Somewhere beyond 10 billion light-years of empty space, a galaxy that once blazed with the fury of a well-fed supermassive black hole has gone quiet. Over roughly two decades of Earth time, the object known as AGN J0218-0036 dimmed by a factor of 20 or more, a collapse in brightness so steep and so fast that it defies the standard playbook for how these cosmic engines wind down.
The finding, led by Tomoki Morokuma of the Chiba Institute of Technology and published in the journal PASJ, documents what may be one of the clearest cases ever recorded of a quasar switching off in something close to real time. Most theorists expected that process to unfold over millions of years. This one appears to have done it in less than a single human generation.
A galaxy caught mid-fadeout
The story begins with archival images taken in the early 2000s, when J0218-0036 still looked like a typical active galactic nucleus, its central black hole gorging on infalling gas and radiating intensely across multiple wavelengths. Later observations collected through the Subaru Hyper Suprime-Cam (HSC) survey, a wide-field instrument whose calibration is detailed in a dedicated technical overview, revealed that the galaxy’s light output had dropped dramatically. Multiwavelength data extending through 2023 filled in the intervening years and traced a consistent downward slide in brightness.
The peer-reviewed paper reports that J0218-0036, sitting at redshift z = 1.767, faded by roughly 20 to 30 times in optical and near-infrared light over about 20 observed years. An author-posted preprint of the same work goes further: after subtracting the background glow of the host galaxy’s stars, the active nucleus alone appears to have faded by roughly 50 times between the early 2000s and 2023. The team attributes the dramatic decline to a sharp drop in the rate at which gas spirals onto the supermassive black hole.
Those two numbers, 20-fold and 50-fold, are not contradictory. The difference reflects what is being measured: total galaxy light in one case, the isolated nuclear engine in the other. But the gap matters, because the larger figure implies a more extreme collapse in the black hole’s fuel supply.
Why the timing is so unusual
A redshift of 1.767 places J0218-0036 at “cosmic noon,” the epoch roughly 10 to 11 billion years ago when galaxies across the universe were forming stars and feeding their central black holes at peak rates. Catching a quasar shutting down during this era is rare. Most known examples of so-called changing-look active galaxies sit at much lower redshifts, close enough that astronomers can revisit them repeatedly with modest telescopes. Spotting one this far away required the wide-field reach and sensitivity of the Subaru HSC survey.
There is another wrinkle that makes the timeline even more striking. Because the universe has been expanding during the billions of years it took this galaxy’s light to reach us, the signal is stretched by cosmological time dilation. A 20-year dimming observed from Earth corresponds to only about seven or eight years in the galaxy’s own rest frame. Whatever shut off the fuel supply did so with startling speed.
For context, many quasars flicker by factors of two or three over years as turbulent structures churn through their accretion disks. A 20-fold decline, let alone a 50-fold one, over two decades sits far outside that normal range. As independent coverage of the result noted, the team’s modeling favors a rapid shutdown of gas inflow rather than the slow, stochastic flickering that characterizes ordinary quasar variability.
What the data still cannot tell us
The dimming itself is well established across multiple epochs and wavelengths, but several pieces of the puzzle remain missing.
No public X-ray or radio monitoring data from the 2000-to-2023 window have been reported in the study. X-rays trace the innermost regions of the accretion flow, the zone closest to the black hole’s event horizon, and could reveal whether the central engine truly powered down or merely shifted into a different emission state. Radio observations, meanwhile, could show whether any relativistic jets responded to the apparent fuel cutoff. Without either data set, the picture of what happened to the black hole’s immediate environment is incomplete.
Spectroscopic follow-up after 2023 has not been published, leaving open the question of whether the broad emission lines characteristic of an active quasar have vanished entirely or simply weakened. Those lines, produced by gas whipping around at thousands of kilometers per second near the black hole, are a key diagnostic in other changing-look AGN. Without time-resolved spectra, J0218-0036 cannot yet be placed cleanly into that category.
Then there is the dust question. A dense cloud of interstellar dust drifting between the nucleus and our line of sight could, in principle, block much of the optical and near-infrared light while leaving the black hole’s intrinsic power unchanged. The published study argues against this scenario based on the color evolution of the fading and the smoothness of the light curve, both of which look more consistent with a genuine drop in luminosity than with patchy obscuration. But without mid-infrared monitoring, which would be far less affected by dust, the door remains slightly open. Mid-infrared data would also reveal whether the surrounding dust is re-radiating energy it absorbed during the galaxy’s brighter days, a delayed “echo” that standard dust-reprocessing models predict should peak several years after the optical decline.
Finally, the raw light-curve tables from the original Subaru observations have not been publicly released. Independent verification currently depends on the summary figures and statistics in the paper rather than on photometry that other groups could re-analyze from scratch.
What comes next for J0218-0036
The strongest evidence in hand is the peer-reviewed PASJ paper itself, which presents the multiwavelength analysis in detail and passed formal review. The arXiv preprint offers additional discussion and what appear to be clearer figures, though it has not undergone the same editorial scrutiny. Both versions converge on the same qualitative conclusion: something caused the black hole’s accretion rate to plummet on a timescale that existing models struggle to accommodate.
The Subaru HSC survey’s well-documented design and calibration pipeline reduce the chance that the fade is an instrument artifact. And the long baseline, spanning roughly two decades, argues against a one-off calibration error. The decline appears gradually across multiple epochs rather than as a sudden jump between two mismatched data sets.
Still, the most cautious reading as of June 2026 is that J0218-0036 is an unusually distant and extreme candidate for a quasar caught in transition. The observed dimming is secure, the timescale is short in cosmological terms, and the preferred interpretation involves a rapid collapse in accretion rate. But alternative explanations cannot be fully excluded until new observations fill the gaps.
The tests that would settle the question are within reach. X-ray telescopes could probe whether the innermost accretion flow has truly gone dark. Infrared space observatories, including the James Webb Space Telescope, could search for the predicted dust echo and pin down the host galaxy’s stellar mass. Time-resolved spectroscopy could track whether the broad emission lines have disappeared or merely faded. If those observations confirm what the optical data suggest, J0218-0036 would stand as one of the most dramatic demonstrations that supermassive black holes can shut off their engines far faster than anyone thought possible, rewriting a chapter of astrophysics that was assumed to play out over geological time.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
