Canada's CHIME Telescope Just Detected the Glow of the 5-Billion-Year-Old Universe, On Its Own

High in the hills above Penticton, British Columbia, four giant metal troughs stare at the sky all day, every day. Last week, the team behind them announced something cosmologists have been chasing for more than a decade: the Canadian Hydrogen Intensity Mapping Experiment, CHIME, has detected the faint radio glow of hydrogen gas from the deep cosmic past using nothing but its own data. The result, published in The Astrophysical Journal on September 28–29, 2026, is what team leader Mark Halpern of the University of British Columbia called "a bold new step in the global cosmology program and a Canadian success story."

Here's what happened, what it means, and why it matters, in plain language.

Key takeaways

• CHIME detected hydrogen's characteristic 21-centimetre radio glow from when the universe was about 5 billion years old (light travelled ~8 billion years to reach us), at a statistical strength of 12.4 sigma, an unambiguous detection.
• This is the first standalone (auto-correlation) detection at this distance: CHIME found the signal in its own data alone, without checking against galaxy catalogues made by other telescopes.
• The signal was buried under interference up to 10,000 times stronger, Milky Way radio static and human-made radio noise. The team sifted 94 clean nights out of roughly 2,000 days of observing.
• The payoff: a cheaper, faster way to map the universe's large-scale structure and study dark energy, the mysterious force accelerating cosmic expansion.
• One caveat: the filtering used to remove interference also removed the exact spatial patterns needed to measure the expansion rate directly. That measurement is the next mountain.

What CHIME actually did

Every element has a fingerprint in radio light, and hydrogen's is a signal with a wavelength of exactly 21 centimetres. (Why 21 cm? It comes from a tiny flip in the hydrogen atom's electron, think of it as the atom clearing its throat at one very specific pitch.) Because the universe is expanding, that signal gets stretched to longer wavelengths the farther it travels, the same way a police siren sounds lower-pitched as it drives away. This stretching is called redshift, and it tells astronomers how far back in time a signal comes from.

CHIME detected this stretched hydrogen glow at a redshift of about 1.16. In plain terms: the light left its source when the universe was roughly 5 billion years old, a little more than a third of its current age of 13.8 billion, and travelled for about 8 billion years before reaching British Columbia. At that epoch, the team estimates, roughly two percent of the universe's hydrogen was in the form of neutral atomic gas, the raw material stars are made of.

The detection was reported at 12.4 sigma. "Sigma" is how physicists measure certainty: 5 sigma is the gold standard for "this is real," so 12.4 is emphatically not a fluke.

Why "on its own" is the big deal

To understand the achievement, you need to know what hydrogen intensity mapping is, and why astronomers have always needed a crutch.

Traditional galaxy surveys find individual galaxies one by one, like photographing every tree in a forest to map the forest. Intensity mapping instead measures the total radio glow from huge patches of sky containing many galaxies at once, like sensing the overall greenness of the forest from above. It's faster and cheaper, but the signal is absurdly faint.

The problem has always been the foregrounds. Our own Milky Way blasts out radio static (called synchrotron emission) that is up to 10,000 times brighter than the hydrogen glow, and human radio interference adds more. Until now, the field's workaround was cross-correlation: pair the messy radio map with a catalogue of galaxies made by an optical telescope, and the noise, which lives in only one dataset, cancels out. CHIME did this itself in 2023 and 2024, stacking its data against galaxy surveys at redshifts up to 2.3.

What changed now is auto-correlation: CHIME found the signal using only CHIME data, with no external catalogue to lean on. That's harder, and it's more valuable, because an independent measurement can disagree with existing surveys instead of being forced to echo them. As recently as 2022, working cosmologists could still say no experiment had a clean auto-detection at all. CHIME's paper acknowledges earlier auto-detection claims by South Africa's MeerKAT telescope at lower redshifts (0.32 and 0.44) but frames its own as the first at redshift ~1.

The cost of going solo: the team selected just 94 interference-free nights from roughly 2,000 days of archived observations, then spent over a year, in some accounts more than two, testing and re-testing that the faint signal wasn't an instrument glitch or leftover interference. Only after that gauntlet did the 12.4-sigma result stand.

How this helps study dark energy

Dark energy is the mysterious force pushing the universe's expansion to accelerate. Nobody knows what it is, but one of the best ways to pin it down is to measure exactly how fast the universe was expanding at different times in its history.

Here's the trick astronomers use. In the early universe, sound waves rippled through hot gas and left a faint pattern in how matter is distributed, slightly denser and thinner regions on scales of about 500 million light-years. These are called baryon acoustic oscillations, or BAO. Because the original size of the pattern is known (measured precisely by the Planck satellite), it works as a standard ruler: measure how big the pattern looks at different cosmic times, and you can work out how much the universe has expanded since. That expansion history is the fingerprint of dark energy.

Hydrogen is the ideal tracer for this job because it follows the same web of matter as galaxies do. And intensity mapping can cover enormous volumes of the universe; CHIME is designed to eventually map over 3% of the entire observable universe, far more than any galaxy-by-galaxy survey, during the epoch when dark energy first started to matter.

The current result is a proof of concept, not yet the payoff: to kill the foregrounds, the team had to throw away exactly the spatial patterns that carry the expansion measurement. So CHIME has proven it can see the glow on its own; the next job is recovering those patterns so the data can actually constrain dark energy. The collaboration, with teams in Canada, the US and India, including the Raman Research Institute, plans to apply the same techniques to nearly seven years of accumulated data to push to earlier cosmic times.

The Canadian story

This one is worth a moment of national pride. CHIME sits at the Dominion Radio Astrophysical Observatory near Penticton, BC. It was completed in September 2017 and looks like nothing else in astronomy: four fixed, half-pipe-shaped reflectors with no moving parts. It doesn't point, the Earth's rotation sweeps the sky across it, and it watches the entire northern sky every single day. It's led by the University of British Columbia, with the University of Toronto, McGill and others in the collaboration.

"It’s a bold new step in the global cosmology program and a Canadian success story," said Mark Halpern, the UBC professor who co-leads the experiment. Co-author Arnab Chakraborty, a University of Toronto postdoctoral fellow, put the science case simply: "Hydrogen is the most common element in the universe and the raw material from which stars form. Its faint radio emission acts like a cosmic tracer, revealing how matter is distributed across space."

There's a policy footnote too. CHIME was built as a dedicated hydrogen mapper, a relatively cheap, specialized instrument, rather than a billion-dollar general-purpose observatory. If intensity mapping works at scale, it offers a faster, lower-cost alternative to galaxy surveys for cosmology's biggest questions. That's the bet the Canadian-led team just validated.

A sense of scale

To appreciate the 12.4-sigma detection, consider what the team was up against:

• 2,000 days of observations sifted down to 94 clean nights, less than 5% of the data survived the interference cut.
• Foregrounds up to 10,000 times brighter than the signal had to be removed.
• Over a year of cross-checks to rule out false alarms from instruments and radio-frequency interference.
• Radio interference around the observatory is growing over time, making the analysis harder with each passing year, the team is racing local radio noise, not just the cosmos.

And the distance: redshift 1.16 means the hydrogen CHIME detected was glowing when the universe was about 5 billion years old, before the Sun and Earth existed (those came ~4.6 billion years ago), and when the galaxies whose collective glow CHIME measured were young.

Practical next steps

• Read the coverage, not just the headlines: the result is in The Astrophysical Journal (Sept 28–29, 2026); accessible write-ups include The Brighter Side and ScienMag's explainers.
• Watch for the follow-up: the collaboration's next papers will apply the same auto-correlation technique to nearly seven years of data, and the real prize will be the first dark-energy constraint from CHIME's solo data.
• If you're in BC: the Dominion Radio Astrophysical Observatory near Penticton hosts occasional public open houses, the most Canadian way to see cosmology in action.
• For the night-sky curious: CHIME's whole project starts with the same dark skies stargazers chase. Our guide at northern lights viewing guide for Canada covers where Canada's dark skies are best this season.

The bottom line

A Canadian telescope with no moving parts just heard the universe's hydrogen whisper from when the cosmos was 5 billion years old, and it did it without any other telescope's help. The detection itself is a landmark; the prize it unlocks is a cheaper, faster way to map the universe and probe dark energy. The expansion-rate measurement isn't there yet, but the path is. For a country that built its cosmology program on bold, specialized instruments, this one is exactly what it looks like: a Canadian success story, 12.4 sigma strong.

Sources

• Particle News, CHIME Detects Faint 21-cm Hydrogen From About 8 Billion Light-Years
• The iBulletin, CHIME Detects 21 cm Hydrogen Signal at 12.4 Sigma
• The Brighter Side, Canadian radio telescope (CHIME) is turning hydrogen into a map of cosmic expansion
• ScienMag, Canadian Telescope CHIME Captures Universe's Earliest Hydrogen Glow in Landmark First
• Wikipedia, Canadian Hydrogen Intensity Mapping Experiment