A new arXiv v1 preprint reports that star-formation rate (SFR) did not add a statistically significant signal to either the spatially resolved or integrated mass-metallicity relation after the relevant stellar-mass measure was accounted for. For the resolved test, that conclusion came under the JY22 calibration; for the integrated sample, residual metallicity did not correlate with total SFR or specific SFR after total mass was controlled.
Here, metallicity is the study’s oxygen-abundance estimate for gas, and the mass-metallicity relation is the pattern between that estimate and stellar mass. The paper tests whether star formation adds a second layer to that pattern and whether the answer changes with calibration choice.
Mass carried the stronger resolved signal
For the spatially resolved test, the authors kept 3.42 million star-forming MaNGA spaxel measurements from 5,058 galaxies. They excluded any spaxel with a relevant emission line below signal-to-noise 3 or with Hα equivalent width below 6 Å.
Within that selected sample, the logarithmic stellar-mass surface-density scale ran from 5.8 to 9.8, measured in solar masses per square kiloparsec, while metallicity ranged from 7.75 to 8.95 on the 12+log(O/H) scale.
To look for an SFR contribution after the mass relation had been accounted for, the authors used partial correlation, a measure of association that holds another variable constant. At fixed stellar-mass surface density, the absolute partial correlations for local specific star-formation rate (sSFR) and SFR surface density were both below 0.2, while the coefficient for total stellar mass was 0.69.
That led the authors to judge the resolved fundamental metallicity relation (FMR)—the proposed secondary dependence on star formation—not statistically significant under JY22.
The integrated test pointed the same way
For its integrated analysis, the study selected 4,548 galaxies with more than 50 qualifying spaxels. It estimated each galaxy’s metallicity at a reference radius called Reff by averaging measurements from 0.75 to 1.25 times that radius.
After total stellar mass was controlled for, residuals in the integrated mass-metallicity relation did not correlate with total SFR or sSFR, and the paper found no statistically significant secondary dependence on SFR.
Among the mass subgroups, galaxies above 10^10.2 solar masses were described as having SFR independent of metallicity. Lower-mass curves were mostly flat but showed wiggles, and their lowest-SFR ends contained few data points.
The measuring method mattered
Metallicity was inferred from N2, S2 and R3 ratios of selected emission lines using a Bayesian photoionization-grid analysis, with flat priors for metallicity and ionization parameter.
Among the compared calibrations, JY22 showed the best consistency with direct-method metallicities. But the authors acknowledge that the inferred values depend on assumptions about the ionizing spectrum, dust depletion and the relation between secondary elements and metallicity; different models can fit the same line ratios.
A shortcut with a boundary
The paper also proposes a simpler line-ratio proxy for JY22. It combines the [N II] line at 6584 with the [S II] lines at 6716 and 6731, then adds 0.410 times the log ratio of [N II] 6584 to Hα at 6563.
Across all spaxels, the proxy had a reported root-mean-square error (RMSE) of 0.034 dex and was reported to fit intermediate and high metallicities above 8.4 on the 12+log(O/H) scale, making it a proposed simplified substitute for JY22. The paper also notes reduced sensitivity and somewhat larger scatter at the lowest metallicities.
A conclusion with a defined reach
The manuscript is an arXiv v1 preprint, and its conclusion has a defined reach. It describes associations in the selected star-forming MaNGA sample under the tested calibration; it does not show that changing SFR changes metallicity everywhere, and the authors acknowledge that alternative photoionization assumptions can produce different metallicity estimates from the same line ratios.
Paper data and sources
Original title: Re-evaluating the resolved mass-metallicity relation with a self-consistent metallicity calibration
Authors: Ziming Peng, Renbin Yan, Zesen Lin, Xihan Ji
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-20
DOI: Not available
Original paper · Full text