BALTIMORE — Dwarf galaxies are leaking the very element their stars work so hard to make, according to a new census reported on 2 November 2026, which found that 37 nearby dwarf galaxies keep only between 4 and 25 per cent of the oxygen produced by their stars, with an average of just 11 per cent.
The study, from the Galaxies Losing Oxygen via Winds (GLOW) project, used NASA's Hubble Space Telescope and other observatories to track oxygen — the most abundant element after hydrogen and helium — through stars, gas and dust in small, gas-rich galaxies in the nearby universe. The missing oxygen, the researchers say, was almost certainly blown out of its home galaxy by the winds and explosions of massive stars, although exactly where it ends up is still an open question.
Illustration: A small dwarf galaxy losing glowing gas to supernova winds. This is an illustration, not a photograph of a real galaxy.
What Happened
The GLOW team examined 37 low-mass, low-metallicity, gas-rich dwarf galaxies within about 6 megaparsecs of Earth — close enough, in cosmic terms, for Hubble to pick out their individual stars. The galaxies span a stellar-mass range where theorists expect feedback from stars to shape galaxy evolution most strongly, from about three million to about three billion times the mass of the Sun.
By combining star-formation histories derived from those resolved stars with measurements of the gas in each galaxy and its chemical composition, the researchers calculated how much oxygen the stars in each galaxy should have made over cosmic time. They then added up how much oxygen could actually be found today in the galaxy's stars, gas and dust. The gap was striking: in every case, most of the oxygen was gone.
Across the sample, galaxies retained between 4 and 25 per cent of the oxygen produced by stellar nucleosynthesis, with a mean of 11 per cent. Nearly all of the oxygen that did stay behind was found in the gas, rather than locked up in stars or dust. The result was reported by NASA's Space Telescope Science Institute and published in The Astrophysical Journal as "GLOW. II. A Census of Oxygen in Low-mass Galaxies".
"Oxygen is produced by stars, which are inside of galaxies. We know that the oxygen had to be inside of the galaxies at some point in the past, so the oxygen must have been ejected," said co-investigator Grace Telford, an assistant professor in the Department of Physics and Astronomy at the University of Utah. "And to do that, you need energy from massive stars and supernova explosions to drive material out of the host galaxies and into the space between galaxies."
The researchers also found that the winds appear to be more effective at stripping oxygen from low-mass galaxies than current models predict. That matters because simulations of galaxy formation rely on getting this balance right: if models keep too much oxygen — and, by extension, too many heavy elements — inside small galaxies, they will misjudge how those galaxies grow, form stars and enrich their surroundings.
Background
Oxygen is a useful tracer because its story is unusually well understood. Massive stars forge oxygen in their cores and scatter it when they explode as supernovae. Because oxygen is relatively easy to detect and measure, and because astronomers understand how much of it stars should produce, it acts as a reliable indicator of how chemical elements move through a galaxy — and out of it.
"Oxygen is one of the primary elements formed by stars. Being able to trace how much oxygen gets made, and then how much is distributed inside and outside of galaxies, helps us understand galaxy evolution," said Kristen Wingfield McQuinn, the GLOW principal investigator. McQuinn is also mission head of the Science Operations Center for NASA's newly launched Nancy Grace Roman Space Telescope at the Space Telescope Science Institute in Baltimore, and an associate professor at Rutgers University. "It also helps us learn where and how much galaxies are enriching the space between galaxies, called the intergalactic medium."
Illustration: The Hubble Space Telescope observing faint dwarf galaxies in deep space. This is an illustration, not a photograph of a real observation.
The new census builds on earlier work. A 2014 study of Milky Way-like, star-forming galaxies found that those larger galaxies also lost a large fraction of their oxygen, some of it turning up in the diffuse haloes of gas surrounding galaxies outside their disks, known as the circumgalactic medium. The GLOW study extends that accounting to much smaller galaxies, across a wide range of masses and environments — a regime where the losses were expected to be larger, but had not been measured so systematically.
To do the accounting, the team used archival optical imaging from Hubble to resolve stellar populations and reconstruct each galaxy's star-formation history and composition. They combined that with data from NASA's retired Spitzer Space Telescope and from ground-based radio observatories, the Very Large Array and the Australian Telescope Compact Array, to trace the oxygen atoms produced by stars through the galaxies' gas, dust and stars.
The 37 galaxies were chosen deliberately. Their masses fall in the range where simulations and theory predict that an increasing share of oxygen will be expelled by supernova explosions, and all of them are close enough for Hubble to resolve individual stars. That last condition is rare: the vast majority of known dwarf galaxies are too far away for this kind of star-by-star study.
Why It Matters
Dwarf galaxies may be small — the sample ranges from galaxies with only a few thousand stars to systems with hundreds of millions — but they are among the most common galaxies in the universe, and they are sensitive laboratories for galaxy physics. Because their gravity is weak, the push from supernovae can more easily overwhelm it, so the tug-of-war between star formation and ejection plays out more dramatically than in a giant spiral like the Milky Way.
The GLOW result sharpens that picture in three ways. First, it puts numbers on the leak: a typical dwarf galaxy in the sample keeps only about one-tenth of its oxygen. Second, it shows that the retained oxygen sits overwhelmingly in the gas, which is the reservoir from which future stars will form — a detail that models of chemical enrichment need to reproduce. Third, it reveals an environmental twist: galaxies in denser surroundings, and those within about a million parsecs of a massive neighbour galaxy, showed greater scatter toward higher oxygen retention, suggesting that environment affects how much oxygen a galaxy keeps, recycles or gains from its neighbours.
Just as surprising was what did not matter. Contrary to expectations, the researchers found that oxygen retention did not correlate with a galaxy's position on the mass-metallicity relation — the well-known link between a galaxy's mass and the chemical richness of its gas. That non-result, the team says, is itself a clue: it suggests that the processes setting how much oxygen stays behind are not simply a smooth function of mass and metallicity, and that wind strength and outflow composition may be doing more of the work than some models assume.
The findings also feed into a bigger question: how did the space between galaxies get enriched at all? If small galaxies are constantly blowing their heavy elements outward, they may be important contributors to the chemical enrichment of the intergalactic medium over cosmic time — the same medium from which later generations of galaxies, stars and, eventually, planets form.
Illustration: Oxygen-rich gas drifting from a small galaxy into the space between galaxies. This is an illustration, not a photograph of a real event.
What Happens Next
The biggest loose end is the fate of the missing oxygen. Ionised oxygen has been detected in the circumgalactic medium around dwarf galaxies, but the GLOW team says it is still unclear whether most of the missing oxygen sits in that halo gas or has been expelled entirely from the galaxies' surroundings. Distinguishing between those two outcomes matters: oxygen parked in a halo could eventually fall back and fuel later star formation, while oxygen lost to intergalactic space is, for that galaxy at least, gone for good.
Future observations should help close that gap. McQuinn has an approved programme to use the Roman Space Telescope to find even lower-mass, fainter and tinier galaxies across a wide swath of sky. Once those systems are identified, she hopes to follow them up with Hubble and NASA's James Webb Space Telescope to obtain deeper observations of their stars, allowing the team to reconstruct how they formed stars across cosmic time.
Telford, meanwhile, plans to use Hubble to zoom in on the individual massive stars that produce oxygen and eject it from their host galaxies, using some of the nearby GLOW targets to characterise in detail how those stars differ in small galaxies. That star-level view could explain why the winds are so effective — and whether the most massive stars in dwarfs behave differently from their counterparts in larger galaxies.
Hubble itself remains central to this work more than three decades after launch. The telescope is a project of international cooperation between NASA and the European Space Agency; NASA's Goddard Space Flight Center manages mission operations, and the Space Telescope Science Institute conducts Hubble science operations. For dwarf galaxies too faint and distant for most other instruments to resolve star by star, that longevity is what makes a census like GLOW possible at all.
For now, the message is a humbling one for small galaxies: they manufacture the ingredients of future worlds, but they keep only a small fraction of them. The rest is on the wind — enriching the darkness between galaxies, and perhaps, in time, seeding the next generation of stars.
Sponsored: explore today's featured sponsored offers
FAQ
What is a dwarf galaxy?
A dwarf galaxy is a small galaxy, typically containing far fewer stars than a large spiral such as the Milky Way. The galaxies in the GLOW sample range from systems with only a few thousand stars to ones with hundreds of millions of stars.
How much oxygen do dwarf galaxies keep?
According to the GLOW census, the 37 dwarf galaxies studied retained between 4 and 25 per cent of the oxygen produced by their stars, with a mean of 11 per cent. Nearly all of the retained oxygen was found in the gas, not in stars or dust.
Where does the missing oxygen go?
The researchers say it was most likely ejected by supernova-driven winds. Some ionised oxygen has been detected in the circumgalactic medium around dwarf galaxies, but it is still unclear whether most of the missing oxygen sits there or has been expelled entirely into intergalactic space.
Why study oxygen specifically?
Oxygen is the most abundant element after hydrogen and helium, it is relatively easy to detect and measure, and astronomers understand well how much of it stars produce. That makes it a reliable tracer for how heavy elements move through and out of galaxies.
What comes next for the GLOW project?
The team plans to use the Nancy Grace Roman Space Telescope to find even lower-mass, fainter dwarf galaxies, then follow them up with Hubble and the James Webb Space Telescope. Telford also plans Hubble observations of the individual massive stars that drive the outflows.
Sources
- NASA / Space Telescope Science Institute, via Astrobiology, "Dwarf Galaxies Retain Only A Small Fraction Of The Oxygen Made By Their Stars," 2 November 2026.
- McQuinn et al., "GLOW. II. A Census of Oxygen in Low-mass Galaxies," The Astrophysical Journal, 2026 (arXiv:2608.07665) — 37 galaxies, 4–25 per cent retention, mean 11 per cent.
- NASA / Space Telescope Science Institute, Hubble mission background (NASA–ESA cooperation, Goddard mission operations, STScI science operations).
Comments
Post a Comment