Cosmic hydrogen from 5 billion light-years was hidden in MeerKAT data since 2018

When Sourabh Paul and his collaborators turned to the archive, they were looking for a signal most astronomers assumed would need two telescopes to find. What emerged from 96 hours of MeerKAT radio data was a faint 21-centimeter whisper, the specific radio signature of neutral hydrogen, carrying information from two moments in cosmic history when the universe was roughly nine and ten billion years old. The signal had crossed 3.67 and 4.76 billion light-years to reach the 64 dishes arrayed across South Africa’s Northern Cape.

The distance matters less than what it took to hear it. Every previous attempt to detect hydrogen this way required cross-referencing the radio observations with a catalog of galaxies identified by an optical telescope. This time, the MeerKAT team needed no such catalog. The detection, published in The Astrophysical Journal Letters, is the first direct hydrogen intensity mapping result from a single radio observatory working alone.

How they heard it

Neutral hydrogen atoms emit radio waves at a precise wavelength of 21 centimeters, produced when the single electron in a hydrogen atom flips its spin. It is one of the most abundant signals in the universe, but also one of the most diluted over cosmic distances. Hydrogen intensity mapping does not try to resolve individual galaxies. It measures the combined radio emission from millions of unresolved galaxies simultaneously. “With intensity mapping, we do not need to detect every individual galaxy,” said Dr. Zhaoting Chen, a co-author on the paper. The technique is closer to hearing the background noise of a crowded room than to picking out a single conversation.

Recommended ReadingStar birth is down 60% — the universe’s hydrogen supply is mostly intact

The obstacle is foreground contamination. The Milky Way’s own radio emissions are many orders of magnitude stronger than the cosmological hydrogen signal. Terrestrial interference, instrumental artifacts, and radio-emitting sources in the nearby universe all compete with the faint whisper from billions of light-years away. The team developed calibration methods to separate what belongs to our galaxy from what belongs to the deep past, working entirely with MeerKAT’s radio archive.

The result covers two redshift windows, z ≈ 0.32 and z ≈ 0.44, corresponding to hydrogen variations measured across scales of several million light-years, roughly comparable to the separation between the Milky Way and Andromeda. It is a narrow slice of cosmic history. But it is a real one, extracted without the support of any optical data.

What required two telescopes before

The previous standard approach, cross-correlating radio data with an optical galaxy survey, worked because optical redshifts give precise positional anchors for where the galaxies are. The radio signal is too faint to attribute to specific sources on its own, so pairing it with an optical catalog let astronomers verify that what they were seeing was genuine. The cross-correlation technique produced detections, but it imposed a ceiling: hydrogen could only be mapped where an optical survey had already cataloged the sky.

Direct detection lifts that ceiling. A radio observatory can now survey hydrogen across regions of sky and epochs of cosmic time where no optical survey exists, or where optical surveys cannot reach because the galaxies are too faint. The cosmic web, the filaments and voids of dark matter and ordinary matter that give the universe its large-scale structure, can be traced through hydrogen regardless of whether optical surveys have covered the same patch.

Lead author Dr. Sourabh Paul put it directly: “Detecting it directly with MeerKAT shows that this technique is becoming a practical tool for cosmology.”

What the signal does not yet tell us

Two redshift windows from a single 96-hour dataset are not a map of the universe. The current result covers a limited patch of sky and two specific slices of cosmic time. Extended surveys, covering more sky and more epochs, are required before hydrogen intensity mapping can constrain the cosmological parameters that the technique was designed to measure: the Hubble constant, dark energy density, the geometry of the universe at different ages.

The foreground subtraction, while successful, introduces systematic uncertainties that are difficult to quantify precisely. Any imperfection in removing the Milky Way’s own emission leaves a residual that could contaminate the cosmological signal. Independent verification with different observatories or with fresh MeerKAT data would help settle the question of signal versus artefact.

The two redshift windows reached in this detection correspond to the universe at roughly 70 and 66 percent of its current age. These are not the epochs where dark energy and structure formation create the most useful cosmological contrasts. Future observations pushing to higher redshifts, where the competition between dark energy and matter density is most visible in how structure grows, will be the real test of hydrogen intensity mapping as a cosmological probe.

Common questions about hydrogen intensity mapping

What is the 21-centimeter line and why does it matter?

Neutral hydrogen atoms emit a faint radio signal at exactly 21 centimeters wavelength, caused by an energy transition in the atom’s single electron. Because hydrogen is the universe’s most abundant element, this emission is everywhere in the cosmos. As the universe expands, the signal stretches to longer wavelengths. Measuring that stretch tells astronomers both the distance of the source and when the light left. It is the universe’s built-in broadcast signal, transmitted continuously since the first atoms formed.

Why hadn’t MeerKAT made this detection earlier?

The data were collected in the telescope’s first year of science operations. Identifying a workable analysis method, developing the foreground removal techniques, and applying them to the archival dataset took years of calibration work. The signal was always in the data. Extracting it cleanly required the methodology to catch up. Professor Mario Santos noted that it is “particularly remarkable” the detection came from data collected so early in MeerKAT’s operational life.

What is the Square Kilometre Array, and how does this result inform it?

The Square Kilometre Array Observatory (SKAO), under construction in South Africa and Australia, will have roughly 50 times more collecting area than MeerKAT. Hydrogen intensity mapping is expected to be one of its central science programs. MeerKAT is the SKAO’s direct precursor. Demonstrating that direct hydrogen intensity mapping works in practice gives SKAO science teams validated methods and a baseline against which to plan their first surveys. Professor Laura Wolz called the MeerKAT result a pointer “to future observations with SKAO.”

Does hydrogen intensity mapping detect dark matter?

Not directly. The 21-cm signal comes from ordinary hydrogen atoms. What it traces is the large-scale distribution of ordinary matter, which follows the gravitational scaffolding built by dark matter over cosmic history. By comparing hydrogen maps to galaxy distribution and gravitational lensing data, cosmologists can infer the dark matter density and clustering without detecting dark matter particles themselves.

The team plans to extend the survey to wider sky coverage and longer observing runs, pushing the detection to higher redshifts and denser hydrogen maps. MeerKAT’s archive holds data from its first years of operations not yet analyzed for intensity mapping. The signal heard from 2018 observations may be the first of several waiting to be found.

Reference: Paul, S., Chen, Z., Santos, M.G., and Wolz, L., “A Direct Detection of Neutral Hydrogen Intensity Mapping on Mpc Scales at z ≈ 0.32 and z ≈ 0.44,” The Astrophysical Journal Letters, 2026. DOI: 10.3847/2041-8213/ae808f

Tags: astronomy, 21-cm line, hydrogen intensity mapping, MeerKAT, radio astronomy, SKAO

More Like This

Sagittarius A*: NASA Telescopes Support Event Horizon Telescope in Studying Milky Way’s Black Hole

JWST found ancient galaxies hiding four times their measured mass in small stars

The gaps astronomers used to trace dark matter are made by the Milky Way itself

JWST found water at Sagittarius A*’s doorstep — galactic cores do chemistry after all

James Webb mapped 164,000 galaxies to chart the cosmic web back a billion years

Roman will survey 1 billion galaxies to crack dark energy — it launches Saturday