A novel approach to correcting $T_e$-based mass-metallicity relations

TL;DR

This paper identifies biases in the $T_e$-method for measuring metallicity in galaxies caused by temperature fluctuations and proposes a calibration using simulations to correct these biases, improving metallicity estimates.

Contribution

It introduces a calibration between ratio temperature and line temperature based on high-resolution simulations to correct $T_e$-based metallicity biases in low-mass galaxies.

Findings

Corrects the metallicity bias, increasing the mass-metallicity relation normalization by 0.18 dex.

Flattens the slope of the mass-metallicity relation from 0.87 to 0.58.

Demonstrates the importance of accounting for temperature fluctuations in metallicity measurements.

Abstract

Deriving oxygen abundances from the electron temperature (hereafter the $T_e$-method) is the gold-standard for extragalactic metallicity studies. However, unresolved temperature fluctuations within individual HII regions and across different HII regions throughout a galaxy can bias metallicity estimates low, with a magnitude that depends on the underlying and typically unknown temperature distribution. Using a toy model, we confirm that computing $T_e$-based metallicities using the temperature derived from the [O III] $\lambda$4363/$\lambda$5007 or [O II] $\lambda\lambda$7320,7330 / [O II] $\lambda\lambda$3727 ratio ('ratio temperature'; $T_{\rm ratio}$) results in an underprediction of metallicity when temperature fluctuations are present. In contrast, using the unobservable 'line temperatures' ($T_{\rm line}$) that provide the mean electron and ion density-weighted emissivity yield an accurate metallicity estimate. To correct this bias in low-mass galaxies, we demonstrate an example calibration of a relation between T_ratio and T_line based on a high-resolution (4.5 pc) RAMSES-RTZ simulation of a dwarf galaxy that self-consistently models the formation of multiple HII regions and ion temperature distribution in a galactic context. Applying this correction to the low-mass end of the mass-metallicity relation shifts its normalization up by 0.18 dex on average and flattens its slope from 0.87 to 0.58, highlighting the need for future studies to account for, and correct, this bias.