This Star Is Older Than the Universe — And the Math Might Be Right

There is a star close enough to see with a decent pair of binoculars on a clear night. It sits in the constellation Libra, roughly 190 light-years away, which by galactic standards makes it practically a neighbor. Astronomers have known about it for over a century. It has a catalog number — HD 140283 — and a nickname, Methuselah, borrowed from the oldest human in the Hebrew scriptures. The nickname is not decorative. For years, the best stellar age estimates placed HD 140283 at somewhere between 14 and 16 billion years old. The universe, according to the most precise measurements we have from the cosmic microwave background and the expansion rate of space itself, is 13.8 billion years old. That is not a rounding error. That is a star that appears, by the numbers, to predate everything.

The obvious response — that someone must have made a mistake — turns out to be both correct and deeply unsatisfying, because pinning down exactly where the mistake lives has proven surprisingly hard. Stellar age estimates carry real uncertainties. So do the cosmological measurements that produced that 13.8-billion-year figure. The Methuselah problem is not a single calculation that went wrong. It is a collision between two separate measurement frameworks, each with its own compounding error bars, each refined over decades of careful work, and each stubbornly resistant to producing a perfectly clean answer. When scientists try to reconcile them, they do not find a simple fix. They find a set of deeper questions about how we know what we think we know about the age of everything.

HD 140283 belongs to a class of stars called metal-poor subgiants. In astronomy, "metal" means any element heavier than helium — a terminology that sounds wrong to a chemist but reflects the universe's actual composition history. The first stars formed almost entirely from hydrogen and helium, the only elements the Big Bang produced in significant quantities. Heavier elements — oxygen, carbon, iron — were forged later, inside stars, and scattered through supernovae. A star with very little iron and very few heavy elements is, by inference, ancient: it formed early, before the galaxy had seeded itself with much stellar debris. HD 140283 has an iron abundance roughly 250 times lower than the Sun's. That alone marks it as a relic.

Its orbit reinforces that picture. The star does not travel in a calm circular path through the disk of the Milky Way the way younger stars tend to. It moves in a long, looping trajectory that carries it through the galactic halo, the sparse outer region populated by the oldest structures in the galaxy. It is passing through our neighborhood at roughly 800,000 kilometers per hour relative to the Sun, a hypervelocity that reflects a gravitational history far older and more chaotic than anything in the orderly stellar disk. HD 140283 has the biography of something ancient. The question is how ancient, and whether the answer breaks physics.

How We Age a Star

Aging a star is not like reading a label. There is no direct clock. What astrophysicists do instead is build theoretical models of how stars evolve — how they burn through hydrogen in their cores, how their luminosity and surface temperature change over millions and billions of years — and then compare those models to what they actually observe. The key observable for age-dating is the position of a star on the Hertzsprung-Russell diagram, particularly the point called the main sequence turnoff, where a star begins transitioning away from stable hydrogen burning as its core fuel runs low. The more evolved a star is toward that turnoff, the older it is. The problem is that determining luminosity precisely requires knowing the star's distance precisely, and for most of stellar history, that was genuinely difficult.

The European Space Agency's Hipparcos mission, launched in 1989, gave astronomers their first high-quality parallax measurements for a large number of nearby stars — parallax being the slight apparent shift in a star's position as Earth moves around the Sun, from which distance can be calculated geometrically. When Hipparcos measured HD 140283's parallax, the resulting distance implied a luminosity that pushed the age estimate above 14 billion years. That number alarmed people. When the Hubble Space Telescope subsequently measured the distance more carefully using the Fine Guidance Sensors^[1] — instruments designed to stabilize the telescope but sensitive enough to do precision astrometry — the parallax came in slightly different, shifting the age estimate downward, though still into uncomfortable territory. The revised figure settled near 14.46 billion years, with an uncertainty range extending down to roughly 13.4 billion. That lower bound is just barely consistent with a 13.8-billion-year-old universe. Just barely.

“The revised age of HD 140283 was just barely consistent with the age of the universe — and 'just barely' is doing an enormous amount of work in that sentence.”

The uncertainty range matters enormously here, and it is worth understanding what goes into it. Stellar evolution models depend on the star's chemical composition, which is inferred from spectroscopy — the way the star's light is absorbed at specific wavelengths by atoms in its atmosphere. They depend on the assumed helium abundance, which cannot be directly measured from the spectrum of a cool star and must instead be estimated from models of Big Bang nucleosynthesis. They depend on how efficiently energy moves through the stellar interior, a process governed by opacity calculations that are themselves dependent on composition. Each of those inputs carries uncertainty, and the uncertainties compound. A 2013 analysis using Hubble data^[1] drove the central age estimate down meaningfully by refining the distance, but it did not make the uncertainty disappear. What it did was move the problem from "definitely impossible" to "possibly consistent, depending on where all the uncertainties actually land."

The Other Clock: How We Dated the Universe Itself

The 13.8-billion-year figure for the age of the universe comes primarily from analysis of the cosmic microwave background radiation — the faint glow of microwaves that fills the sky in every direction, the afterglow of the hot plasma that filled space when the universe was roughly 380,000 years old. The Planck satellite, operated by ESA and collecting data from 2009 to 2013, mapped that radiation with extraordinary precision, measuring the tiny temperature fluctuations that encode information about the geometry, composition, and expansion history of the universe. From those measurements, within the framework of the standard cosmological model — Lambda-CDM, which describes a universe containing ordinary matter, dark matter, and a cosmological constant driving accelerated expansion — Planck extracted a Hubble constant of approximately 67.4 kilometers per second per megaparsec^[2]. The Hubble constant is the rate at which the universe expands, and it is the key ingredient in calculating how long expansion has been going on.

This is where things become interesting in a way that extends well beyond HD 140283. There are two primary methods for measuring the Hubble constant, and they do not agree. The Planck CMB method gives roughly 67 to 68. The other method — measuring the distances to galaxies directly, using a chain of distance indicators that runs from Cepheid variable stars through Type Ia supernovae — consistently produces values in the range of 72 to 74. That discrepancy, now known as the Hubble tension, has survived years of additional measurement and systematic-error hunting. It is not going away, and it matters for the Methuselah problem directly. A higher Hubble constant means the universe has been expanding faster, which means it reached its current size in less time, which means it is younger. If the true Hubble constant is closer to 73 than to 67, the universe might be more like 12.5 to 13 billion years old — which makes HD 140283's age estimate even more difficult to accommodate, not less.

“If the Hubble constant is higher than Planck suggests, the universe gets younger — and HD 140283's age problem gets worse, not better.”

Some proposed resolutions to the Hubble tension involve modifications to the early universe that would change the sound horizon scale encoded in the CMB — the precise size of the acoustic oscillations that Planck measured. Early dark energy models, for instance, posit a brief period of additional energy density in the early universe that would shift the CMB-derived Hubble constant upward without breaking the overall fit to the data. Others involve interacting dark matter scenarios or new relativistic species. None of these are established. All of them are active research programs, and the James Webb Space Telescope is now providing independent measurements of Cepheid distances that will either tighten or further complicate the picture. So far, early JWST results have not resolved the tension. If anything, they have sharpened it.

What the Star's Chemistry Actually Tells Us

HD 140283 is not simply old — it is also scientifically legible in ways that make it a laboratory for early-universe chemistry. High-resolution spectroscopy from large ground-based telescopes, including instruments on the Very Large Telescope in Chile and the Keck Observatory in Hawaii, has allowed astronomers to measure its atmospheric abundances with remarkable precision. The star shows an oxygen-to-iron ratio significantly higher than the Sun's, which is characteristic of material enriched by massive stars that ended as core-collapse supernovae before significant iron production from Type Ia supernovae had accumulated in the galactic environment. That chemical signature is consistent with the star forming from gas enriched by only one or a few generations of massive stellar predecessors — meaning it formed very early in the galaxy's history, when the interstellar medium was still nearly pristine.

The star also shows a lithium abundance — or more precisely, a relative lack of it — that touches one of the older unsolved problems in astrophysics. Big Bang nucleosynthesis predicts a specific amount of lithium-7 should have been produced in the first few minutes after the Big Bang. But old metal-poor stars like HD 140283 consistently show lithium abundances about three times lower than that prediction. This is the cosmological lithium problem, and it has been unresolved for decades. It could mean that lithium was destroyed inside these old stars over their long lifetimes by mixing into hotter interior layers. It could mean the BBN predictions have a systematic error. It could mean there was some non-standard physics operating in the early universe. The Methuselah star is not just an age puzzle — it is a chemistry puzzle sitting on top of an age puzzle.

Where the Uncertainty Actually Lives

Astronomers who work on stellar ages tend to be precise about where their uncertainty comes from, and it is worth following that precision rather than letting the problem blur into a vague cloud of "we don't know." The biggest remaining source of uncertainty in HD 140283's age is almost certainly the oxygen abundance. Oxygen affects stellar opacity and the equation of state inside the star, and a higher oxygen abundance shifts model isochrones — the theoretical lines of equal age on the HR diagram — in ways that reduce the inferred age. Some analyses, incorporating revised oxygen measurements, have pushed the central estimate closer to 13.7 billion years, with error bars that make contact with 13.4. Others, using slightly different spectral analysis methods or model atmosphere grids, land higher. The star's age is not precisely known. What is known is that it is genuinely ancient, that it formed very early in the history of the Milky Way, and that the lower end of its plausible age range is at least technically consistent with the universe's age — provided you accept the Planck CMB value and hold the model assumptions fixed.

The Gaia space observatory, ESA's ongoing astrometric mission, has now measured HD 140283's parallax with extraordinary precision — better than Hipparcos, better than the Hubble Fine Guidance Sensors. The Gaia Data Release 3 figures^[3] pin the star at approximately 190.1 light-years, with an uncertainty small enough to constrain the luminosity significantly. When stellar evolution models are run against that distance and against the best current spectroscopic measurements of the star's composition, the age still comes in old — very old — but the central value has migrated downward compared to earlier estimates. The word "migration" matters here: the star has not been proven young. It has been proven consistent with being around 13.5 to 14.5 billion years old, which is still a range that requires careful handling against the cosmological timeline.

“Gaia didn't prove Methuselah is younger than we feared — it proved the uncertainty is finally small enough to matter.”

Why a Single Star Puts Pressure on Everything

A cosmological model is only as trustworthy as its most uncomfortable boundary case. HD 140283 is that boundary case. When age estimates for the oldest stars exceed the age of the universe derived from expansion history, that is not a public relations problem for astronomy — it is a signal that at least one of the measurements, or one of the model assumptions behind a measurement, is wrong. The history of cosmology is full of exactly this dynamic. The original Hubble constant measurements from the mid-twentieth century implied a universe younger than the oldest globular clusters. That discrepancy drove decades of distance-scale revision and eventually contributed to the discovery of dark energy, when supernovae observations in the late 1990s showed that the universe's expansion was not decelerating as expected but accelerating. The Methuselah problem is structurally similar: a star that seems too old is telling us something real about either stellar physics or cosmology or both, and the work of figuring out which one is ongoing.

What makes the current situation genuinely interesting — as opposed to simply uncomfortable — is that the tools available to resolve it are better than they have ever been. JWST is measuring stellar populations in distant galaxies with enough precision to constrain stellar evolution models independently. Gaia continues refining distances for thousands of metal-poor halo stars, building a statistical picture of galactic archaeology that no individual parallax measurement could provide alone. Ground-based spectrographs with resolving power unavailable a decade ago are producing chemical abundance measurements that constrain stellar interiors more tightly. And the Hubble tension, rather than fading into a known systematic, has hardened into a genuine crisis that is forcing theorists to consider modifications to the standard cosmological model that would have seemed radical ten years ago. HD 140283 did not cause any of these pressures. But it sits at the intersection of all of them — a nearby, well-studied, persistently awkward star that keeps asking the same question the field keeps struggling to answer cleanly: how old is old, and how certain are we about the ruler we used to measure it?

The star itself is indifferent to the argument. It has been moving through the galaxy's halo since before the Milky Way's disk had fully assembled, burning slowly on what little hydrogen it has left, its core gradually contracting toward the subgiant phase that will eventually swell it into a red giant. Whatever the exact number turns out to be — 13.5 billion years, 14 billion, somewhere in between — HD 140283 formed from gas that was already ancient by the time the Sun's nebula began to collapse. It is not impossible that it predates every star currently visible in our night sky by a billion years or more. The age of the universe does not feel abstract when something that old is close enough to resolve with instruments we actually built and launched and pointed at it. The uncertainty that remains is not a failure of nerve or precision. It is what honest measurement looks like at the edge of what the cosmos will tell us so far.

References

1. HD 140283: A Star in the Solar Neighborhood that Formed Shortly After the Big Bang (arxiv.org)
   Provides Hubble Space Telescope parallax measurement of HD 140283 that refined distance and age estimate to 14.46 billion years with uncertainty range.
2. Planck 2018 results. VI. Cosmological parameters (arxiv.org)
   Supplies the Planck satellite's Hubble constant measurement of 67.4 km/s/Mpc used to calculate the universe's age of 13.8 billion years.
3. The age of the Methuselah star in the light of stellar evolution models with tailored abundances (aanda.org)
   Provides updated parallax and distance measurements for HD 140283 used to refine stellar age estimates in recent analyses.