In a new study, University of Maryland astronomers spotted a supermassive black hole lurking at the edge of a galaxy. It’s the first time scientists detected a dormant black hole so far away from a galaxy’s center. Normally invisible, the celestial abyss revealed itself with a fleeting flash of light as it swallowed a star. The finding was published in The Astrophysical Journal Letters on July 27, 2026.
“This is a novel result. What’s new is that, until now, we've started with the assumption that supermassive black holes reside in the centers of massive galaxies,” said study co-author Suvi Gezari , an associate professor of astronomy at UMD. “This discovery will have a huge impact. It means that we're going to find many more examples of wandering black holes, and we can understand how galaxies and their black holes merge and build up over time.”
Theorists have predicted that “wandering” black holes would roam the outskirts of galaxies as galaxies collide and merge. But because many of these black holes are quiescent—meaning they aren’t swallowing material or emitting light—they tiptoe around undetected by earthly sensors.
The quiescent black hole described in this study was first detected through the Zwicky Transient Facility (ZTF), which “searches the whole universe,” said study lead author Robert David Stein , a Neil Gehrels Prize Postdoctoral Fellow at the Joint Space-Science Institute , a research partnership between UMD’s Departments of Astronomy and Physics and NASA’s Goddard Space Flight Center (GSFC).
ZTF scans the entire northern sky every two days using two telescopes based at the Palomar Observatory in San Diego County, California. It’s impossible to check for black holes in each of the hundreds of thousands of celestial events that ZTF documents each night. So, the researchers developed an artificial intelligence (AI) program that identifies a characteristic light pattern that occurs when a black hole shreds a star, an astronomical phenomenon called a tidal disruption event. Those events are well-documented at galaxy centers, but the AI looks for similar flares anywhere in the sky. The UMD team launched its AI program in August 2025 and made its black hole discovery just three months later.
“I remember the moment we discovered it very clearly. It was a Saturday, and everyone was very excited to be messaging. We dropped everything and started triggering all kinds of other instruments to get more data,” Stein said. “We weren't really sure we would be successful so quickly, so it's amazing that we found one so fast.”
The black hole described in this study is located 9.3 kiloparsecs (about 30,000 light-years) away from the center of its galaxy. It’s about the same mass as the black hole at the center of our galaxy, the Milky Way. But surprisingly, it has no observable galaxy swirling around it, said study co-author and UMD Astronomy Professor Sylvain Veilleux.
“To have such a big black hole outside of a galaxy is surprising to me,” he said. “There should be a Milky Way-like object around it—and that's definitely not the case.”
This black hole likely originated from two colliding galaxies. One possibility is that a larger galaxy cannibalized a smaller one and then slowly stripped away its stars until only the core remained.
A second possibility is that one galaxy already had two black holes orbiting closely at its center—this is called a binary black hole. If a third black hole from another merger enters the mix, complex three-body interactions would kick the smallest one out. Continued observations of this tidal disruption event could favor one of these two theories over the other, the authors said.
Characterizing how quiescent black holes behave is important because most black holes—including the one at the center of the Milky Way—are dormant. It’s not known if our galaxy hosts any wandering black holes on its outskirts, but Stein said we shouldn’t fear getting sucked into one: “We're very unlikely to meet one, at least in our lifetime.”
The team is now scanning the skies for more wandering black holes so they can “understand how galaxies form and how many black holes are whizzing around,” Stein said. This latest discovery shows that it’s possible to detect them without expensive techniques beyond standard sky surveys and machine learning.
“It is super exciting,” Gezari explained. “It’s an example of machine learning opening up a whole new area of research.”
Stein expects they’ll detect dozens or even hundreds of wandering black holes each year using the NSF-DOE Vera C. Rubin Observatory, which debuted last June in Chile. The observatory provides extraordinarily detailed images using the world’s largest digital camera.
Vellieux added that the Lowell Discovery Telescope and its Rapid infrared IMAger-Spectrometer —which debuted in June 2025 as a collaboration between GSFC, the UMD Department of Astronomy and Lowell Observatory—will allow astronomers to detect tidal disruption events from even greater distances from Earth than previously imagined.
“This is the strongest case of a wandering black hole that we know,” Veilleux said. This is going to set the standard.”
###
UMD Astronomy Adjunct Professor Stephen Bradley Cenko and postdoctoral associate Jillian Chin Rastinejad coauthored this article with Stein, Gezari and Vellieux.
The paper, “TDE 2025abcr: A Tidal Disruption Event in the Outskirts of a Massive Galaxy,” was published in The Astrophysical Journal Letters on July 27, 2026.
This research was funded by the U.S. National Science Foundation (Award Nos. 2407588 and 1106171); the Gordon and Betty Moore Foundatation through the Data-Driven Investigator Program and a dedicated grant to SkyPortal; the W. M. Keck Foundation; the Heising-Simons Foundation; William and Marina Kast; the University of California Observatories; Google; Deutsche Forschungsgemeinschaft under Germany’s Excellence Strategy – EXC-2094/2 – 390783311.939; the Australian Research Council under ARC LIEF grant LE130100104; Astronomy Australia Limited and the Australian Government through the Commonwealth’s Education Investment Fund and National Collaborative Research Infrastructure Strategy, particularly the National eResearch Collaboration Tools and Resources and the Australian National Data Service Projects. This article does not necessarily reflect the views of these parties and organizations.
The Astrophysical Journal Letters
Observational study
TDE 2025abcr: A Tidal Disruption Event in the Outskirts of a Massive Galaxy
27-Jul-2026