Astronomers Confirm Third "Wandering" Black Hole Caught Shredding a Star Far From Its Galaxy's Core

Astronomers have confirmed a third case of a massive black hole caught shredding a star far from the center of its galaxy, adding to a small but fast-growing catalog of what researchers call "wandering" or "off-nuclear" black holes — evidence that large galaxies may harbor dislodged black holes that drift undetected until, by chance, one of them rips a passing star apart.

The event, cataloged as the tidal disruption event TDE 2023mfm, was described in a preprint posted by a 27-member research team led by Itai Sfaradi, with co-authors including Ryan Chornock, Yuhan Yao and Raffaella Margutti, in late September and is only now drawing wider coverage. The black hole involved sits roughly 1.08 kiloparsecs — about 3,500 light-years — from the nucleus of its host galaxy, a massive system with its own, separately active central black hole. That makes the newly described object the second-most distant such offset yet confirmed, and the latest data point in an effort to answer a question that has nagged at astrophysicists for years: how many black holes are quietly loose inside galaxies, far from where theory says they belong?

A flare in the wrong part of the galaxy

A tidal disruption event, or TDE, happens when a star strays close enough to a black hole that the black hole's gravity stretches and ultimately shreds it. Roughly half of the star's gas falls inward, heating to tens of thousands of degrees and producing a flare that can briefly outshine every star in the host galaxy. Because supermassive black holes almost always sit at a galaxy's dynamical center, astronomers have treated TDEs for two decades as, in effect, a way of finding galactic nuclei — flares are expected to appear at the center and nowhere else.

TDE 2023mfm broke that pattern. It was first picked out of survey images from the Zwicky Transient Facility, a wide-field camera at Palomar Observatory in California that scans much of the night sky every few days looking for anything that brightens or fades. Follow-up spectroscopy at the Lick and Keck observatories showed the broad hydrogen emission line characteristic of a hydrogen-rich TDE, and additional observations with the Swift satellite's ultraviolet telescope, the Chandra and XMM-Newton X-ray observatories, the Hubble Space Telescope and the Karl G. Jansky Very Large Array were used to pin down its location, temperature and evolution. Those measurements placed the flare's source well away from the galaxy's nucleus — not a measurement glitch, but a second, independent object.

What the data show

According to the preprint, the disrupting black hole weighs roughly 1.6 million solar masses, a figure the authors derived through three independent estimation methods that converged on a broader range of about one million to ten million solar masses. The flare's temperature stayed unusually hot, around 22,000 kelvin, for more than a month after peak brightness, and its visible-light brightness remained above half its maximum for about 48 days, somewhat longer than many previously studied TDEs. The host galaxy itself is a heavyweight, with a stellar mass of roughly 100 billion suns, and harbors a faint, separately active nucleus of its own — meaning the galaxy appears to contain two black holes, one at its center and one well outside it.

Unusually, radio emission from the disrupted star's debris did not appear until roughly two and a half years after the optical flare was first detected, a delay the authors say is still unexplained and markedly different from the radio behavior seen in a comparable off-center TDE studied previously. Deep, late-time imaging with Hubble also turned up a faint, unresolved clump of stars at the exact position of the flare, with an estimated mass of 10 million to 100 million suns — too faint and compact to be a normal galaxy, but plausibly the stripped remnant of a small galaxy that was absorbed by the larger one long ago.

A separate, independent analysis published earlier this year in Research Notes of the American Astronomical Society, led by Wenkai Li, used the Very Large Array to resolve two distinct radio sources in the system — one at the galaxy's nucleus, the other at the flare's location — providing confirmation, from an entirely different dataset, that the two are not the same object.

Why a black hole would be off-center at all

The Sfaradi-led team lays out two main explanations. The one they favor is that the black hole is the relic of a minor galaxy merger: a smaller galaxy, with its own central black hole and a surrounding knot of stars, was swallowed by the larger galaxy long ago, and the stripped core — including its black hole — continues to orbit, undigested, inside its new host. The faint stellar clump Hubble detected at the flare's position is consistent with being exactly that kind of leftover. The alternative explanation is more exotic: the black hole could have been given a "kick" by the recoil of gravitational-wave emission, either from a black hole merger or from a dynamically unstable system of three black holes. The authors estimate that mechanism could plausibly account for only about one percent of all tidal disruption events, making the merger-remnant scenario the more likely explanation in this case. A third possibility — that the galaxy's two black holes are themselves a merging pair, kicked apart by their own interaction — is largely ruled out, the authors write, because the host nucleus shows signs of an active black hole of its own, implying two separate black holes were already present before this event.

A small population coming into focus

TDE 2023mfm joins two previously confirmed off-nuclear disruptions. The first, AT2024tvd, was identified through Hubble imaging by a team led by Yuhan Yao and announced by NASA and the Space Telescope Science Institute in 2025; its black hole sits about 2,600 light-years from its galaxy's center and is roughly ten times less massive than the supermassive black hole at that galaxy's core. The second, announced by NASA's Swift mission team this past July, involved a roughly million-solar-mass black hole caught devouring a star more than 30,000 light-years from the center of a galaxy 750 million light-years away — described at the time as the clearest example yet found of a wandering supermassive black hole. Robert Stein, the astronomer who led that discovery and who is also a co-author on the new TDE 2023mfm paper, said of the broader search effort at the time:

"With this discovery, which is one of just a couple that have been confirmed so far, we've validated a new technique"

— a machine-learning method, built to flag flares appearing away from a galaxy's center, that the same research network used again on TDE 2023mfm.

The stakes extend well beyond cataloging oddities. Computer simulations of how galaxies grow through repeated mergers have long predicted that large galaxies should carry populations of black holes that never settled into the center — survivors of swallowed satellite galaxies, drifting through the stellar halo largely undetected because they have nothing nearby to feed on. Until recently, that population was almost entirely theoretical. Each confirmed tidal disruption at an off-center location gives astronomers a rare, if brief, chance to measure one of these black holes directly: its mass, its surroundings and, through cases like this one, clues to its origin.

What comes next

The authors say continued radio monitoring of TDE 2023mfm is needed to understand why its radio afterglow brightened so late and so differently from other off-nuclear cases, and deeper imaging could help determine whether the faint stellar clump at its position really is the remains of a cannibalized dwarf galaxy. More broadly, the same survey techniques that caught this event and its two predecessors are expected to be applied systematically to years of archival data from the Zwicky Transient Facility, and eventually to the much larger stream of transients expected from newer wide-field surveys such as the Vera C. Rubin Observatory. If wandering black holes are as common as merger simulations suggest, researchers expect the current handful of confirmed cases to grow into a proper statistical sample within the next few years — enough, potentially, to test directly how often galaxies grow by swallowing their smaller neighbors, black holes and all.