Abstract illustration of glowing gas and infrared spectral lines.

JWST Calibrates a Neon Gauge for Oxygen in Dusty Nebulae

By comparing JWST mid-infrared spectra with direct oxygen measurements in a sample of 18 nebulae, astronomers have produced the first empirical calibration of Ne23, a diagnostic that estimates oxygen abundance without the electron-temperature uncertainties of optical methods.

Apr 29, 2026

Inside the glowing clouds of ionized gas around young massive stars, oxygen and neon are made in the same furnaces. Both elements are forged in massive stars and scattered into space when those stars explode. That shared origin makes neon a possible stand-in for oxygen, the element astronomers use to gauge a galaxy's chemical enrichment. But a mid-infrared method for using neon lines to estimate oxygen had never been calibrated directly against real nebulae. A new JWST study has now done that.

The trouble with taking a nebula's temperature

Astronomers often measure metallicity—the abundance of elements heavier than helium—by looking at oxygen in H II regions, the glowing clouds of ionized gas around young massive stars. The most trusted approach uses optical emission lines from oxygen ions. It is called the direct method: faint auroral lines reveal the electron temperature of the gas, and that temperature helps convert line strengths into an oxygen abundance.

That method has limits. The optical lines are sensitive to electron temperature, so small temperature variations can bias the result. In metal-rich nebulae, the gas cools efficiently and the faint auroral lines become hard to detect. Dust can hide the optical light altogether. Other common optical strong-line diagnostics also suffer from a double-value problem, where the same line ratio can point to two different oxygen abundances.

Neon follows oxygen through the same stellar furnaces

The new diagnostic, called Ne23, uses two mid-infrared fine-structure lines of neon: [Ne II] at 12.81 micrometers and [Ne III] at 15.56 micrometers. It compares their combined strength with a hydrogen line, Hu alpha at 12.37 micrometers. The result is a logarithmic ratio that the researchers calibrated against oxygen abundance.

Neon is useful because it evolves alongside oxygen in massive stars, the paper explains. It also does not freeze onto dust grains the way some oxygen can. And the mid-infrared neon lines arise from energy levels so low that their strengths are largely insensitive to electron temperature. Their wavelengths are also close together, so dust attenuation affects them only slightly. That combination makes Ne23 attractive for dusty, metal-rich environments where optical diagnostics struggle.

Matching JWST apertures to ground-based spectra

The team used JWST's Mid-Infrared Instrument in its Medium Resolution Spectroscopy mode, MIRI MRS, which is an integral field spectrograph: it can take spectra across a small patch of sky rather than through a single narrow slit. That flexibility allowed the astronomers to extract one-dimensional mid-infrared spectra from apertures matched to existing optical spectra.

Ten of those H II regions come from the CHAOS project, observed with the Large Binocular Telescope's MODS spectrograph. Those optical spectra provided direct measurements of electron temperature and oxygen abundance. The researchers then added archival MIRI observations of eight low-metallicity galaxies that also had direct oxygen abundances from LBT/MODS spectra. In total, the study assembles 18 nebulae spanning 1.5 dex in 12+log(O/H), a standard logarithmic measure of oxygen abundance; the main cubic fit uses 17 of them because one object has only an upper limit on Ne23. One dex is a factor of ten; 0.06 dex is about 15 percent.

A tight relation, and a model mismatch

The result is a strong correlation. The Spearman rank-order correlation coefficient is 0.988, with a null-hypothesis probability of about 2×10^-4, which the paper reports as extremely unlikely to arise by chance. The best-fit relation is a cubic function because the lowest-metallicity nebulae bend away from a straight line. At fixed Ne23, the scatter in oxygen abundance is just 0.06 dex—at least a factor of two smaller than the scatter typical of existing empirical optical strong-line diagnostics.

The empirical calibration also differs from theory. The Hii-Chi-mistry-IR photoionization model grid predicts oxygen abundances that are, on average, about 0.23 dex higher than the new observations at the same Ne23. That offset echoes disagreements seen between models and observations for common optical strong-line diagnostics. In other words, the neon lines are not just a theoretical convenience; they needed a real-world calibration.

Where the neon gauge works, and where it does not

According to the paper, the relation can estimate oxygen abundance from JWST/MIRI MRS observations of ionized nebulae out to redshift z≈0.8. It should be especially useful in highly attenuated regions and in the metal-rich interstellar medium, where optical lines are faint or hidden. Future JWST surveys such as CLASSYIR and MIRAGE are expected to test the relation with larger samples. The authors also note that a future far-infrared mission concept, PRIMA, and its FIRESS spectrograph could observe all the needed Ne23 lines in bright galaxies beyond z≈0.95, though that is a forecast rather than a current measurement.

The caveats are important. The main cubic calibration rests on 17 nebulae (an 18th object provides only an upper limit on Ne23), a small sample by survey standards. The authors warn against using it outside the measured range of Ne23 from 0.904 to 1.99. It should not be applied to active galactic nuclei, whose harder radiation can push neon into higher ionization states that Ne23 does not track. Nor should it be applied to individual spaxels in a spectral map; the paper recommends using integrated emission from a star-forming nebula. The direct oxygen abundances used for calibration have their own uncertainties, and variations in the Ne/O ratio or differences between the gas traced by optical and infrared light could add scatter. The team found no clear correlation between the residuals and ionization, density, dust, or Ne/O, but larger samples will be needed to confirm that.

A new window into dusty chemistry

The new calibration does not replace direct optical measurements. It gives astronomers another way to read the chemical history of galaxies, one that works in places where optical light is dimmed or lost. Because neon and oxygen are released by the same generations of massive stars, the mid-infrared glow of neon can now serve as a proxy for oxygen across a wide range of metallicities. In the dusty hearts of galaxies and in nebulae too cool for traditional diagnostics, JWST can begin to read a chemical record that was previously hidden.

The First Empirical Calibration of the MIR Abundance Diagnostic Ne$_{23}$ with JWSTNoah S. J. Rogers, Evan D. Skillman, Danielle A. Berg, Karla Z. Arellano-Córdova, Richard W. Pogge, Alessandra Aloisi, Leslie K. Hunt, Anne E. Jaskot, Matilde Mingozzi, Ryan J. Rickards Vaught, Karin M. Sandstrom, O. Grace Telford, Macarena G. del Valle-Espinosahttps://arxiv.org/abs/2604.27056v1