Stellar Merger in Andromeda: Astronomers Watch a Giant Star Swallow Its Companion
Astronomers followed a star slightly lighter than the Sun as it spiralled into a supergiant in the Andromeda galaxy over seven years, and Webb observed the scene just 47 days after the 2025 eruption peaked.

For the first time, astronomers have followed a stellar merger from its slow beginning to its violent end. In the outskirts of the Andromeda galaxy, about 2.5 million light-years away, a star a little lighter than our Sun spent around seven years spiralling into a swollen supergiant before the two became one. The final eruption lit up in October 2025, and the James Webb Space Telescope was looking at the scene just 47 days after the outburst peaked.
The system is formally named AT 2025abao, but the research team calls it Bao. The results were published in Nature on 18 November 2026, and NASA announced them the same day. "Stellar mergers are very, very difficult to observe in real time because they are hard to predict," said lead author Viraj Karambelkar, a NASA Hubble Fellow at Columbia University in New York, in NASA's announcement.
Big Bao and Little Bao: the two stars in this stellar merger
The pair was very uneven. The larger star, nicknamed Big Bao, was a puffed-up supergiant of about 7 solar masses. It had swollen to roughly 350 times the radius of the Sun and shone about 18,000 times brighter. Its partner, Little Bao, had only about 0.7 times the Sun's mass, so it was a small, dim star by comparison.
The two had orbited each other for thousands of years. Big Bao kept growing as it aged, the way massive stars do near the end of their lives, until its outer layers reached Little Bao. From that moment the smaller star was effectively moving inside its partner's thin, extended atmosphere. Astronomers call this stage common envelope evolution: one star sinks through the gas of the other, losing orbital energy as it goes.
What surprised the team was the speed. According to NASA, once Little Bao touched the giant's surface, the merger took only about seven orbits, or roughly seven years. Earlier mergers of smaller star pairs needed thousands of orbits. "We thought an outburst from [such] an extended star would take years," said co-author Nadia Blagorodnova, an associate professor at the University of Barcelona in Spain.
Seven years of clues before the eruption
The most unusual thing about Bao is not the merger itself but how much of it was recorded. Most stellar outbursts are noticed only after they happen, and astronomers then search old images to see what came before. Here, several different surveys had been watching, mostly by chance.
Archive data from NASA's NEOWISE infrared mission showed the star holding a steady brightness before 2018. By February 2023, the same data revealed an odd, varying star, which is when the team first flagged it. In 2024, the WINTER infrared survey camera at Palomar Observatory showed the system brightening and forming dust. In mid-2025, NASA's SPHEREx mission saw it fading in visible light while growing brighter in infrared, a sign that cool dust was building up around it.
Then, on 17 October 2025, the Zwicky Transient Facility in California caught the eruption itself. NASA's Swift observatory and the Gemini North telescope in Hawaii followed up in November, and Webb observed the system on 24 December 2025 under a fast-response Director's Discretionary Time programme. "Webb observed Bao only 47 days after the stars' eruption peaked, the earliest ever follow-up of an event like this," Karambelkar said. Counting back, that puts the peak of the outburst around 7 November 2025.
Sven Buder, an astrophysicist at the Australian National University who was not part of the team, summed up why this matters in an interview with ABC News: "It's like investigating a crime scene, except that the cameras were already rolling before everything happened."
What Webb added
The earlier surveys showed that something was changing. Webb's infrared view, taken weeks after the peak, gave the team the detail they needed to reconstruct the whole event before, during and after the merger. Infrared light passes through dust far better than visible light, which matters here because merging stars wrap themselves in a cocoon of freshly made dust.
The researchers estimate that the eruption threw off around 0.1 to 0.2 solar masses of gas. Their models also suggest that the energy released as Little Bao sank inward was shared out in ways standard theory does not fully capture. Co-author Kishalay De of Columbia University described the event as "almost like a bomb exploding inside a living star." Simulations led by Karel Temmink, a PhD student at Radboud University in the Netherlands, helped the team test how the inspiral played out.
The paper, "An inspiral of a star inside a common envelope with its giant companion," measures how efficiently this kind of engulfment pushes gas away. That number matters well beyond Andromeda, because common envelope evolution is thought to be one of the main ways the universe builds very tight star pairs. Many close pairs in which a white dwarf pulls gas from a partner, the kind of system behind a nova outburst like the one expected from T Coronae Borealis, are thought to start this way. So are pairs of compact objects whose final collisions shake space-time, like the merging black holes in our recent story on black holes born from failed supernovae.
A "rejuvenated" star, but not yet a closed case
What is left at the centre now appears to be a single star, which NASA and the team describe as possibly "rejuvenated". A merger like this can add fresh material to a star and mix its interior, which could change how it evolves from here. Its final mass and state are still unknown, and the team says more observations will be needed to confirm conclusively that the two stars fully merged rather than leaving behind a tight new pair.
That caution is normal for a result like this. The interpretation rests on a long chain of observations from different instruments and on computer models of a process that is hard to simulate. Buder noted in the same interview that the gas environment the smaller star moves through is "incredibly hard to model." Watching how Bao fades and changes over the next few years is the best test.
Can you see where it happened?
Bao itself is far too faint to see without a large telescope. Its host galaxy, though, is one of the few galaxies visible to the naked eye, and late November is one of the best times of year to find it. The Andromeda Galaxy (M31) appears as a faint, elongated smudge of light, and from a dark site it is easy to spot in binoculars. It reaches its highest point around 8 to 9 p.m. local time in late November.
| Latitude | Example cities | Andromeda's highest altitude |
|---|---|---|
| 45–55° N | London, Paris, Berlin, Seattle | about 76–86°, almost overhead |
| 30–45° N | New York, Madrid, Tokyo, Beijing | about 79–89°, almost overhead |
| 10–25° N | Mexico City, Dubai, Mumbai, Manila | about 59–74°, high in the north |
| 0–10° S | Singapore, Nairobi, Jakarta | about 39–49°, in the north |
| 20–30° S | São Paulo, Johannesburg, Brisbane | about 19–29°, low in the north |
| 33–40° S | Sydney, Cape Town, Melbourne, Auckland | about 9–16°, very low in the north |
Altitudes are our own calculations for late November 2026. Viewers in the far south need a clear, dark northern horizon, and around the 24 November full Moon the bright sky will make the galaxy much harder to spot. If you want to know what else is above your horizon, the deep-sky objects visible tonight list works for any location.
What comes next
Bao shows that stellar mergers can give years of warning if the right telescopes are watching. "There's going to be a large influx of similar discoveries just ahead," Karambelkar said. New surveys will scan the sky far more often and more deeply, including the Vera C. Rubin Observatory in Chile and NASA's Nancy Grace Roman Space Telescope, which will image huge areas of sky in infrared.
The study also gives astronomers a rare real-world test for theories about how massive stars live and die. Many of the most extreme objects in the universe, from supernovae to merging black holes, begin in close binary systems. Watching one of those systems change in real time, rather than reconstructing it from its remains, is exactly what theorists have been waiting for.
Source: NASA / Nature ↗
More to read
All astronomy news →


From our guidesHubble vs James Webb: Orbits, Wavelengths and TrackingDeep Space · Aug 21, 2026
