Centauri Dreams
Imagining and Planning Interstellar Exploration
A Deeper Dive into the Proxima Centauri Swarm
I’m always interested in how work on interstellar concepts gets funded. After all, although the Nancy Grace Roman telescope is now ready to fly, with a launch some time this fall, there was a real chance the project might get canceled along the way. Trying to predict what will happen to NASA’s budget is harder now than ever. Thus I followed Marshall Eubanks and team’s work on swarm technology missions to Proxima Centauri with interest, learning in their new paper that their NIAC funding continues along with a grant from Breakthrough Starshot’s Communications Group. That last is itself interesting, as communications was, I’ve been told, the toughest nut to crack in setting up swarm strategies for tiny sailcraft – a few grams each – for Proxima Centauri b. Some of this work was performed at the Jet Propulsion Laboratory as well.
Imagine our swarm as consisting of 1000 lightsails launched in a one-month window, boosted by the kind of laser array Breakthrough Starshot has advocated, an Earth-based installation high in the Chilean desert. The research team refers to these sailcraft as ‘coracles,’ a nod to a traditional bowl-shaped boat common to the northern British isles and Ireland. Reaching a velocity of 20 percent of lightspeed, the probes are to be assembled into a coherent swarm using drag from the Interstellar Medium (ISM). At these velocities, this flow of neutral and charged particles can shape them into coherency on the order of 100,000 kilometers transverse separation; i.e., perpendicular to the path of the swarm.
Image: Figure 3 from the paper. Caption: The beta-plane of a swarm flyby of Proxima Centauri b, with the swarm shown lying in that plane. (Note that the planned swarm dispersion is much smaller than is indicated in this artist’s impression, and that in practice the swarm will not be exactly centered on Proxima b’s position due to ephemeris errors.)
The individual probe is currently envisioned at 4 meters in diameter, and on the order of 10 mm thick (aerographene is a leading material candidate). The total probe mass is 3.6 grams, 2.6 of which is allocated to the laser sail. Instrumentation is placed directly on one side of the sail, a phase-coherent array of metamaterial flat optics. As shown below, it contains spaces for 169 smaller 200-mm annular apertures, although not all of these are necessarily used depending on the profile of the mission being flown. These optical apertures when combined produce the light collecting area for a single coracle approximating a 0.5 m telescope.
Image: This is Figure 4 from the paper. Caption: Oblique view of the top/forward of a probe (side facing away from the launch laser) depicting an array of phase coherent apertures for both imaging and for sending data back to Earth. Credit: Eubanks et al.
An earlier Centauri Dreams article Reaching Proxima b: The Beauty of the Swarm gives background particulars, but the concept is now being brought forward with a great deal more detail. Swarm concepts are useful because the high number of probes heightens the chances that some of the probes may move past both sides of the target for maximum coverage. It’s noteworthy that the authors, taking into account launch as well as voyage and encounter losses, assume only 300 of the original 1000 will be left for communications back to Earth. As we’ll see, some of the probes are to be ‘sacrificed’ as they serve the communications needs of the mission.
Working with the Medium
But let’s get back to the question of the interstellar medium. Each of the probes is to rotate 90 degrees at the end of the boost phase, the idea being to reduce erosion during the cruise phase by traveling edge-on. We have to get through the comparatively dense interplanetary zone before exiting into interstellar space – here it’s interesting that given the direction of Proxima Centauri from the ‘nose’ of the heliosphere, the movement through the heliopause should occur at roughly the same distances experienced by the Voyagers – 125.6 and 119 AU. We’re moving, of course, considerably faster, and at .20 c, exit the Solar System in less than four days. It took Voyager 2 41 years to make this passage. From the paper:
It is not possible to increase speeds with drag from the ISM wind caused by the probe’s velocity; we use ISM drag to implement a velocity on target technique, slowing down the later launched probes so that velocities come to match as probes approach each other. Once the solar system risk zone is passed this technique will be initiated by rotating swarm members into a “face-down” sail-side up configuration, increasing the drag by having the sail-side face into the ISM wind. In the face-down configuration, the main communications lasers on the instrumentation side will be facing the Earth, enabling high-bit-rate communications without exposing the instrumentation and electronics to ISM wind damage. Note that the first probe launched will not have to enter a facedown configuration, and the later launched probes will advance to join it.
So we have differential thrust between edge-on probes and face-on probes, the result being a swarm that is gradually assembled over 2.79 years. Remember that the plan is to boost the entire swarm into space in a period of no more than one month. The swarm begins to coalesce after launch because the launch velocity of each new probe is increased, allowing later-launched probes to catch up with earlier ones. The probes all return to an edge-on configuration after swarm assembly, coordinating communications through six lasers per probe.
Image: Leaving the heliosphere, we move into the interstellar medium’s gas, plasma, dust, cosmic rays, and magnetic fields. Can we use this ‘interstellar wind’ to shape the Proxima Centauri swarm? Credit: JHU/APL.
I mentioned above the attrition of the swarm along the route, which is not entirely due to encounters with material in the ISM. The authors also turn individual probes into a face-down configuration to manage data communications with Earth, creating higher drag that pulls them out of the swarm. Meanwhile, the 30 percent of the swarm thought to be remaining at the Proxima Centauri system can target Proxima Centauri b or break into sub-swarms, perhaps targeting other planets in the system. One week before the encounter, the first probes will rotate their instrument side into the forward direction of motion, relaying observations to the rest of the swarm. The entire swarm will go face-down after the encounter for relaying data to Earth. The data return phase is assumed to require no less than a year.
Bringing the Data Home
The communications problem vexed Breakthrough Starshot designers, so the solution posed here catches the eye. Among the options are having probes return data independently or, far better, creating a time-coherent swarm which sends communications pulses that arrive at Earth simultaneously. More challenging but perhaps the most worthy of future study is to create a sparse phased array for communication, one that allows swarm antennas to act as a single higher-gain antenna. The thin, ultra-lightweight optical elements are phase-locked to achieve a synthetic aperture of considerable size, but one that demands maintaining probe positions at the nanometer level. From the paper:
The advantage of this latter approach is that the synthesized beam pattern in the main lobe at Earth is equivalent to that of the single transmission reflector with area equal to the sum of the areas of all the probes, although this would be a sparse array and the beam shape would not be the same as the beam formed by a solid antenna with the same extent. Note that this approach would require maintaining the positions of the probe members at the few 100 nm level or better, roughly 6 orders of magnitude better than the time coherent swarm approach. We do not consider this last sparse phased array approach further in this paper due to the extreme difficulty of phase coordination across the swarm.
Image: This is Figure 2 from the paper. Caption: Artist’s impression of a Coracle approaching Proxima b (and reflecting the light of Proxima Centauri). The 12,000nm intra-swarm “Side Lasers” (see Subsection 6.3) are for intra-swarm probe-to-probe communications. Each round ring on the top (instrumentation) side of the sail visible here is the 200 mm annulus aperture of a folded optic camera (see Figure 6 and discussion) shared between imaging and communications with Earth at 432/539-nm. Conceptual artwork by Mark Garlick. (Note: Seeing other probes apparently nearby at encounter is artistic license!)
Data broker ‘agents’ can be used to filter and select data from the many terabytes collected during the flyby, managing the data return to Earth. In this way redundant data can be filtered out of the data flood, using what the authors call Observe-Evaluate-Select-Flood (OESF) loops, in which the swarm is essentially divided into nested sets of probes. This part of the concept deserves more attention than I can give it here, but it’s essentially applying an AI approach not only to managing collected data but also to analyzing imagery for further consideration. Even so, this statement pulled me up short:
Although the techniques of developing swarm coherence and agent-based data selection certainly require work, there seems to be no fundamental limitation to the return of gigabytes of data over interstellar distances with large swarms of laser-sail spacecraft.
I believe the statement is true insofar as we can come up with a solution consonant with physics to make this happen, but gigabytes of data with this particular mission concept seems too much to hope for. That’s the judgment of a layman, however, and it will be fascinating to see how these communications concepts play out in the literature as this project continues to be refined. The concepts here are ingenious, even startling, and deserve further investigation.
Moving into the Proxima Centauri System
The prospect of instrumentation in the Proxima Centauri system is exciting indeed. Given the number of probes entering this zone, the authors believe at least one is likely to pass within a single diameter of Proxima b, which would provide spectroscopic analysis of the planet’s atmosphere as well as imaging in considerable detail. Mapping of the surface on the day side of the planet would allow us to search for the so-called ‘vegetation red edge’ and any biology there. The search for biosignatures and technosignatures could get down to the level of features like coral reefs or even night-time city lights.
High-velocity flybys pose huge imaging challenges, given the needed length of exposure time and the movement of the planet in the field of view. The result: enormous image smear. To attack the problem, the authors point to Time Delay Integration (TDI), Velocity Shift Integration (VSI) and high dynamic range imaging (HDR), three techniques explained in the paper. The close flyby of Proxima b itself will last less than a minute. Note the ramifications of this not only on data return but the necessary computational resources of the swarm:
In 0.01 s the spacecraft would move ∼600 km, which, at a distance of 10,000 km… would cause noticeable distortions of the images being stacked; these are predictable and can be removed. Iterative HDR can remove rotations of the spacecraft during the image, correct for ephemeris errors during imaging, and also correct smearing due to objects with different relative velocities in the image plane. In a 10 second flyby with 106 mega-pixel images per second per aperture a single probe with multiple aperture arrays might obtain billions of images, mostly greatly underexposed. This will form the raw material for searches for small bodies and unanticipated features in the Proxima system. It will never be possible to send all of this raw material back to Earth; extracting as much useful information as possible from it after the encounter will be a major computational task for the probes in the swarm.
Image: This is Figure 1 from the paper. Caption: Artist’s impression of the approach of a swarm towards Proxima b; at this point, a few seconds before closest approach, the swarm could be examining the planet’s nightside for techno- or bioluminescence. (This image is based on the artistic work of Dr. Mark A. Garlick.)
Orienting the probes after the system flythrough to communicate with Earth, the swarm will be able to observe the Proxima system as it recedes and observe the interactions of the star’s heliosphere with its local interstellar medium (and recall the New Horizons imagery of Pluto after that spacecraft’s encounter). Moreover, a distant encounter with Proxima A and B will occur about a year after the Proxima Centauri event, although the approach as conceived here would be on the order of 10,000 AU. Planets in the habitable zone of both stars should be observable from this distance. Much better, of course, to have a separate Centauri AB flyby mission, but for now one system at a time.
Navigation will be difficult given that we need highly accurate ephemeris information – in other words, we have to know exactly where Proxima Centauri b is, an obvious point, but it’s problematic because given current data from Gaia, the possible error in the star’s proper motion amounts to a 260,000 kilometer error over the mission’s flight time. A better determination of Proxima b’s orbit is also critical, which is why the authors consider a possible precursor mission several years before the first swarm mission to improve the ephemeris.
I won’t list all the authors of this paper but many will be familiar to Centauri Dreams readers, including Jean Schneider and Pierre Kervella (Paris Observatory), Andreas Hein (I4IS/University of Luxembourg), Robert Kennedy (I4IS), Slava Turyshev (JPL) and Philip Lubin (UC-Santa Barbara). The kind of investigation mounted by this team is how we move the ball forward in interstellar studies. Drawing on recent work including the deep investigations of the Breakthrough Starshot scientists, Eubanks and colleagues have enlarged the speculative space especially in terms of communications and swarm computational options, all making an interstellar crossing in decades rather than centuries possible. This paper should be studied by anyone seriously following our increasingly refined strategies for making such a crossing happen.
The paper is Eubanks et al., “Science from the In Situ Exploration of the Proxima Centauri System,” available as a preprint.
Moving a Civilization: The Caplan Thruster
Because I’ve been talking about enormous structures lately and describing them as ‘big dumb objects,’ I thought it would be fun to revisit the origin of that term. BDOs emerged in what was intended as an April Fool’s joke by writer and critic Peter Nicholls, famed as editor (with John Clute) of The Encyclopedia of Science Fiction, now online in its fourth edition. He describes the genesis of the term in a well known essay called “Big Dumb Objects and Cosmic Enigmas: The Love Affair between Space Fiction and the Transcendental”:
“All these matters were in the forefront of my mind when I came to revise The Encyclopedia of Science Fiction, a task in which my primary responsibility was to rewrite and rethink all those entries dealing with the themes of science fiction. This brings us to April Fool’s Day, 1992, that being a day in which practical jokes are traditionally carried out. On that day I was exhausted writing theme entries, and my brain was hurting. So I decided to write an April Fool’s entry. I would pretend that a phrase that I’d always liked, originated by the critic Roz Kaveney but not in general use, was actually a known critical term. I would write an entry called “Big Dumb Objects” in a poker-faced style, suggesting an even more absurd critical term to be used in its place, “megalotropic sf…”
Image: The cover of the first edition of Greg Bear’s novel Eon (1985), which describes a huge, terraformed asteroid that enters the Solar System. This is one of the Big Dumb Object novels Nicholls discusses in his formative essay.
Nicholls soon realized that vast structures were symptomatic of what makes the best science fiction operate, and he relates them to the “…tension between the writer’s respect for and understanding of orderly scientific thought (the classical) and his love for the phenomena which do not submit to this order (the romantic).” If that seems a stretch, read the essay, where he points out that ‘hard’ science fiction, with its adherence to the laws of physics, can inspire in its ringworlds and Dyson spheres and Ramas a deep Dionysian mystery, a sense of the sublime that we can easily relate to the familiar ‘sense of wonder.’ I like Nicholls’ reference to the rituals of Dyonisus, with their ecstaties and trances.
In our talk about stability and BDOs, we home in on the practical matter of whether or not they could actually be built, but again, the laws of physics imply this is an engineering problem that an advanced civilization could well master. It could be, of course, that a Big ‘Dumb’ Object isn’t really so dumb if it needs a constant technological assist to survive, but Colin McInnes, whose essay on stellar engines we examined last week, has also produced a paper covering ringworlds and Dyson spheres that finds modes of stability even there. I’ll give that citation below, and thank Dr. McInnes for his kind note with the reference. So maybe we need another term: ‘Big Smart Objects’?
For today, let’s segue to a relatively new entry in the stellar engine portfolio, as developed by Illinois State University’s Matthew Caplan. Unlike a Dyson sphere or swarm, a stellar engine produces a change in its star’s position, small enough that a planetary system is not disrupted, but large enough that over millions of years, the star’s galactic orbit can be modified. Speculating about what alien civilizations might do takes us deep into the weeds of philosophy and epistemology, an exercise best left for future posts. But let’s take one possibility that seems rational, escaping from one or more nearby supernovae. Thus Caplan:
…ozone depletion in the earth’s atmosphere due [to] ultraviolet radiation from a supernova within 10–100 pc may result in a mass extinction event. Amusingly, mounting evidence for one or several nearby supernova (100 pc) approximately 2 million years ago now forms the basis for recent suggestions that nearby supernova caused climatic shifts which directly influenced human evolution. The effect of a supernova on an exoplanetary biosphere inhabited by an advanced civilization will depend on that planet’s atmospheric composition and biosphere, and may be very different from earth, possibly extending the danger zone of supernova by an order of magnitude relative to earth. A catastrophe such as a supernova could likely be predicted millions of years in advance, at minimum, for an advanced civilization with detailed understanding of star formation and the supernova mechanism.
Of course, when dealing with supernovae, it’s best to move as swiftly as possible. Previous stellar engine ideas have resulted in movement of 10 pc per galactic orbit, the latter being in the range of 225 to 250 million years. It also makes sense to move in a retrograde motion relative to the galactic orbit, which provides maximum exposure to other star systems during the trip. Caplan’s summary of previous stellar engines in the literature is useful, and goes back to Fritz Zwicky’s ideas on inducing a jet in a star through particle beams. And here we can distinguish between ‘passive’ thrusters that operate without intervention (the Shkadov thruster, for example) and ‘active’ thrusters that become, in Caplan’s terms, something like a tug pushing the star through the galaxy.
Image: This is Caplan’s Figure 2. Caption: Artist’s rendition of an operational active thruster around a star with a Dyson swarm, where the solar wind is collected by an engine which drives a jet of exhaust. The sun and ramjet accelerate to the left. Credit: Michelle Buhrmann.
Caplan analyzes both passive and active thrusters, the passive design being essentially the ‘solar sail’ configuration used by Shkadov. Here we can manage 10-12 m/s2 working with a G-class star like the Sun, and as noted last time, this amounts to 20 meters per second after a million years, or 0.03 light-years from its original position. What Caplan manages to do to improve this is to deploy matter collected from the star after heating part of the solar ‘surface’ using a mirror swarm (think Dyson swarm here). This material is then used to fuel fusion reactors, with the result being to achieve speeds a thousand times faster than the passive design. Now we can move the star 50 light-years in a relatively swifter one million years. A civilization stable enough to survive through entire epochs like this might see this as a viable approach.
The plan here is somewhat reminiscent of Benford and Niven’s, at least insofar as it uses a focused beam to disrupt the stellar surface and produce an ejection of helium and hydrogen. But whereas Caplan will use that ejection to fuel fusion reactors, Benford and Niven take a different approach that I’ll cite here for the interest of the comparison. This is from Greg’s afterword to the novel Shipstar (see also his “Building the Bowl of Heaven,” which he wrote for Centauri Dreams in 2014):
There’ve been several Big Dumb Objects in sf, but as far as I know, no smart ones. Our Big Smart Object is larger than Ringworld and is going somewhere, using an entire star as its engine…Our Bowl is a shell more than a hundred million miles across, held to a star by gravity and some electrodynamic forces. The star produces a long jet of hot gas, which is magnetically confined so well it spears through a hole at the crown of the cup-shaped shell. This jet propels the entire system forward – literally, a star turned into the engine of a “ship” that is the shell, the Bowl. On the shell’s inner face, a sprawling civilization dwells.
As you can see, Caplan’s tack is entirely different. Benford and Niven use the reflected light from the Bowl’s surface to produce the disruption on the stellar surface that creates the resultant beam. Lacking a huge reflecting object like the Bowl, Caplan considers the use of a Dyson swarm to produce the needed energy for the essential mass-lifting. From the paper:
Alternatively, such mass lifting may be possible at very high efficiency using similar principles to concentrated solar power. Reflecting large amounts of sunlight directly to one spot or small region of the sun’s surface (perhaps with statite mirrors like those described above) will locally increase the temperature and mass loss rate. Physically, the mirror reduces the area over which the sun radiates and drives up the surface temperature by the Stefan-Boltzmann law. Similar radiatively driven mass loss is believed to occur in Wolf-Rayet stars [29].
In other words, a small change in temperature produces massive energies. My math skills weren’t up to the challenge, but I wondered how long a thruster like this could function before changing the star’s classification along the Hertzsprung/Russell diagram. So I put the matter to Google’s Gemini AI, which produced a figure of 12.6 million years to turn the Sun into a K-dwarf like Centauri B. Keep in mind that we’re burning millions of tons of stellar material every second using the Caplan thruster. That also increases the Sun’s lifetime as it begins to burn its hydrogen at a slower rate. Thus a stellar engine becomes a way to keep a star burning far longer than its earlier classification would allow.
Image: This is Figure 3 in the Caplan paper. Caption: Schematic of an active thruster. Solar wind is collected by large scale electric or magnetic fields, which funnel matter into the engine. H and He are separated, with an He fuel mixture being used to drive a high velocity jet of exhaust away from the sun and out of the solar system, while H is returned to the sun using traditional electromagnetic accelerators, transferring exhaust momentum to the sun. The sun and engine accelerate to the left. Credit: Michelle Buhrmann.
We’re now in the velocity range that Caplan considers sufficient to achieve a retrograde galactic orbit or even galactic escape velocity. So here’s a science fictional thought. Stars now known to be on such trajectories might be candidates for SETI observations given the possibility of mega-scale engineering. He continues (the italics below are mine):
We therefore argue that hypervelocity stars on escape trajectories from the galaxy may be observable candidates for detecting megastructures, even though the operation timescales of stellar engines are short relative to intergalactic flight times. Recent work suggests that known hypervelocity stars may be traveling above the upper limit for classical ejection methods. Such stars may be candidates for detecting megastructures if additional powerful dynamical ejection methods are ruled out.
The Caplan paper is “Stellar engines: Design considerations for maximizing acceleration,” Acta Astronautica Vol. 165 (December, 2019), pp. 96-104 (full text). The Colin McInnes paper mentioned above is “Ringworlds and Dyson spheres can be stable,” Monthly Notices of the Royal Astronomical Society Vol. 537, Issue 2 (29 January 2025), 1249-1267 (full text).
Stability of Interstellar ‘Big Dumb Objects’
‘Big dumb objects’ (BDOs) appear to great effect in science fiction. They come in all manner of sizes and shapes and they fulfill a wide range of functions. An early favorite of mine was Cordwainer Smith’s “Golden the Ships Were Oh! Oh! Oh!,” which I snagged on a long ago trip to a Chicago newsstand, where it appeared in an issue of Amazing Stories. It’s probably found most easily these days in The Rediscovery of Man: The Complete Short Science Fiction of Cordwainer Smith (NESFA Press, 1993), a collection that should be on every science fiction fan’s shelf.
Smith (a pseudonym for Paul Myron Anthony Linebarger, whose life was as remarkable as his fiction) goes to work on structures that are millions of miles long. I won’t say more for fear of spoiling the story for newcomers. More recent BDOs are better known, Dyson spheres and Dyson swarms are no strangers to these pages, and have been the subject of intense scrutiny by Jason Wright and his colleagues at Pennsylvania State University. The G-HAT (Glimpsing Heat from Alien Technologies) project scanned data from the Wide-field Infrared Survey Explorer satellite looking at tens of thousands of galaxies for the waste heat signature of possible Dyson spheres. The idea that megastructures might interest a hugely advanced civilization is reasonable, but we have yet to find evidence that Dyson spheres exist.
Larry Niven’s Ringworld posits a structure that circles an entire star but does not encompass it. A transit signature might give this one away if ever found; imagine the lightcurve. Niven and Gregory Benford later come up with the ‘shipstar’ concept that Greg described some years back on Centauri Dreams. This was an unusual re-thinking of the original ‘Shkadov Thruster,’ a device that could be used to move an entire star. See the Bowl of Heaven trilogy for more.
The work of Russian physicist Leonid Shkadov in 1987, the thruster design used asymmetric light pressure from a huge mirror to move an entire planetary system to a new destination. The physics works, but we’re moving at slow speeds, on the order of 20 meters per second after a million years. On the other hand, a truly long-lived species might find waiting a billion years to reach 20 kilometers per second, with a whopping 34,000 light years shift in position, to be plausible. Shipstar would be able to move considerably faster.
Image: An artist’s conception of the Benford/Niven ‘shipstar’ concept. Think of the ‘bowl’ as half of a Dyson sphere curved around a star whose energies flow into a propulsive plasma jet that moves the entire structure on its journey. Here the notion of living space may remind you of Niven’s Ringworld, that vast structure completely encircling a star, though not enclosing it. The difference is that in the ShipStar scenario, most of the ‘bowl’ is made up of mirrors, with living space just on the rim. Credit: Don Davis.
In conversations with Benford about his shipstar concept a few years ago, I learned that a solid Dyson sphere is unstable, and would need constant adjustment to maintain its position. Concerns over stability plague BDOs. Colin McInnes (University of Glasgow) looks at the problem in a recent paper, noting this about the Shkadov design:
In its simplest form a stellar engine can be considered as a single ideal ultra-large rigid reflective disc in static equilibrium above a central star… As the disc accelerates due to radiation pressure from the star, the centre-of-mass of the gravitationally coupled star-reflector system accelerates, leading to a displacement of the star.
Image: This is Figure 1 from a paper by Duncan Forgan (citation below). Caption: Diagram of a Class A Stellar Engine, or Shkadov thruster. The star is viewed from the pole – the thruster is a spherical arc mirror (solid line), spanning a sector of total angular extent 2ψ. This produces an imbalance in the radiation pressure force produced by the star, resulting in a net thrust in the direction of the arrow. Credit: Duncan Forgan.
That seems straightforward, assuming a civilization so advanced that it could build mirror structures of the needed size. Here too, though, we have stability problems. The McInnes paper is highly interesting, examining megastructure concepts and the possible ways of stabilizing them. While a uniform, rigid reflective disk proves unstable as a star-moving engine, a disk with its mass concentrated at the edges can be stable. Instead of a flat disk, we are looking at something much closer to the shape of a ring. Here passive stability is what we want – i.e., the object does not need continual adjustment by other technologies to maintain its position and function.
In the case of the Schkadov engine, we have this consideration:
…for an ideal reflector subject to gravitational and radiation pressure forces the gradient of these forces across the reflector will induce stresses. While the direction of the radiation pressure force is always normal to the reflector, the direction of the gravitational force will vary across the reflector moving from the centre to the edge. Therefore, while the component of the gravitational force normal to the reflector can in principle be balanced by the radiation pressure force, there will be an in-plane component of the gravitational force which will generate a compressive stress. A thin reflector will clearly be unable to support such compression. However, in principle a zero-stress reflector can be configured for a non-homogeneous, partially reflecting rotating reflector…
The math for a stellar reflector and a stellar ring are laid out in the paper’s appendices.
McInnes thinks that stability is useful as we investigate possible technosignatures in our SETI work, whether they be star-moving thrusters or energy-gathering Dyson objects. The assumption is that passive stability will be sought after because it is efficient and economical, not requiring control systems that must continually adjust position. Remember, too, that in searching for technosignatures, we have the possibility of finding megastructures like these that have survived the demise of their creators. Passive stability is essential for these objects to remain intact and detectable.
What McInnes calls a ‘Dyson bubble’ can likewise be stabilized. Here we’re talking not about a solid Dyson sphere but a constellation of discs, a ‘power swarm’ that allows a civilization to exploit most of the output of its star. The terminology can be confusing but bear with me. The author distinguishes between a cloud of small reflectors in orbit around the central star – huge in number, these form a so-called ‘Dyson swarm’ – and a ‘Dyson bubble,’ by which he means a smaller number of large reflectors in ‘statite’ configuration, so that instead of orbiting, radiation pressure exactly balances gravity. In other words, the ‘bubble’’ components stay stationary relative to the star.
Self-stabilizing techniques are challenged not only by gravitational and radiation pressure but also collisions between the myriad orbiting disks as well as outside perturbing forces. Over large timeframes, passing stars can disrupt the gravitational dance, while interstellar comets, whose numbers are likely to be huge, present a similar risk of disruption. Even so, there are ways around this:
…the Dyson bubble can remain stable when its self-gravity and a simple model of a diffuse background of scattered radiation are included in the dynamics defined in Section 6.4. However, there are now regions of the parameter space where instability can occur, primarily at the edge of the Dyson bubble driven by the diffuse background radiation. In addition, it has been shown that the self-gravity of the Dyson bubble is in itself sufficient to ensure passive stability in the absence of the diffuse background radiation, and indeed it enhances the stability of the Dyson bubble when the diffuse background of scattered radiation is included.
A Dyson swarm if properly implemented can also ensure passive stability. Reflectors must always be configured ‘normal’ (perpendicular) to the central star “…using slighting conical reflectors with the centre-of-pressure displaced behind the centre-of-mass.”
So there are ways of doing these things as long as we abandon the Shkadov concept of a uniform reflector disc in favor of a ring supporting the reflector, or in the case of the two Dyson options McInnes looks at, a dense cloud of reflectors stabilized through orbital mechanics, or a smaller assembly of reflectors in static equilibrium with radiation pressure from the star exactly balancing gravity. But here I’m more interested in the consequences in terms of hunting for technosignatures:
A Dyson swarm can be expected to generate a different technosignature to a passively stable Dyson bubble discussed above. For example, the motion of the discs in a swarm would imply a flickering of the observed luminosity of the central star, with a larger variation expected from a small number of ultra-large discs relative to a large number of small discs. Finally, while an orbiting swarm of reflectors will be susceptible to collisions (B. C. Laki 2025), collisions within a Dyson swarm could in principle be minimised using families of displaced non-Keplerian orbits, where the orbit planes of the reflectors can be stacked in parallel rather than being inclined relative to each other (C. R. McInnes & J. F. L. Simmons 1992).
And what of Shipstar? A recent conversation with Jim Benford reminded me that his brother Greg had worked out a way to stabilize the induced flare on the central star through intense magnetic fields, but as far as I know, this concept has never been rigorously investigated. From the technosignature standpoint, McInnes’ paper reminds us that stability problems can be overcome should an advanced civilization choose to build Dyson-class structures, or undertake star-moving of the Shkadov variety. How to engineer the stability of BDOs should continue to provide insight into possible technosignatures, even if the lack of any trace of Dyson structures despite intensive work at G-HAT remains puzzling. Next week I want to look at an even more recent stellar engine concept as presented by Illinois State University’s Michael Caplan.
The paper is McInnes, “Stellar engines and Dyson bubbles can be stable,” Monthly Notices of the Royal Astronomical Society 546 (2026), 1-18 (full text). The Shkadov paper is “Possibility of Controlling Solar System Motion in the Galaxy,” presented at the 38th Congress of the International Astronautical Federation (IAF) in Brighton, UK. An English translation of the original paper was published in the Journal of Solar System Research Volume 22, Issue 4, pp 210–214 under the title “Possibility of Control of Galactic Motion of the Solar System.” The Forgan paper mentioned above is “On the Possibility of Detecting Class A Stellar Engines Using Exoplanet Transit Curves,” Journal of the British Interplanetary Society, Vol. 66, no. 5/6, 2013 pp. 144–154. Preprint.
No Signs of Atmosphere on TRAPPIST-1 b or c
Waiting to learn what next generation telescopes will reveal is tantalizing in the extreme. In terms of space-based instruments, we’re getting close to launch of the Nancy Grace Roman Space Telescope, which has been the subject of many posts here under its former name WFIRST (Wide-Field Infrared Survey Telescope). Part of its remit will be to image nearby planetary systems, assuming it can survive NASA budget battles that have threatened to cancel it. Launch could occur late this year if these issues are resolved.
Needless to say, the European Space Agency’s PLATO mission (Planetary Transits and Oscillations of Stars), with a 2026 launch expected, has my full attention. Here we have a focus on terrestrial exoplanets in the habitable zones of their stars, to be followed up with ESA’s Ariel (Atmospheric Remote-sensing Infrared Exoplanet Large-survey), designed within a few years to be launched for the study of planetary atmospheres. On the ground, the European Southern Observatory’s work on its 39-metre instrument continues, with first light projected for 2029 and regular observations beginning the following year.
Meanwhile, the James Webb Space Telescope continues to deliver outstanding results. The latest to catch my eye involve the TRAPPIST-1 system, with its seven terrestrial-sized planets orbiting an M-dwarf in Aquarius. At about 40 light years out, this system is close enough to reward intense scrutiny, especially since all seven planets transit the star. In new work just published in Nature Astronomy, we get our first look at planetary atmospheres – or the lack of same – on the two inner worlds, TRAPPIST-1b and TRAPPIST-1c.
Image: This artist’s impression displays TRAPPIST-1 and its planets reflected in a surface. The potential for water on each of the worlds is also represented by the frost, water pools, and steam surrounding the scene. Credit: © NASA/R. Hurt/T. Pyle.
The question of atmospheres is a fraught one given that the tight habitable zones around an M-dwarf mean that planets there are subject to violent flare activity that can potentially strip an atmosphere entirely. The two inner worlds are not in the habitable zone (TRAPPIST 1-e, f and g are, but are not part of this study). We learn in the paper that no atmospheres can be detected here, but the question of the other planets remains open. This is the first time that astronomers have mapped climate features on Earth-sized planets.
We can continue to speculate on tidal lock, which will be a factor on planets in the habitable zone of any red dwarf star. A permanent day on one side, permanent night on the other are the result, but there are mechanisms that could keep a planet like this able to sustain life. Brice-Oliver Demory (University of Bern), a co-author of the study, comments on the importance of the work:
“The presence of an atmosphere around these tidally locked planets could allow for energy transfer between the day and night sides, resulting in more moderate temperatures across the planet, which would have a significant impact on their potential habitability. Successfully detecting the atmosphere of one of these planets has therefore become a key objective for our community, highlighting the importance of the TRAPPIST-1 system with the JWST.”
Sixty hours of observation with JWST tracked the two inner planets in the infrared through a full orbit, allowing readings of surface temperature to a high degree of precision. What tells the story is the marked temperature contrast between night and day sides, with the inner TRAPPIST-1b at 200 degrees C on the dayside, while planet c comes in at 100 degrees C. The night side of each registers at below -200 degrees C, indicating that thermal energy is not being transferred, a likely consequence of early atmospheres being stripped away.
How far out from the star do we have to go to find a surviving atmosphere? Emeline Belmont (University of Geneva) points to our own Solar System as reason for optimism. Whereas Mercury has been stripped of any atmosphere, both Venus and Earth clearly had no problem forming and keeping their own. That would leave the three TRAPPIST-1 worlds in the habitable zone continuing candidates for follow-up, and eventually spectroscopic study of atmospheric components. Will a future telescope register a biosignature on one of these?
We can expect the investigation of TRAPPIST-1 to accelerate. Out of curiosity I ran a quick check on the Astrophysics Data System (ADS), requesting papers with TRAPPIST-1 in their abstracts published since the beginning of this year. 36 entries came up, some of them only referencing the system, but most homing in on various issues involving it. Today’s paper particularly caught my eye given lead author Michaël Gillon (University of Liège), who led the international team that discovered the system in 2016 and subsequently identified its full extent.
The paper is Gillon et al., “No thick atmosphere around TRAPPIST-1 b and c from JWST thermal phase curves,” Nature Astronomy 3 April 2026 (abstract).
Interstellar Probes: Moving Beyond Bracewell
Lately we’ve been discussing interstellar probes, the kind that an extraterrestrial civilization might use to explore the galaxy. Ronald Bracewell’s analysis of such probes dates back to 1960 and was all but coterminous with the emergence of SETI. The problem with Bracewell probes is that we would expect to have one in our Solar System if they exist. Rather than using that notion to add stress to the Fermi question, I’m going to point out that there is a lot of real estate waiting to be searched.
Case in point: What might our ongoing study of the lunar surface through images from the Lunar Reconnaissance Orbiter pick up as we use AI models that have already identified human-made space debris from various missions? A closer look at this project reminds us that while the Moon is an obvious place to look for a ‘lurker’ probe, we can’t discount other locations even though earlier work on the various Lagrange points, a good place for long-term observation of our planet, came up empty (see below). Our capabilities are so much more advanced not only in terms of instrumentation but analytical tools that a continued hunt for artifacts is reasonable.
I’m getting picky here given the wide variety of possible probes, tapping the definition that Bracewell used in his original article. That’s a probe we probably would have noticed by now if it were active. In 1960, Bracewell was offering an alternative to the SETI goal of detecting an interstellar radio signal aimed at Earth. His physical probe would arrive in a planetary system to look for signs of life and technology, duplicating any radio signals it heard so as to re-transmit them to the originators, thus establishing contact. Sagan uses the notion in his novel Contact (1979), where Adolf Hitler’s opening speech from the 1936 Berlin Olympics is found embedded within the message, along with much else.
How would we respond to hearing a signal sent back to us from space? Bracewell thinks we would experiment with it to see what would happen next:
To notify the probe that we had heard it, we would repeat back to it once a:gain. It would then know that it was in touch with us. After some routine tests to guard against accident, and to test our sensitivity and band-width, it would begin its message, with further occasional interrogation to ensure that it had not set below our horizon. Should we be surprised if the beginning of its message were a television image of a constellation?
Bracewell’s notions of dispatching a physical object as opposed to sending a radio signal take advantage of the ‘information density’ available to a physical probe. This is the familiar notion that a box of DVDs in a truck moves information at a far higher rate than fiber-optic cable. But of course you have to get the truck to its destination, and in the case of interstellar flight the latency is huge – perhaps thousands of years or more. A long-lived civilization, thought Bracewell, may nonetheless see purpose in seeding nearby stars if the travel time is a small fraction of its likely civilizational life.
Swarming and Reproducing
Bracewell’s ideas jibe nicely with the Breakthrough Starshot concept of swarms of sails investigating nearby stars. We might imagine the descendants of such tiny flyby probes scattered to all interesting stellar systems within, say, 100 light years. With concepts like Bracewell’s entering the literature, it was left to Robert Freitas to run the first scientific search I am aware of for such probes (citation below). Freitas made a series of visual observations of the various LaGrange points in the early 1980s. But in the early days of SETI (and Bracewell was writing even before the Green Bank meeting in 1961 that produced the Drake Equation), other ideas about how interstellar probes might operate had begun to surface. Ancient probes sent by civilizations far more advanced than ours might still be live, waiting and reporting on our activities (Clarke’s sentinel ‘slabs’ from 2001: A Space Odyssey come to mind . Or they might be long-dead relics.
Version 1.0.0
When Michael Hart went to work on this in 1975, he amplified the probe concept and changed the game. He produced, in fact, what Jason Wright (Pennsylvania State) has dubbed “The most influential formulation of the Fermi Paradox…,” one that compresses the conundrum by homing in on the fact that we observe no intelligent beings on our planet, something Hart called Fact A. The fact that they are not observed tells us that despite the amount of time available for long-lived cultures to have colonized the galaxy, none evidently have. This is no small problem, for as Wright calculates in his new textbook on SETI, even a ‘wavefront’ of probes moving outwards from star to star at Voyager-like speeds would have been able to reach every star within 2 billion years.
Move the dial up in terms of speed to, say, 0.5 c and the numbers get considerably shortened. Imagine relativistic ships that close on lightspeed and we find exponential growth saturating the galaxy in 150,000 years, all contrasting with an Earth that is 4.5 billion years old. Hart saw nothing in the laws of physics that prohibited starflight, and he found the idea that ETI was uninterested in Earth to be unconvincing. What David Brin coined the ‘Principle of non-Exclusiveness’ boils down to the idea that alien species will not all behave the same way. All that is needed is for one civilization to decide to send out probes, and by now such probes should have reached every star.
Image: How quickly would a single civilization using self-replicating probes spread through a galaxy like this one (M 74)? Moreover, what sort of factors might govern this ‘percolation’ of intelligence through the spiral? The answers affect our view of the Fermi question, and thus our own place in the cosmos. Image credit: NASA, ESA, and the Hubble Heritage (STScI/AURA)-ESA/Hubble Collaboration.
Advances in computing led Frank Tipler to push Hart’s views even more strenuously, bringing John von Neumann’s work on self-replicating machines to bear. His insight was to ask what would happen if an extraterrestrial culture began seeding stars with self-reproducing probes, each capable of not only studying a new world but building another probe that could reach yet another star, and so on. Here the numbers become even more telling. Such probes could use local resources in each system to build their next generation, thus nullifying the resource problem. Here’s Tipler on the matter:
…if the motivation for communication is to exchange information with another intelligent species, then as Bracewell has pointed out, contact via space probe has several advantages over radio waves. One does not have to guess the frequency used by the other species, for instance. In fact, if the probe has a von Neumann machine payload, then the machine could construct an artifact in the solar system of the species to be contacted, an artifact so noticeable that it could not possibly be overlooked. If nothing else, the machine could construct a “Drink Coca-Cola” sign a thousand miles across and put it in orbit around the planet of the other species. Once the existence of the probe has been noted by the species to be contacted, information exchange can begin in a variety of ways.
As to the cost of such a vast exploration program, Tipler has this to say:
Using a von Neumann machine as a payload obviates the main objection to interstellar probes as a method of contact, namely the expense of putting a probe around each of an enormous number of stars. One need only construct a few probes, enough to make sure that at least one will succeed in making copies of itself in another solar system. Probes will then be sent to the other stars of the galaxy automatically, with no further expense to the original species.
A ‘Catastrophic’ Answer to Fermi?
Tipler suggested a timeframe of 300 million years to fill the galaxy with these devices, in an argument that drew fire from Carl Sagan and William Newman, who argued in 1983 that his approach was ‘solipsistic’ because the idea that we were alone in producing a technological civilization was anti-Copernican. And here we need to pause on a concept that has surfaced repeatedly in SETI studies not just in the western nations but also the Soviet Union. The idea of ‘mediocrity’ troubled attendees at the Soviet SETI meeting at the Byurakan Astrophysical Observatory in 1964, to be discussed again in a second meeting (with American scientists as participants) in 1971.
Do we just take the Copernican principle as a given? Sagan clearly thought so. His ‘co-author’ on Intelligent Life in the Universe, Iosif S. Shklovskii was far less sanguine on the matter:
Since we do not adequately understand the factors leading to the evolution of intelligence and technical civilizations, we cannot reliably estimate the probability that intelligence and technical civilizations will emerge.
Here I’m drawing on Mark Sheridan in his 2023 book SETI’s Scope (How The Search For Extraterrestrial Intelligence Became Disconnected From New Ideas About Extraterrestrials). Sheridan homes in on the philosophical disagreement between emerging Soviet SETI and the ideas in the Drake Equation. At Byurakan, Soviet mathematician A. V. Gladkii challenged the idea, accepted by Sagan, that mathematics could be a recognizable common ground between all intelligences across the stars. And Sheridan quotes Theodosius Dobzhansky, a Ukrainian-born geneticist later working in the U.S., who in a 1972 paper cast doubt on Sagan’s insistence that because intelligence had arisen on our planet, it must arise everywhere life exists. In his view, the principle of mediocrity was being taken several steps too far. Quoting Dobzhansky:
“Natural scientists have been loathe, for at least a century, to assume that there is anything radically unique or special about the planet Earth or about the human species. This is an understandable reaction against the traditional view that Earth, and indeed the whole universe, was created specifically for man. The reaction may have gone too far. It is possible that there is, after all, something unique about man and the planet he inhabits.”
In a fascinating 2009 paper, Milan Ćirković examines the Fermi question in the context of our basic premises about science. As amplified in his later book The Great Silence: Science and Philosophy of Fermi’s Paradox (Oxford University Press, 2018), the Serbian astronomer points to the focus the ‘where are they’ question places upon both Copernicanism and gradualism. In the former, as clearly stated by Sagan as by many other of the early SETI practitioners, the assumption is that we occupy no privileged place in the cosmos, and thus should expect other civilizations to exist, some of which would be far more advanced than ourselves. Yet we do not observe them.
Many answers can be offered to Fermi’s question, of course, but as we continue probing the cosmos, the silence takes on escalating significance. Must we envision a future in which we abandon Copernicanism and assume that we do not, in fact, occupy a relatively common niche in the cosmos, but rather a rather special one?
Or should we give up on gradualism, the idea that geophysical processes proceed in the future more or less as they did in the past? The concept is foundational to 18th Century geology and remains a commonplace in current thinking. But ‘catastrophism’ is an obvious factor in the development of life, as extreme ruptures like the K–T extinction event that ended the era of the dinosaurs make clear. Are there common factors that could affect planets throughout what is thought of as the Milky Way’s habitable zone?
The question is the focus of recent work on gamma ray bursts and implies, as Ćirković notes, a ‘reset’ of the clock. That could explain our lack of detections, as it would imply that living worlds, no matter their geological age, have had only about the same amount of time we have had to develop intelligence. The Fermi question highlights both of these key assumptions, while our lack of a solution keeps the tension tight.
The Bracewell paper is “Communications from Superior Galactic Communities,” Nature Volume 186, Issue 4726 (1960), pp. 670-671. Abstract. On the LaGrange search, see Freitas, “A search for natural or artificial objects located at the Earth-Moon libration points,” Icarus, Volume 42, Issue 3 (June, 1980) p. 442-447 (abstract). Michael Hart’s paper on galactic expansion is “Explanation for the Absence of Extraterrestrials on Earth,” Quarterly Journal of the Royal Astronomical Society, Vol. 16, p.128 (full text). Frank Tipler’s paper on self-reproducing probes is “Explanation for the Absence of Extraterrestrials on Earth,” Royal Astronomical Society, Quarterly Journal, vol. 21 (Sept. 1980), p. 267-281 (full text). Milan Ćirković’s paper on Fermi and Copernicanism is “Fermi’s Paradox – The Last Challenge for Copernicanism?” Serbian Astronomical Journal 178 (2009), 1–20. Preprint.
On Artemis and Starshot
Watching Artemis lofting skyward I relived the Apollo launches, experiencing feelings that no subsequent missions ever engendered. Artemis involves taking humans back into exploration mode with our spacecraft. Getting people out of low-Earth orbit again is a thrill despite the astonishing cost of the SLS launch vehicle. Obviously finding alternatives that would make more frequent flights possible has a major place on the agenda if we are to contemplate a continuous presence on the Moon, not to mention Mars. But for now, what a kick to see that big bird climb.
The distance between actual goals and dreams sometimes shrinks, and we saw recently that Breakthrough Starshot has made serious progress in developing the engineering concepts for an interstellar flyby. Both Artemis and the evolving Starshot design remind me that while most of the population in any era does not venture far from home, there are always a few who do, and those few change the shape of their civilization. Spaceflight obviously demands hardware and missions. Just as obviously, it demands scientists working on ways to push the envelope to attain still more distant goals. And it demands informing the public about where we are.
Be aware that Jim Benford’s recent interview on the matter is now available online. It’s part of a series of presentations offered by Paul Davies, Sara Walker and Maulik Parikh from the Beyond Center for Fundamental Concepts in Science at Arizona State University. This interview powerfully makes the case for funding Phase 2 of the Starshot program and developing early prototypes. The public needs to know about what has been accomplished and what steps lie ahead.
