Astronomers find a gravitational-wave imprint that may be the first “dark star” signal from the dawn of the Universe
For years, astronomers have been listening to a faint, almost imperceptible hum in the fabric of spacetime — a background tremor at nanohertz frequencies, picked up by networks of millisecond pulsars scattered across the galaxy. The leading explanation has always been familiar: pairs of supermassive black holes, locked in slow gravitational spirals, generating ripples that accumulate into a cosmic murmur.
Now a new study suggests that hum may carry something far older embedded within it — the fingerprints of hypothetical stars that burned not on nuclear fusion, but on dark matter itself, more than 13 billion years ago.
A hum in the fabric of spacetime
The gravitational-wave background is fundamentally different from the dramatic single-event detections that made headlines when LIGO first picked up colliding black holes in 2015. Rather than a sharp signal from one cosmic collision, it’s a continuous, overlapping wash of waves from countless sources — ambient noise rather than a single note.
Pulsar Timing Arrays are the instruments built to hear it. Millisecond pulsars spin with extraordinary regularity, making them natural cosmic clocks. When gravitational waves pass through the galaxy, they shift the arrival times of pulsar radio pulses by tiny fractions of a second. By monitoring dozens of pulsars simultaneously over years, international PTA collaborations have confirmed a stochastic background at nanohertz frequencies.
The most accepted explanation has been populations of supermassive black hole binaries generating overlapping gravitational ripples across the relatively recent Universe. That explanation remains credible — but it may not be the complete picture.
The mystery of the first giant black holes
Here’s the deeper puzzle. For the PTA signal to look the way it does, it needs contributions from binary systems with combined masses above roughly a billion solar masses. That points back to an older, unresolved question: how did black holes get that enormous so early in cosmic history?
Observations from the James Webb Space Telescope and the Chandra X-ray Observatory have revealed unexpectedly massive black holes already in place when the Universe was very young. Their existence implies that seed black holes must have formed quickly and grown fast — faster than standard formation models comfortably explain.
Researchers Sohan Ghodla and Cosmin Ilie, based at Colgate University, focused on two candidate seed mechanisms: direct collapse black holes and black holes born from the collapse of supermassive Dark Stars. Their study, published as a Letter in Physical Review D, asked whether descendants of those ancient seeds could still be shaping the gravitational-wave signal we measure today.
What are Dark Stars — and why do they matter?
Dark Stars are hypothetical primordial objects unlike anything that exists in the present-day Universe. Rather than burning through nuclear fusion, they’d be powered primarily by heat generated through dark matter — specifically, in the WIMP scenario the study considers, by the annihilation of weakly interacting massive particles trapped within the star. This unusual energy source would keep them relatively cool and diffuse, letting them accumulate mass without collapsing prematurely.
Under the right conditions, a Dark Star might grow to a million times the mass of the Sun before finally collapsing into a massive black hole seed.
Direct collapse black holes — the other candidate mechanism — appear far less promising. The study found their characteristic number density to be around 10⁻⁶ per cubic megaparsec, making them too rare to contribute significantly to the PTA signal. Dark Star remnants, modeled at roughly 10⁻³ per cubic megaparsec, could dominate it.
Dark Stars have never been directly observed. They remain a theoretical construct, and this study doesn’t change that. What it does suggest is that their indirect fingerprints may already be hiding in data we’ve already collected.
How the signal connects past to present
Ghodla and Ilie didn’t simply propose that Dark Stars existed — they traced a plausible path from cosmic dawn to the present. They modeled how early black hole seeds would have grown alongside their host dark matter halos, how frequently those halos would have merged, and what gravitational-wave output those mergers would eventually produce.
The key result is a kind of Goldilocks constraint. As Ghodla put it, produce too many massive seeds and you over-produce the PTA signal; produce too few, and you need other unexplained mechanisms to assemble supermassive black holes later. Seed densities in the 10⁻² to 10⁻¹ per cubic megaparsec range would already exceed what current PTA observations allow.
That constraint is more useful than it might sound. PTA data — collected from mergers happening in the relatively recent Universe — can still place limits on populations of objects that existed at redshifts greater than 10, more than 13 billion years ago. The signal carries memory of its origins.
A new window onto cosmic dawn
The broader implication of this work is that gravitational-wave astronomy, dark matter physics, and the study of the first luminous objects are no longer separate conversations. PTA measurements now offer a way to probe populations from cosmic dawn — not by seeing them directly, but by reading what their descendants left behind.
“Dark Stars were originally proposed as objects that might be seen directly at cosmic dawn,” Ilie said. “This work points to a completely different way of testing their possible role in cosmic history. Their descendants could leave a gravitational-wave imprint that persists all the way to the present-day Universe.”
As PTA networks grow more sensitive and JWST continues to reveal early black holes in unexpected places, the range of viable formation models may narrow considerably. The hum in spacetime has been growing louder for years — and it may be telling a much older story than anyone anticipated.
More information is available here: Sohan Ghodla, Cosmin Ilie. Reconstructing PTA measurements via early seeding of supermassive black holes. Physical Review D, 2026; 114 (4) DOI: 10.1103/hvfd-8fkr
Read the whole thing?
Get the week's signal, not the noise
Our sharpest reporting on energy, climate and nature — free, once a week.
