Schematic illustration of the hierarchical structure of the Universe, from a galaxy group (a collection of galaxies) to an individual galaxy and the supermassive black hole at its center. Although a black hole is more than 100 million times smaller than the radius of its host galaxy, it plays a crucial role in the galaxy's central region. (Credit: Tohoku University)
A Nearby Black Hole Is Reshaping Its Galaxy Cluster Far More Violently Than Ever Measured
In A Nutshell
- A supermassive black hole at the center of a nearby galaxy cluster appears to be pumping energy into surrounding gas far more efficiently than earlier studies had estimated.
- The XRISM space telescope detected gas motions of roughly 300 kilometers per second, nearly twice the speeds seen in typical cluster cores.
- Researchers estimated a feedback efficiency of at least 1%, potentially as high as 10%, compared with earlier estimates below 0.01%.
- The result finally matches the efficiency levels that leading cosmological simulations have long assumed but never confirmed with direct observation.
For decades, astrophysicists have suspected that the supermassive black holes powering quasars, some of the brightest objects in the universe, play a major role in shaping how galaxies grow. But catching one in the act of pumping energy into the hot gas surrounding an entire galaxy cluster has proven elusive. Now, an international team of researchers publishing in Nature Astronomy has done exactly that, and the efficiency they inferred dramatically exceeds earlier observational estimates, by a factor of hundreds to thousands.
Using Japan’s XRISM space telescope, scientists observed H1821+643, the closest known galaxy cluster with a quasar at its center. The hot gas filling the cluster showed a velocity spread far stronger than motions previously measured in comparable cluster cores. The black hole’s winds may be driving a pressure wave through that gas, though the exact nature of the motions has not been definitively established. More telling still, researchers estimated a feedback efficiency, how much of the black hole’s radiated energy actually reaches the surrounding cluster gas, of at least about 1%, potentially reaching 10% if the proposed shock extends roughly 300,000 light-years out. That 1% figure is the conservative lower limit; the higher estimate depends on the size of a shock that has not been directly detected. Previous estimates from similar systems had put the number below 0.01%.
That gap is not a rounding error. It is the difference between feedback being a minor footnote and a major factor in how clusters are shaped over time.
Gas in This Cluster Moves Nearly Twice as Fast as Normal
XRISM, short for X-Ray Imaging and Spectroscopy Mission, carries an instrument called Resolve, a detector sensitive enough to catch tiny shifts in the energy of light from hot gas, the same way a train whistle changes pitch depending on whether the train is approaching or pulling away.
Scientists pointed XRISM at H1821+643, a cluster roughly 3.7 billion light-years from Earth, across six days in September 2024, collecting more than three days’ worth of usable observing time. Researchers analyzed X-ray light from highly charged iron atoms in the hot gas that fills the space between galaxies in the cluster. In clusters without powerful black holes at their centers, that iron light shows relatively narrow features, indicating gas motions of no more than about 160 kilometers per second. In H1821+643, those same features were dramatically broader, pointing to a velocity spread of approximately 300 kilometers per second.
To confirm the signal came from cluster gas, not the black hole itself, the team cross-referenced data from NASA’s Chandra X-ray Observatory. More than 90% of the iron emission originated between roughly 65,000 and 326,000 light-years out, well beyond the galaxy itself and firmly within cluster territory.
This Black Hole Is Far More Forceful Than Scientists Thought
Previous studies showed that even clusters with bright, actively feeding black holes tend to produce relatively gentle gas motions. H1821+643 is different. Its black hole is radio-quiet, with only weak twin radio jets, yet radiates one to five orders of magnitude more powerfully than others of its kind studied so far.
A plausible interpretation is that its winds are driving a pressure wave too weak to leave a clean shock signature in Chandra imaging. Whether the vigorous gas motions in the XRISM data trace turbulence, velocity shear, bulk flows, or some combination, they are unmistakably real.
Non-thermal energy tied up in those motions came out to about 8.4% of the cluster’s total thermal energy in the relevant region, higher than XRISM and Hitomi measured in clusters with less luminous black holes, where the fraction typically runs 1% to 5%.
Its Feedback Now Matches What Simulations Require
Cosmologists building large simulations of galaxy formation have long needed this kind of feedback strong enough to stop runaway star formation in massive galaxies, or clusters would end up packed with far more stars than astronomers actually see. For years, simulations simply assumed feedback efficiencies of a few percent to make the numbers work, without direct proof that real black holes behaved that way. This measurement is that proof. Major simulations, including COLIBRE and IllustrisTNG, use similar ranges, though the authors note the comparison is a rough one rather than an exact match, since simulations define efficiency differently than a direct cluster measurement can.
H1821+643 also occupies an unusual place in cosmic history. Close enough to study in detail, it behaves like the powerful black holes thought to have dominated galaxy growth much earlier in the universe’s history, when they were far more common. Most galaxy clusters this close to Earth are governed by far quieter black holes. This one, feeding at roughly 30% to 60% of its maximum possible rate, looks like a nearby stand-in for the black holes that shaped galaxy growth during the universe’s most active period of construction.
A jump of two to four orders of magnitude over earlier estimates will need testing in other clusters before astronomers call it settled. Still, in this one system, a black hole appears to be actively reshaping its surroundings, at an efficiency that finally lands in the range the universe’s large-scale structure seems to require.
Paper Notes
Limitations
Several important caveats apply to these findings. Although a velocity spread of roughly 300 kilometers per second is clearly detected, the researchers acknowledge uncertainty about whether those motions represent true turbulence, bulk flows, velocity shear, or a combination including gas sloshing. Chandra imaging of H1821+643 shows features that could indicate sloshing, though the authors note that the near-zero net bulk velocity and the large velocity spread make a highly symmetric sloshing configuration unlikely. Additionally, the feedback efficiency estimate depends strongly on the assumed shock radius, and the comparison between the observationally derived efficiency and values used in simulations is described by the authors as qualitative, since simulation efficiencies are defined at a subgrid injection level and are not strictly equivalent to the cumulative, radius-integrated estimate made here. XRISM’s moderate angular resolution also required careful correction for signal leaking between detector regions, introducing modeling uncertainties. Resonance scattering of the iron emission lines could also affect the line width measurements, though accounting for it still yields a velocity dispersion well above other clusters.
Funding and Disclosures
This work was supported by the Japan Society for the Promotion of Science (JSPS) through various KAKENHI grants awarded to multiple authors. Additional support came from the RIKEN Special Postdoctoral Researcher Program, the Program for Forming Japan’s Peak Research Universities, the Kavli Institute for the Physics and Mathematics of the Universe (Kavli IPMU) and associated initiatives, an SNSF Consolidator grant, a Fondecyt Regular grant, an ANID BASAL project, the China-Chile joint research fund, the Kyoto University Foundation, and the University of Geneva, which provided open access funding. Specific grant numbers are listed in the paper’s acknowledgments section. Authors declare no competing interests.
Publication Details
Title: Vigorous turbulence driven by quasar-mode feedback in a cluster core | Authors: Satoshi Yamada, Shutaro Ueda, Hirofumi Noda, Yutaka Fujita, Misaki Mizumoto, Kentaro Nagamine, Claudio Ricci, Shoji Ogawa, Taiki Kawamuro, Shinya Yamada, Yuichi Terashima, and Yoshihiro Ueda | Journal: Nature Astronomy | DOI: https://doi.org/10.1038/s41550-026-02939-x | Received: October 15, 2025. Accepted: July 2, 2026.
