Centauri Dreams
Imagining and Planning Interstellar Exploration
Sakurai’s Object: A Stellar Rebirth
Given how long stars live in comparison to human lifetimes, I always do a double-take at science fiction tales of starships dropping in to study a nova just as it’s about to go off. I suppose we have to assume the starship civilization has found a way to time such matters. There’s a Star Trek: The Next Generation episode that involves rescuing a planetary population from a star that’s about to explode (“A Fury Scorned”), but of all the nova arrivals, I like Samuel Delany’s 1968 novel Nova the best. Here, in order to harvest a rare element dubbed ‘Illyrion,’ the exact moment of the explosion has to be known and exploited through a mind-bending run through the debris.
Today’s paper doesn’t involve a nova, but it does involve a star that is doing things on a very short timeframe indeed. The star is known as Sakurai’s Object (V4334 Sagittarii). Japanese amateur Yukio Sakurai observed the object in 1996, noting how it appeared to be brightening. A 1976 detection of the progenitor star had recorded a magnitude of 21, whereas when Sakurai tagged it, it had reached magnitude 11. The apparent eruption here was first thought to be the result of a nova, but that was discounted for spectral reasons.
Several years later the unusual brightening began to decline. That was evidently the result of loss of mass and subsequent condensation of stellar material, which wound up hiding the star behind a dusty screen. It now shows a spectrum similar to what is known as a Wolf-Rayet star. Sakurai’s Object appears, however, to be in the much less massive Wolf-Rayet class known as a [WR] star. A conventional Wolf-Rayet star is massive, stripped of its hydrogen by huge stellar winds that leave the underlying helium-burning layers exposed. The odd [WR] notation turns out to be needed because low-mass objects like these can mimic a Wolf-Rayet star, so it was necessary to show that this is a different kind of object and a much smaller kind at that, though with Wolf-Rayet features.
There are also sub-categories of [WR] including [WC] that don’t need to occupy us now, although the paper explains what they mean. What’s intriguing about Sakurai’s object is its pace. According to current calculations it’s now roughly 6 times hotter than it was 30 years ago, and is climbing back toward the white dwarf temperatures it had before the eruption that made it observable by Sakurai. Albert Zijlstra (Jodrell Bank Centre for Astrophysics, University of Manchester), describes its significance:
“Most stars evolve so slowly that major changes take place over timescales far longer than a human lifetime. As a result, we usually have to piece together snapshots of stellar evolution by comparing different stars at different stages of their lives. Sakurai’s Object offers something far rarer. It is one of the very few stars known to have changed dramatically within just a few decades, giving us the opportunity to watch stellar evolution unfold in real time. With our observations, we can test theories of how stars evolve and gain new insights into one of the shortest and least understood phases in the life of a dying star.”
Image: The rapid brightening of Sakurai’s object allows astronomers to study the final phases of the stellar evolution in only a few decades. The left and right panels show the brightening. The middle panel is an image obtained with the radio telescope ALMA, showing the material ejected after the star re-ignited. The material currently extends over a size similar to our entire solar system. Credit: Stefan Kimeswenger, University of Innsbruck; Peter van Hoof, Royal Observatory Belgium. The observatory also has a short video showing the brightening.
What we apparently have here is a star once similar to the Sun that had finished its nuclear burning and was in the process of turning into a dense white dwarf star about the size of the Earth. But an event known as a ‘very late thermal pulse’ seems to have occurred. This happens when helium deep inside the star reignites, forcing a rapid expansion along with the ejection of stellar materials, so that the star has, at least for the time being, been reinvigorated. Along with another star called V605 Aquilae, Sakurai’s Object is one of only two stars that have been directly observed going through this process.
Let me turn to the paper to home in on the matter of thermal pulses:
Depending on when the pulse occurs, it is classified as an AGB final thermal pulse (AFTP), a late thermal pulse (LTP; post-AGB), or a very late thermal pulse (VLTP; see M. M. Miller Bertolami 2024). In the most extreme case (VLTP), the star is already on the white-dwarf cooling track when the flash causes rapid expansion and cooling, accompanied by substantial mass ejection into the circumstellar environment. The star is thus “born again” and returns close to its former AGB position in the HR diagram. Subsequent evolution involves reheating, and possibly an additional cooling excursion, as the star evolves back towards the white-dwarf domain with a markedly altered surface composition (T. M. Lawlor & J. MacDonald 2003).
No wonder this was a hard object to observe. The ejected gas and dust following the 1996 outburst caused the star to became hidden to direct imaging. Low density gases are coming off the star even as the outflowing atmosphere, moving at e 500 kilometers per second, is optically thick. Much of the spectrum produced is nebular. The team compared data from the VLT with computer models originally developed to study the processes that power the atmospheres and winds of Wolf-Rayet stars. The emission lines studied in this paper are embedded in the stellar wind, as opposed to the ejecta and dust. Much of the credibility of the analysis depends upon separating these factors.
The analysis reveals a stellar temperature at the surface pegged at between 27,000 and 36,000 degrees Kelvin. Scientists will now be able to study the process of the reheating following the eruption that occurred thirty years ago. Early indications are that the process is occurring more slowly than some models had predicted, so that Sakurai’s Object becomes a useful laboratory, like V605 Aquilae, into stellar behavior in this extremely rare class of stars. At some point, Sakurai’s Object will return to its fate as a white dwarf, only now under close observation as we ponder massive changes in a very short order.
Griet Van de Steene (Royal Observatory of Belgium), a co-author of the paper adds:
“Sakurai’s Object evolved much more quickly than pre-existing models for stellar evolution predicted. This led to a new generation of models that we now need to test. Our measurements show that the star is reheating more gradually than some of these new models predicted. That gives us an important way of testing which theories best describe what happens when a dying star briefly springs back to life. As we continue to monitor the star over the coming years, we expect to learn much more about this remarkable phase of stellar evolution.”
The paper is Marcolino et al (2026). The emergence of a [WC] star in Sakurai’s object. Monthly Notices of the Royal Astronomical Society, 552(1), Article stag1533. Full text.
Disrupted Systems: Implications for Life
Anthropocentric thinking is a persistent problem when we’re talking about extraterrestrial civilizations. Some of our most cherished notions can be invoked so effortlessly as to defy the imagination. The Copernican idea that life must exist elsewhere because elsewhere is bound to be more or less like here has had a long lifetime in SETI studies. It seems like the most basic common sense. And yet, as we’ll see once again today, scientific results continue to make it apparent that the fact that we are here does not mean that they are there.
Finding stellar systems more or less like our own continues to be difficult. I’m not going full ‘rare Earth’ here but acknowledging that our circumstances may be unusual enough to tamp down estimates of the number of life-supporting planets. For that matter, our lack of knowledge about abiogenesis itself makes the case that we cannot necessarily expect it around other suns. On this matter, at least, we should be able to gather data soon, perhaps with a Mars lander.
Which gets me to today’s interesting take on binary stars. The binaries – HD 129171 and HD 129209 – are G-class, both having formed from the same molecular cloud. This is a common enough scenario, because binary systems are extremely common. In fact, about half of the stars in the Milky Way have a companion star, and thus are likely to share a common chemistry.
About 180 light years from the Sun (based on Gaia data) in Boötes, the system is proving unusually helpful. Using the UVES spectrograph on ESO’s Very Large Telescope (VLT) in Chile, an international effort led by Anne Rathsam at the University of São Paulo in Brazil is using this binary as a laboratory to study chemical differences between the stars that may be telling us something about planet formation and orbital evolution. What is driving the differences?
All this has a bearing on life, because for life to occur, conditions must exist for long enough to let the necessary processes go to work. And that may be a problem. HD 129171 turns out to be enriched in refractory elements, while its companion HD 129209 is not. Refractory elements condense at high temperatures, as opposed to volatiles, and the proportion of one to the other can be an interesting diagnostic tool, telling us something about the system’s history. A star depleted in refractories is likely one that has spawned planets, while one heavy in these elements has likely swallowed planets whole.
Rathsam’s work, available in a paper just published in Astronomy & Astrophysics, focuses on beryllium, whose abundance here is interesting because it is not produced in stellar interiors. Measured properly, its presence points to a star that consumed planetary material long after formation. Lithium, beryllium, and boron, in fact, share characteristics that make them exceptional in terms of chemical emergence in the galaxy. Jorge Luis Melendez Moreno (USP), who served as study advisor to Rathsam on this paper, explains:
“All other chemical elements originate from primordial nucleosynthesis [the formation of the first atomic nuclei in the minutes following the Big Bang] or stellar nucleosynthesis [the nuclear fusion process that occurs inside stars throughout their lifetimes]. But not beryllium and boron. They primarily arise through a process called ‘cosmic spallation,’ in which high-energy particles fragment heavier nuclei, such as carbon, nitrogen, and oxygen, producing lighter elements.”
Image: The binary system of HD 129171 and HD 129209. Credit: Digital Sky Survey/Aladin/Anne Rathsam.
Lithium has its own uses as a marker, but the USP researchers found that beryllium is the more reliable tool, with a longer lasting chemical signature. HD 129171’s surplus of refractory elements (including magnesium, silicon, calcium and titanium) compared to its companion HD 129209 is accompanied by lithium and beryllium excess as well. The authors see evidence for rocky material equivalent to more than eleven times the mass of the Earth. This would have been ingested in the star, raising interesting questions about planet dynamics.
The infall of entire planets into a host star can be explained in various ways, from gravitational perturbations from other planets on eccentric orbits or skewed inclinations, to interactions within the early circumstellar disk driving young planets onto migratory trajectories. Although I hadn’t seen this paper before, the authors cite a 2025 study by Soares and team (citation below) that draws on simulations to show that about half of the stars simulated should ingest planets, leaving a chemical signature in perhaps 20 percent of them.
Evidently we can imagine system disruptions as a common occurrence in the Milky Way. We might add to this the fact that planet configurations with giant planets in circular outer orbits and rocky planets on inner ones are not common. Our observational data so far tends to confirm this, although tracking outer Jupiter-class planets with our current detection methods is always problematic given their sparse transit signatures. Overall, though, the picture that is emerging is that systems similar to the Solar System seem rare. If many systems are going through upheavals through planet migration and ingestion into the host, stable orbits may be trickier than we’ve thought. Rathsam comments:
“In our planetary system, the planets have relatively stable, low-eccentricity orbits. However, if planetary engulfment is common, it suggests that many systems undergo violent dynamic phases.”
Implications?
“Life wouldn’t just need billions of years to emerge and evolve. The planet would also have to remain in a sufficiently stable orbit to survive significant gravitational perturbations.”
If you’re wondering about our own star, it’s light in refractory materials relative to volatiles, which at least one recent study sees as a sign of planet formation at an early era. What became rocky planets and planetesimals, in other words, was never accreted into our star. I give that citation below. This and subsequent work seems to peg the Sun as the center of a dynamically quiet system compared to many, but be aware that the debate on this matter continues.
How useful is beryllium as a marker in such an analysis? The authors argue that despite the fact that beryllium is eventually depleted in a stellar interior, it persists long enough to provide reliable information. From the paper (my italics):
Refractory elements offer an excellent way to distinguish between the planet engulfment or the proto-cloud inhomogeneity scenarios. Since they have high condensation temperatures (≳1000 K), they are the primary constituents of the rocky material in planetary systems – terrestrial planets and cores of gaseous planets. In case of engulfment by a Sun-like star, this material is accreted by the star and then dissolved and mixed in the convective envelope, increasing the stellar surface abundances after the event (Sandquist et al. 2002). This process, however, produces a metal-rich outer layer with an unstable mean molecular weight gradient, which triggers thermohaline mixing (Théado & Vauclair 2012; Sevilla et al. 2022). For fragile elements such as Li and Be, this thermohaline mixing induces depletion, as it can carry these elements below the convective zone, into their burning regions. Thus, the chemical enrichment caused by the engulfment disappears over time. Nevertheless, Sevilla et al. (2022) demonstrated through simulations that the Li engulfment signature in stars with masses close to solar can be detected for ≥1 Gyr.
As far as I can tell, it’s the treatment of beryllium as a marker of engulfment that gives this paper its significance, as heretofore it has been used primarily as a diagnostic for the mixing of elements in stellar interiors. More broadly speaking, though, any reminders of the factors that make our Solar System adaptable for life should help us in identifying systems where life is less likely. That’s worth keeping in mind as we move beyond easy assumptions about life’s ubiquity and dig into realities that may make it more unusual than we thought.
The paper is Rathsam et al. (2026). Planet engulfment in the chemically anomalous HD 129171/HD 129209 pair. Astronomy & Astrophysics, 710, A236 (full text). The interesting paper on beryllium’s persistence is Soares et al. (2025). Assessing the processes behind planet engulfment and its imprints. Astronomy & Astrophysics, 693, A47 (preprint). The paper on the composition of our Sun is Meléndez et al. (2009). The peculiar solar composition and its possible relation to planet formation. The Astrophysical Journal Letters, 704(1), L66–L70. Abstract.
Pandora: Tuning Up our Data on Exoplanet Atmospheres
Launching a flagship-class mission like the Roman Space Telescope is always exciting, but with Roman in space, let’s also keep an eye on Pandora, a smaller though fascinating NASA mission (through its Astrophysics Pioneers program) that is now beginning its own work in exoplanet science. 20 exoplanets are targeted here, the idea being to work with data from transits to characterize their atmospheres via transmission spectroscopy, but with an additional twist. Elisa Quintana,(NASA GSFC) is principal investigator:
“Pandora’s data will help close a major gap in our knowledge about planets and their host stars because, right now, we can’t be entirely sure how the star’s light affects measurements of what makes up exoplanet atmospheres. We designed the Pandora spacecraft and its in-depth observing program to better understand this vexing issue.”
Launched January 11 of this year, Pandora is now beginning observations of its target stars. It’s in that category of smallsats that fascinate me because their low cost also gives them the option of higher tolerance of failure, making it possible to push the limits without great financial risk (the cost cap on this class of mission is $20 million). A tool like this can be tightly focused on specific targets without compromising larger instruments that are already besieged with obligations to existing observing programs. Word from the Pandora team is that science work begins with all instruments performing as expected.
Image: Artist’s concept of NASA’s Pandora mission, which will help scientists untangle the signals from exoplanets’ atmospheres and their stars. Credit: NASA GSFC/Conceptual Image Lab.
What makes Pandora so interesting is that it’s a multi-wavelength mission. Pandora’s telescope, an aluminum Cassegrain instrument 45 centimeters in diameter, feeds detectors that work in both visible light for long-baseline photometry as well as near-infrared (NIR) wavelengths. Starspots or other stellar activity can be separated from the near-infrared spectroscopy data on the planet’s clouds and atmospheric hazes. This is information that will inform subsequent work with telescopes like JWST and Roman as we refine our tools for removing noise in the data. A recent paper on Pandora makes the case:
Alongside the opportunities explored throughout this work, Pandora provides a unique opportunity to help overcome many hurdles inherent in observations with current instruments. Namely, multi-epoch observations of exoplanet atmospheres with JWST are often observed at different points in the stellar rotation, making them difficult to fit jointly (e.g., E. M. May et al. 2023). By providing information about the stellar activity and allowing for corrections at the data level, joint Pandora–JWST programs will unlock new insights into planets around active hosts. Additionally, these same insights will inform the presence of offsets between different instruments and epochs, which have been difficult to constrain thus far with JWST alone (e.g., A. L. Carter et al. 2024).
Pandora is going to give us useful data on stellar contamination in our transmission spectroscopy work. That means a clearer and more reliable look at the composition of exoplanet atmospheres. Stellar contamination should stand out, with a minimum of 10 transit observing sessions per target, each lasting about 24 hours. The plan is to accumulate, for each target, roughly 240 hours of pointing time, which should yield more than 120 hours of science data once Earth occultations and unavoidable gaps for downlinks, etc. are taken into account. An instrument like JWST can collect transmission spectroscopic data for exoplanets but only with relatively short observing sessions involving one or a small number of transits. The combined datasets of the two observatories will cleanly excise star signals from the data on planetary ones.
The finalized target list was published last January and contains 19 host stars, one of which is orbited by two of the target worlds. The stars range from M-dwarfs to K-class. 16 of these planets have been observed previously by JWST, which will allow the retrofitting of the Pandora measurements onto the JWST spectra. As far as I can tell, earlier provisional target lists included planets that didn’t make the final list, with the emphasis shifting toward larger warm Neptunes and Saturns. That makes sense given that these offer higher signal-to-noise ratios as we put these techniques to work together for the first time.
Image: This is Figure 1 from the paper. Caption: The absorption cross sections of common absorbers in exoplanet atmospheres considered in this work, shown at a pressure and temperature of 0.1 mbar and 1000 K. The wavelength coverage of Pandora’s NIR detector (NIRDA), JWST’s NIRCam F322W2 and F444W filters, and HST’s WFC3 instrument (for the G141 grism) are shown. Pandora/NIRDA covers absorption bands of H2O, CH4, NH3 and the wing of the K doublet, making it most sensitive to these absorbers. Credit: Rotman et al.
Using these methods, Pandora’s science observations should be more accurate and reliable than any planetary atmosphere readings ever taken before. The paper is Rotman et al., “NASA’s Pandora SmallSat Mission: Simulated Modeling and Retrieval of Near-Infrared Exoplanet Transmission Spectra,” accepted at The Astronomical Journal (preprint).
Fermi Explorer: Building the First Mission to Another Star?
The question of when to launch an interstellar mission has occupied us many times in the past. Specifically, how long do we wait so that travel times are reduced to something like the lifetime of a researcher working on the project? But there is another approach to all this. Someone is going to launch an interstellar mission that will be the first human effort to send a payload to another star. It’s all about intentionality and the choice of targets.
A symbolic act? Sure, but don’t write the idea off. We can learn a lot from symbolic acts, and if we only have, at our current level of technology, the ability to reach Voyager-like speeds, we can still work on issues like equipment lifetimes, self-healing technologies, navigational issues and more. We can also work to refine existing AI tools to achieve the most efficient design.
If we give ourselves 80,000 years to reach Alpha Centauri, we have to contend with the fact that the system is constantly moving. On this timeframe, by the time the craft would arrive, Centauri A and B would be a bit over 6 light years from the Sun as opposed to their current 4.365 light years. Trajectory analysis going this far into the future is going to be an interesting challenge.
I mention all this because a call to mount such a mission has now arisen. It bears the name Fermi Explorer, and according to its new website, its intention is to get a spacecraft with a 1 kilogram, 10X10X10 cm payload to the barycenter of the binary Centauri A and B system. In other words, the target is not either star itself but the common center of mass between the two as they orbit.
Some particulars: The mission should launch before the end of 2029 if the effort succeeds, and is intended to cost less than $15 million to design, build, launch and operate. Mission co-founder Philip Johnston is going to have his hands full.
As to departure, Fermi Explorer would take a year and a half moving out of Earth orbit. Then, using a series of Oberth maneuvers taking it to within 0.42 AU of the Sun, the craft would rely upon what the site calls a ‘perihelion pump,’, which involves multiple close solar flybys over 12 years to build the energy to achieve an escape trajectory that, after climbing out of the Sun’s gravity well, attains 23.64 km/sec. That’s a bit higher than Voyager 1’s 17 km/sec. Final Solar System departure would be, after a 2029 launch, around the year 2043. Ahead for the spacecraft would be an unpowered cruise of over 70,000 years.
What the craft will carry is not yet determined, although I notice the plan to put a copy of the Voyager Golden Record and similar materials onboard (I’m assuming this is to be done digitally). Ahead is a three-month period for solicitations for other items of cultural value. Likewise, scientific instruments will undergo their own period of solicitation. The emphasis is on flight-proven hardware with little research and development necessary. To quote from the website:
We will soon put the mission out for open tender to all the major satellite manufacturers, and we aim to open-source as much of the design as we can. The four primary objectives are considered non-negotiable. Everything else is negotiable. For example, the manufacturers can determine the power system, antenna strength, propulsion, mission profile, and whether to include gravity assists, etc. We anticipate that we can do the mission with around a 100-200 kg small solar-powered satellite with just electric propulsion, doing what we call a perihelion pump maneuver… We expect the mission will not have a large antenna for communication, and so we expect we will lose connectivity relatively quickly, and so much of the mission will be autonomous. It will be too small to track and will lose power once it leaves the solar system.
Can crowdfunding build an interstellar craft? The hope is clearly that enough people will become interested to help, with the site offering engraved names and physical objects in the payload itself, so the scientific payload, already tightly squeezed, will have a mass budget with even tighter constraints.
And with all the attention AI is getting in the press, note its use here. The website points to a key technical report called “Interstellar Precursor Mission to Alpha Centauri: Technical Feasibility Assessment,” dated July of 2026. Specifically, the report is said to be: “Prepared with PSI’s Autonomous Physics-Research Platform,” under which is stated “Physical Superintelligence’s agentic research system produced the analyses, simulations, and proof-grade verification in this report end-to-end under staged independent audit.” And again: “This report did not undergo comprehensive human peer review.”
This gets interesting. Writing for MIT Technology Review, Michelle Kim has a fine piece on the use of AI for Fermi Explorer that fills in the background. PSI is a research laboratory called Physical Superintelligence, and its AI system is what came up with the trajectory Fermi Explorer would follow. According to Kim, PSI’s AI went to work on the problem of getting a small spacecraft like this up to speed:
A week later, the AI system turned up a novel trajectory… It combined well-known orbital maneuvers in a way the Fermi team had not considered, according to a paper that has not been peer-reviewed. It suggested that the spacecraft could first slow down so its orbit swings in close to the sun—closer than Mercury. On each close pass, it would fire its engine so that the solar panels get four times the light, and a burst of thrust delivered at high speed would buy more energy than the same burst anywhere else. Because the engine would run only near the sun, the solar panels could stay small and the spacecraft light.
The Fermi Explorer site also links to a separate mission analysis which cross-references the PSI report and seems to agree with its results almost completely. I’m assuming human peer review is going to come into play if momentum for this mission builds. But watching the development of these models for physics and their tweaking along the way is a fascinating exercise.
Roman Space Telescope Launches Tomorrow
We’ll have a lot to talk about when observations from the Nancy Grace Roman Space Telescope start coming in. But first we’ve got to get it out to its halo orbit around the L2 Sun-Earth Lagrange point. With the Sun, Earth and Moon permanently blocked out, the seeing should be good from this vantage some 1.5 million kilometers out. Just now I’ve had a note from Jim Benford reminding me that launch is currently scheduled for tomorrow, August 30, at 0726 ET (1126 UTC). The additional good news is that Jim’s son Dominic will be one of two hosts of the live coverage from NASA. I remember meeting Dominic some years back at Goddard Space Flight Center, where I had the chance to see the JWST telescope being prepared to go through its vibration testing.
Dominic has spent more than a decade working on the Roman telescope, so the honor of working the NASA live coverage is well deserved. The photo below shows him during preparations for the Sunday coverage, which starts at 0620 ET (1020 UTC). You can tune in on Facebook, Instagram, Twitch, X, Discovery+, Amazon Prime, YouTube, or NASA+ (see https://www.nasa.gov/live/ for links and updates).
And as Jim reminds us: “As with any launch, timing can change because of weather or other conditions, so check NASA for the latest schedule. Fingers crossed!”
A ‘Time Variable’ Feature on Pluto?
With New Horizons now 63 AU out, it’s time to ponder that record hibernation period that the spacecraft just went through. Over 10 months long, this one had the same goal as previous such periods, to hold down operating costs but also give a break to onboard instrumentation in hopes of augmenting mission lifetime. As a kid, I often thought how cool it would be to be involved with a distant spacecraft. For some reason I always visualized this as being here on Earth while working with a craft beyond the Solar System. So maybe I was dreaming more of being a flight controller than an astronaut.
I hadn’t realized until reading Alan Stern’s recent update that while hibernation means there are no commands and responses going back and forth, the craft does continue active data gathering from its Solar Wind at Pluto (SWAP) plasma detector and the Pluto Energetic Particle Spectrometer Science Investigation (PEPSSI) instrument. The Student Dust Counter is also in operation. Since awaking early in the summer, New Horizons has been delivering data from these instruments back to Earth.
I see that future flybys of Kuiper belt objects (KBOs) are still a distinct possibility, assuming that one can be found within range of the vehicle. And yes, the Vera Rubin Observatory is already in use in the search for such objects, with planning extended to requests for time on the Roman Space Telescope, now scheduled for launch within days. Writing about New Horizons always comes with a tinge of melancholy as I contemplate the fact that no successor is ready, but it’s a relief to see how active this tough little craft continues to be. There are good successor concepts out there, but none are currently funded.
Meanwhile, we have the interesting news about possible liquid flows on Pluto’s surface, revealed by New Horizons images of Sputnik Planitia, the nitrogen glacier whose northern regions reveal convection cells separated by features that suggest liquid nitrogen has flowed there in the past. The new work out of Southwest Research Institute (SwRI) and published in the Planetary Science Journal, suggests that this process continues, doubtless through liquid nitrogen upwelling from beneath the glacier.
SwRI’s Kelsi Singer, one of the study’s co-authors, comments:
“The surface of Sputnik Planitia is quite young, probably less than one million years based on modeling of the surface overturn, and thus these features that we are looking at must have formed since then. Pluto has many unique terrains seen nowhere else in the solar system, and this area of Sputnik Planitia is one of them. Its surface provides a different set of conditions compared to what we are used to on Earth, and exploring that allows us to better understand how materials behave in environments that are difficult to produce on Earth.”
Image: Pluto’s northern Sputnik Planitia glacier (in the western or left side of Pluto’s bright heart) is shown here in a color mosaic made from NASA’s New Horizons imagery. The direction of north is shown on the image. The image is ~ 700 x 350 kilometers across. The red box has been added to show most of the region containing dark features attributed to the wetting of the glacier by liquid nitrogen sourced from a “basal melting” process beneath the glacier. NASA/Johns Hopkins APL/SwRI.
These findings are based on studies of a glacial surface far from Earth and involve liquid nitrogen rather than water, but it’s interesting that the paper reports that the surface patterns on this region of Sputnik Planitia have darkened in ways that are similar to Earth glaciers that have undergone their own wetting by subsurface liquid. The team worked with images from the Landsat 9 satellite that included the Greenland ice sheet. Computer models of glacial melting led by Orkan Umurhan (SETI Institute) reveal a transport mechanism to the surface and indicate an icy flow that can persist on the surface for short periods, which would account for the dark features.
And there is another Earth analogue that perhaps comes into play. From the paper:
We have posited that the striking, dark, irregular linear and diffuse features prevalent at the northern margin of SP are manifestations of the upwelling of liquids or liquid slurries from beneath the SP ice sheet. We have further posited that these liquids were created as a basal melt at depth below the glacier, and we have quantitatively shown the plausibility that such a mechanism can then advect through conduits to the surface of SP where it will horizontally spread out, leaving a darkened surface akin to that observed in northern SP. The buoyant ascent of liquid nitrogen through the denser N2 ice is akin to terrestrial volcanism where less dense basaltic melt rises through cold continental crust.
Does this process occur on other dwarf planets in the Kuiper Belt? We won’t know that until we have mapping missions to objects like Eris. The paper also mentions the obvious candidate at Neptune, the large moon Triton, as a possible example of such processes factoring into the geysers Voyager 2 observed during its flyby. For that matter, it’s fascinating to consider that despite the success of New Horizons at Pluto/Charon, over half the planet (I know I’ll get emails about calling it that) remains unmapped at high resolution because of the geometry of the encounter. So for the time being, we don’t even know whether apparent flows like that at Sputnik Planitia occur elsewhere.
The SETI Institute’s Umurhan points out that the physics of solid nitrogen materials under stress and strain are not well understood, and have to this point “never been studied in real detail in the laboratory.” We need, in other words, to go out there and look, and New Horizons has just reminded us why.
The paper is Stern et al., “Evidence for Possible N2 Basal Flow beneath Pluto’s Northern Sputnik Planitia,” Planetary Science Journal Vol. 7, No. 7 (31 July 2026), 185 (full text).
