire, and wondering how much better off (and more numerous) they would be if the campfire were larger. It’s better to learn how to make fire so you can have one wherever you are, wherever you desire to be, where it is most useful.

    It’s unlikely that an advanced ETI would construct a Dyson structure because…they’re advanced.

  • Rob Henry October 16, 2012, 14:25

    @Eniac, once well into the process we can use the aid of gravitational assist manoeuvres but that was not really what I was getting at either. If we started afresh, and aimed at the nearest star that our delta V allowed as course correction, then my guess is that it should be at < 100 light years and would take in the order of a hundred thousand years to get to our target closest approach of about 1000au. A mid course correction over that huge distance that would alter our arrival position by just a few au over that journey could set us up to bullseye our next target this time in less than ten thousand years (due to increased choice in targeting). Note that even a correction of 10m/s would move us 2000au from our original path of arrival over a thousand year journey.

    The key is, if we get a million times closer to each target star, than we are on leaving the previous star, then our parabolic orbit with that star will allow us a gain factor in targeting the next star after that (at least in terms of altering angles of trajectories by mid course adjustments within our delta V budget) of about a million also.

  • Christopher Phoenix October 16, 2012, 17:05

    Yes, of course. However, it is far, far more interesting from a megastructure point of view….

    How so? I agree with Ron S- this whole discussion of massive spheres surrounding stars with residential space on the interior does sound a a bit like caveman wondering how much better off they would be if they had a REALLY BIG campfire to huddle around, like, the size of the whole forest outside!! Tapping the power of a star is a fascinating technology, but it is best to explore techniques that could actually work, and these megastructures could resemble the classic Dyson Shell as little as our modern electrical grid resembles a forest-sized campfire.

    An actual advanced spacefaring society will probably be fooling around with megastructures that actually work- like massive space colonies located on planets, orbiting habitats, Asimov Arrays, artificial stellar lasers, etc. We don’t even need to be limited by our ability to tap solar power if we mine the outer planets and comets for fuel, which will be useful for those in the outer solar system.

    The “Dyson Sphere FAQ” actually mentions that if you could build a Dyson Sphere, you would not need to here. The level of super-technology is so ludicrously high, you’d be able to do what you want without bother with Dyson Shells- unless this is just a giant toy for a super-beings amusement, I suppose…

  • Eniac October 16, 2012, 21:18


    The “Dyson Sphere FAQ” actually mentions that if you could build a Dyson Sphere, you would not need to here. The level of super-technology is so ludicrously high,

    You mean this:

    …but since supertechnology is already assumed for building a rigid shell, it seems almost trivial. As somebody put it, if you can build a dyson shell you don’t need it…

    Which is very apt, since my point was exactly that the technology needed is actually not that “ludicrously high”. You can build the 99% reflective sphere at 10 AU with known material, and I outlined a viable process of its assembly. In principle. What remains ludicrously high is the scale, of course. Self-replicating machines could help out in that department, but the mind remains boggled.


    Big toys for big boys?



    An actual advanced spacefaring society will probably be fooling around with megastructures that actually work- like massive space colonies located on planets, orbiting habitats, Asimov Arrays, artificial stellar lasers, etc.

    For exactly the reasons you guys mention, I will (and think I have) completely refrain(ed) from guessing what advanced ETI are actually going to build. Or even what it might be useful for, if anything. These are all just entertaining thought experiments, including the ones you list, and I will let my whimsy guide me in which I consider interesting at any given time…. :-)

  • Eniac October 16, 2012, 21:34

    @Rob Henry: I think you still fail to appreciate what I think I have shown: The acceleration of a star when its energy is redirected will not exceed a few km/s per MILLION years. I think that means it will take about a million years just to approach a neighboring star. You may be right that by proper choice of the angle of incidence you could gain a few more km/s during that encounter, set course to the next suitable star, and proceed in an ever accelerating game of cosmic billiards to wherever you want in the galaxy.

    I am just put off by the three looming dampers to this adventure: 1) The initial million year wait, and 2) the realization that there isn’t really anywhere better in the Galaxy to go to than where we already are, and 3) the probable damage to the solar system during such close encounters.

  • NickCD October 17, 2012, 2:34

    In Kepler data, both a single sphere and swarms evenly distributed around the star are indetectable: there needs to be an assymetric distribution in the swarm or a smaller megastructure to cause changes in the light curve. Mass esimation is also important to identify artificial structures — efficient artificial energy collection and radiation structures seen face-on would show up looking like planets with extraordinarily low density.

    However due to the Malthusian imperative, which applies to all forms of life we have ever observed, we are astronomically unlikely to find anything artificial in the Kepler data or anywhere else in our galaxy outside the work of humans. It is orders of magnitude more productive to look for Kardashev 3 civilizations in other galaxies than type 1 or 2 in our own. I hope this is a sideshow for Marcy or he’s just ruined his career.

    So for all useful purposes the sizes of ETI’s structures are completely irrelevant. Whether they are observable depends not on their singular size but to what extent they collectively form an observable proportion of the observable surfaces in a distinguishable region of the sky.

    The three things we could actually see are

    (1) The ways they dispose of waste energy. Presumably they will be efficient and thus dispose of it in the infrared or perhaps even microwave (if engines that consume infrared energy and radiate in the microwave, harnessing the difference, are plausibly useful). Efficient radiators may or may not be difficult to distinguish from natural dust clouds — it’s a subtle issue.

    (2) The ways they use energy for illumination (broadly speaking, including energy transmission and communications via radiation, as well as illumination proper for seeing). Presumably the most efficient forms of this are monochromatic quantum devices (advanced descendants of our crude lasers and LEDs) but if there are good reasons to suppose other technologies would be better I’m all ears. Such monochromatic quantum devices produce extremely distinguishable and artificial-looking spikes in spectra.

    (3) Surface engineering optimized for thermal and optical properties, again leading to very artificial looking spectra.

    So theorizing about whether a Dyson sphere could really be a single sphere is pointless speculation, but theorizing about the above three points can actually lead us to testable hypotheses, assuming there are any civilizations out there at all to observe.

    More on this in my comment over at Cat Dynamics:


  • Christopher Phoenix October 17, 2012, 8:40


    Which is very apt, since my point was exactly that the technology needed is actually not that “ludicrously high”. You can build the 99% reflective sphere at 10 AU with known material, and I outlined a viable process of its assembly. In principle.

    The Dyson Sphere FAQ discusses rigid Dyson Shells, for which the level of technology is indeed ludicrously high. The light pressure shell is an intriguing idea, but you still have the problems of finding the materials with which to construct it. Also, given the performance of solar sails, it is likely that this light-pressure shell will have to be both high-reflectivity and very low mass, otherwise the light pressure won’t support the sphere against its own weight. A big, thin reflective balloon is not exactly that useful, at least for habitation.

    You know, it might be interesting to do a search for the signatures of artificial stellar lasers amongst Kepler data. If the beam of such a device happened to flash across our view, it would probably be visible- if it happened to be aimed in our direction at the time, which is probably very unlikely. Especially if someone is actually using the beam to power something- still, it is an interesting idea.