Remarkable new study determines there is 'no evidence' of an accelerating universe

Astronomers piece together the story of our expanding universe, and whether or not that expansion is accelerating, by studying how bright certain stellar explosions look from Earth – Type Ia supernovae.

These massive cosmic eruptions are a great measurement tool because all reach nearly the same peak brightness and follow predictable patterns as they flare up and fade away.

By analyzing each supernova’s color and light curve, scientists can adjust for those details and plot the results on a distance-versus-redshift chart known as the Hubble diagram. That cosmic map reveals how quickly space itself has been expanding over time.

Major paradigm shift

A new study tests a quiet, but key, assumption in that process.

Most analyses have treated all corrected Type Ia supernovae as if they behaved the same way, regardless of where they erupted or when their progenitor stars formed.

The authors of this study show that this is not the case.

After the usual corrections, the standardized brightness still correlates with the average age of stars in the host galaxy: younger progenitors tend to produce slightly fainter standardized events; older ones appear slightly brighter.

Astronomers call the leftover offset a “Hubble residual,” and here it tracks the host age with strong statistical support.

“Our study shows that the universe has already entered a phase of decelerated expansion at the present epoch and that dark energy evolves with time much more rapidly than previously thought,” explained lead researcher Professor Young-Wook Lee, of Yonsei University in South Korea.

“If these results are confirmed, it would mark a major paradigm shift in cosmology since the discovery of dark energy 27 years ago.”

Reconciling an accelerating universe

Observers record a supernova’s light curve – the rise to maximum and the decline afterward – along with its color across filters. Those two features dominate the corrections that bring Type Ia explosions onto a common brightness scale.

Color can reveal dust and intrinsic spectral differences; the light‑curve width indicates how much radioactive nickel the explosion produced, which affects luminosity.

Analysts then fold these parameters into a model and treat the event as a standard candle within quoted uncertainties. In that language, teams “standardize” each supernova using a shared recipe.

Different surveys often use compatible tools to fit the data, but every approach relies on the idea that once analysts correct for color and light‑curve width, the residual brightness should not depend on where the star system lived or on how old its parent population was.

Subtle age trend

Galaxies evolve. Long ago, they formed stars more rapidly, which means the typical stellar population astronomers sample at large cosmic distances skews younger on average.

If, after standardization, younger progenitors are slightly dimmer and older ones are slightly brighter, then a trend with distance appears that is unrelated to the actual expansion geometry. Without accounting for this effect, the analysis can bias the Hubble diagram.

This is exactly the kind of systematic error that can sneak in because distance and lookback time are linked.

The farther you observe, the further back in history you reach. A small age‑linked shift can masquerade as a changed expansion rate if it is not modeled correctly.

Accounting for accelerating universe “bias”

The team estimates how the average age of Type Ia progenitors varies with redshift by combining the universe’s star formation history with the typical delay between a stellar birth and the eventual explosion.

That expectation yields a modest brightness adjustment with distance – nudging standardized brightness slightly fainter as the mean age decreases – calibrated directly against the observed relation between host age and residuals. The correction applies from nearby space out to several billion light‑years.

They also build a special sample that avoids model‑dependent corrections. By selecting supernovae whose hosts are comparably young at both low‑ and high‑redshift, the average host age no longer drifts with distance.

This “no‑evolution” stress test removes the suspected bias by construction. The age‑corrected full sample and the age‑matched subsample agree.

Stated more clearly, even after adjusting for luminosity, supernovae from younger stars consistently appear dimmer, while those from older stellar populations shine more brightly.

Using a much larger sample of 300 host galaxies, the new study confirmed this trend with extremely high confidence – 99.999% – indicating that the apparent dimming of distant supernovae isn’t driven solely by cosmological factors, but also by the underlying physics of the stars themselves.

Combining the data

The authors reanalyze two major modern supernova catalogs and compare them with two other pillars of precision cosmology: the cosmic microwave background and baryon acoustic oscillations.

The three datasets are in notably tighter agreement when dark energy is allowed to vary slowly with time instead of acting as a perfectly constant “cosmological constant.”

The combined picture even hints that, at the present epoch, the universe might not be accelerating the way the standard model says.

The standard framework still works well in many respects, yet a slightly more flexible description fits these data better.

Small nudge to Hubble‑tension

A well‑known puzzle hangs over modern cosmology: local measurements of the Hubble constant run higher than values inferred from early‑universe data.

The local route relies on a “distance ladder,” where Cepheid‑calibrated, nearby Type Ia supernovae set the zero‑point for more distant ones.

If the nearby anchors come from older progenitors on average, and the next rung samples relatively younger systems, the age effect biases the ladder upward.

Correcting for age narrows that gap by a few percent. The discrepancy remains, but the shift reduces the difference and acknowledges extra uncertainty until ages can be estimated on an object‑by‑object basis.

Where do we go from here?

Moving beyond an average correction to clearly prove the universe is not actually accelerating will require measuring host‑galaxy ages for a far larger set of supernovae across the full distance range.

New facilities will supply the galaxy volume: the Rubin Observatory’s decade‑long survey will multiply the Type Ia sample by orders of magnitude, and the Roman Space Telescope will deliver precise distances in the infrared with minimal dust interference.

With ages in hand for individual events, analysts can standardize each supernova with an explicit age term rather than a global trend.

A bigger, age‑aware sample also strengthens cross‑checks with galaxy‑clustering measurements and with the early‑universe snapshot encoded in the microwave background.

If the age trend persists across instruments, sky areas, and selection methods, it will become a standard ingredient in distance work, much like color and light‑curve width today.

Cosmic expansion studies improve when we fold in the physics of where each supernova lived and when its star system formed. The key lesson from this study is clear: the most reliable yardsticks in the sky are not truly “one size fits all.”

The full study was published in the journal Monthly Notices of the Royal Astronomical Society.

—–

Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.

Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.

—–