The physical origin of the ~0.3 dex metallicity deficit of low-mass galaxies at high redshift: why galaxies of the same stellar mass are more metal-poor at $z\gtrsim2$ than at $z\sim0$

Galaxies with stellar masses $\log(M_{\star}/M_{\odot})\lesssim9$ observed as [OIII]λ4363 emitters at $z\gtrsim2$ are, on average, about 0.3 dex more metal-poor (in gas-phase oxygen abundance $12+\log(\mathrm{O/H})$) than galaxies of the same stellar mass at $z\sim0$. Laseter et al. (2026) showed that below this mass the canonical Fundamental Metallicity Relation (FMR) does not exist at $z\sim0$, so the offset cannot be attributed to SFR differences via the FMR, and asked why the high- and low-redshift Universe differ in the same mass regime when star formation and enriched outflows dominate the metallicity variations. The open problem is to identify which physical mechanism — evolving ambient circumgalactic/intergalactic gas metallicity, a different initial mass function or massive-star yields, higher star-formation efficiency, or stronger/more metal-efficient outflows at early times — dominantly produces this offset. A self-consistent, homogeneous reanalysis of the low- and high-redshift samples, acknowledged in the source and still unpublished, would firm up the observational basis of the comparison; the scientific target is the physical origin of the offset.

Background

The gas-phase oxygen abundance ($12+\log(\mathrm{O/H})$) of a star-forming galaxy correlates with its stellar mass — the mass–metallicity relation (MZR) — and, at $\log(M_{\star}/M_{\odot})\gtrsim9$, residual scatter about the MZR anti-correlates with star formation rate (SFR), defining the Fundamental Metallicity Relation (FMR), a three-parameter relation between $M_{\star}$, SFR, and metallicity (Mannucci et al. 2010, MNRAS 408, 2115). The FMR was originally described as approximately redshift-invariant out to $z\sim2.5$, which made it a standard benchmark for chemical-evolution models. Metallicities for this kind of work are measured either by the 'direct' method — the faint auroral oxygen emission line [OIII]λ4363, detected only in low-metallicity or high-excitation gas, yields the electron temperature $T_e$, from which the oxygen abundance follows — or by 'strong-line' calibrations, which infer the abundance from ratios of bright, easily detected forbidden emission lines; requiring an [OIII]λ4363 detection selects a biased ('auroral-emitter') population. A useful summary statistic is the effective yield $y_{\mathrm{eff}}$, the oxygen mass per locked stellar mass inferred from the observed abundance; in the analytical infall/outflow models of Dalcanton (2007, ApJ 658, 941), departures of $y_{\mathrm{eff}}$ from the true stellar yield $y_{\mathrm{true}}$ diagnose pristine gas infall (which dilutes) versus enriched outflows (which remove metals). JWST (James Webb Space Telescope) spectroscopy has now measured direct-method metallicities for hundreds of [OIII]λ4363 emitters at $z\gtrsim2$, and many studies report that these systems depart from the $z\sim0$ FMR. Laseter et al. (2026, ApJ 1007, 26; arXiv:2510.15024) closed the missing comparison anchor: they compiled $\sim700$ low-mass ($\log(M_{\star}/M_{\odot})\lesssim9$) [OIII]λ4363 emitters at $z\sim0$ and found no evidence of the FMR at these masses by any standard fitting method, a weak positive (rather than the canonical negative) metallicity–SFR correlation, gas-rich systems with effective yields spanning $\sim3$ dex, and a pattern of $y_{\mathrm{eff}}$ versus FMR offset that, in the Dalcanton framework, indicates bursts that rapidly enrich and eject metals before the gas reservoir re-balances. With this local characterization, they then showed that high-redshift [OIII]λ4363 emitters really are more metal-poor than $z\sim0$ counterparts at fixed $M_{\star}$, by $\Delta12+\log(\mathrm{O/H})\simeq0.3$ dex: their comparison covers $z\gtrsim2$ systems ($\sim320$ objects) and the result is quoted for $z\gtrsim3$, as a difference of sample means ($\langle12+\log(\mathrm{O/H})\rangle=7.96$ at low $z$ versus $7.66$ at high $z$) that the authors note is subject to sample bias. Their Discussion leaves the physical cause open: candidate explanations are (a) the ambient circumgalactic-medium/intergalactic-medium (CGM/IGM) gas being less metal-enriched at high redshift (analytical models predict a $\sim0.8$ dex decrease in ambient gas metallicity between $z=0$ and $z=6$, comparable in order of magnitude to the observed offset; Hernquist & Springel 2003, MNRAS 341, 1253), (b) a systematically different stellar initial mass function (IMF, the distribution of stellar masses at birth) or massive-star nucleosynthetic yields at early times, (c) higher star-formation efficiencies, and (d) stronger or more metal-efficient feedback and outflows at early times. Two auxiliary observational gaps feed the same question: the gas fractions of the high-redshift inter-burst ('mini-quenched') population are unknown, and spatially resolved gas kinematics and multiphase outflow metal-loading (entrainment) factors are too poorly sampled at low mass to close the causal chain from starburst to metal ejection to metallicity dilution.

Problem Statement

Determine the physical origin of the redshift evolution of gas-phase metallicity in low-mass galaxies. Concretely, building on Laseter et al. (2026), identify which physical mechanism dominantly produces the $\Delta12+\log(\mathrm{O/H})\simeq0.3$ dex deficit of $\log(M_{\star}/M_{\odot})\lesssim9$ galaxies at $z\gtrsim2$ (the offset to be explained is the low-mass comparison over $z\gtrsim2$; the quoted $0.3$ dex value is the source's $z\gtrsim3$ measurement) relative to $z\sim0$ counterparts at fixed $M_{\star}$ — the lower metallicity of ambient circumgalactic/intergalactic accreted gas at early times, a different stellar IMF or massive-star yields, higher star-formation efficiency, or stronger/more metal-efficient (enriched) outflows. Because each candidate can on its own produce a metallicity shift of the required order at fixed mass, an attribution requires evidence that discriminates the dominant mechanism from the other candidates rather than a reproduction of the offset's magnitude alone. Whether the present data can discriminate among the candidates at all is itself part of the question, and non-identifiability given the present data is a possible outcome of the analysis. The observational basis of the offset carries acknowledged caveats — the two-epoch comparison combines heterogeneous stellar-mass, SFR, and metallicity determinations for [OIII]λ4363-selected samples, the source quotes the 0.3 dex figure for its $z\gtrsim3$ systems and notes that sample bias may be present, and the source's planned self-consistent reanalysis remains unpublished — and how much of the offset survives a homogeneous treatment of these systematics is itself an open observational question. This serves one objective: explaining why galaxies of the same low stellar mass were more metal-poor in the early Universe than they are today.

References

1. Laseter, I. H., Maseda, M. V., Bunker, A. J., Cameron, A. J., Curti, M., & Simmonds, C. (2026). An Investigation into the Low-Mass Fundamental Metallicity Relation in the Local and High-z Universe. The Astrophysical Journal, 1007, 26. DOI 10.3847/1538-4357/ae7d25. https://doi.org/10.3847/1538-4357/ae7d25 (preprint: arXiv:2510.15024, https://arxiv.org/abs/2510.15024; source of the open question)
2. Mannucci, F., Cresci, G., Maiolino, R., Marconi, A., & Gnerucci, A. (2010). A fundamental relation between mass, star formation rate and metallicity in local and high-redshift galaxies. Monthly Notices of the Royal Astronomical Society, 408, 2115-2127. DOI 10.1111/j.1365-2966.2010.17291.x. https://doi.org/10.1111/j.1365-2966.2010.17291.x
3. Dalcanton, J. J. (2007). The Metallicity of Galaxy Disks: Infall versus Outflow. The Astrophysical Journal, 658, 941-959. DOI 10.1086/508913. https://doi.org/10.1086/508913
4. Hernquist, L., & Springel, V. (2003). An analytical model for the history of cosmic star formation. Monthly Notices of the Royal Astronomical Society, 341, 1253-1267. DOI 10.1046/j.1365-8711.2003.06499.x. https://doi.org/10.1046/j.1365-8711.2003.06499.x
5. Curti, M., Maiolino, R., Curtis-Lake, E., Chevallard, J., Carniani, S., D'Eugenio, F., Looser, T. J., Scholtz, J., Charlot, S., Cameron, A., et al. (2024). JADES: Insights into the low-mass end of the mass–metallicity–SFR relation at 3<z<10 from deep JWST/NIRSpec spectroscopy. Astronomy & Astrophysics, 684, A75. DOI 10.1051/0004-6361/202346698. https://doi.org/10.1051/0004-6361/202346698 (preprint: arXiv:2304.08516, https://arxiv.org/abs/2304.08516)
6. Jain, S., Sanders, R. L., Khostovan, A. A., Jones, T., Shapley, A. E., Reddy, N. A., Garcia, A. M., Torrey, P., & Coil, A. (2025). A Uniform Analysis of Gas-phase Metallicity Evolution with 1-3 Gyr Time Sampling over the Past 12 Billion Years. arXiv:2508.18369. https://arxiv.org/abs/2508.18369
7. Giménez-Alcázar, A., Amorín, R., & Vilchez, J. M. (2026). Shape of Direct-Method Mass-Metallicity Relation with JWST: Fast-Track Nitrogen and Helium Enrichment. arXiv:2605.05327. https://arxiv.org/abs/2605.05327
8. Isobe, Y., Curti, M., Maiolino, R., Duan, Q., McClymont, W., Puskás, D., D'Eugenio, F., Rinaldi, P., Trussler, J. A. A., Scholtz, J., et al. (2026). JADES: the mass-metallicity relation at z=1-10. New calibrations, extremely metal-poor galaxies, and chemical diversity. arXiv:2606.11345. https://arxiv.org/abs/2606.11345
9. Lam, N., Clarke, L., Shapley, A. E., Sanders, R. L., Topping, M. W., Brammer, G. B., Reddy, N. A., & Karthikeyan, S. (2026). The JADES Mass-Metallicity and Fundamental Metallicity Relations at z≳2 Using New High-Redshift Metallicity Calibrations. arXiv:2605.30513. https://arxiv.org/abs/2605.30513
10. Khostovan, A. A., Sanders, R. L., Shapley, A. E., Topping, M. W., Reddy, N. A., Garcia, A. M., Berg, D. A., Clarke, L., Cullen, F., Ellis, R. S., et al. (2025). The AURORA Survey: The Mass-Metallicity and Fundamental Metallicity Relations at z∼2.3 Based Purely on Direct Te Metallicities. arXiv:2512.16989. https://arxiv.org/abs/2512.16989
11. He, X., Li, Z., Wang, X., Cai, Z., Jones, T., Treu, T., Vulcani, B., Malkan, M. A., & Glazebrook, K. (2026). Revisiting the mass metallicity relation and the fundamental metallicity relation of dwarf galaxies at cosmic noon with NIRISS. arXiv:2605.20810. https://arxiv.org/abs/2605.20810
12. Wang, K., Baugh, C. M., Boardman, N. F., Bose, S., Cai, Z., Cole, S., Frenk, C. S., Lacey, C., Norberg, P., Peng, Y., et al. (2026). Inflow-driven galaxy evolution - I. Revealing the physics of the fundamental metallicity relation. arXiv:2608.04784. https://arxiv.org/abs/2608.04784
13. Carnevale, L., Garcia, A. M., Torrey, P., Ellison, S. L., Jain, S., Zhu, P., Grasha, K., Sanders, R. L., Hernquist, L., Kewley, L. J., et al. (2026). The Mass Dependence of the Fundamental Metallicity Relation in Observations and Simulations. arXiv:2608.24826. https://arxiv.org/abs/2608.24826
14. Pollock, C. L., Gottumukkala, R., Heintz, K. E., Brammer, G. B., Roberts-Borsani, G., Oesch, P. A., Witstok, J., Arellano-Córdova, K. Z., Cullen, F., Scholte, D., et al. (2026). Novel z∼10 auroral line measurements extend the gradual offset of the FMR deep into the first Gyr of cosmic time. Astronomy & Astrophysics, 708, A203. DOI 10.1051/0004-6361/202556032. https://doi.org/10.1051/0004-6361/202556032 (preprint: arXiv:2506.15779, https://arxiv.org/abs/2506.15779)
15. Garcia, A. M., Torrey, P., Ellison, S. L., Grasha, K., Chen, Q.-H., Hemler, Z. S., Zimmerman, D. H. T., Wright, R. J., Zovaro, H. R. M., Nelson, E. J., et al. (2024). Does the Fundamental Metallicity Relation Evolve with Redshift? II: The Evolution in Normalisation of the Mass-Metallicity Relation. Monthly Notices of the Royal Astronomical Society, 536, 119-144. DOI 10.1093/mnras/stae2587. https://doi.org/10.1093/mnras/stae2587 (preprint: arXiv:2407.06254, https://arxiv.org/abs/2407.06254)
16. Boardman, N. F., Wild, V., Scholte, D., Wang, K., Vale Asari, N., & Saintonge, A. (2026). Beyond the Fundamental Metallicity Relation: galaxy sizes encode the link between inflow and metallicity. arXiv:2606.13124. https://arxiv.org/abs/2606.13124
17. Scholte, D., Cullen, F., Moustakas, J., Zou, H., Saintonge, A., Arellano-Cordova, K. Z., Stanton, T. M., Andrews, B., Sui, J., Aguilar, J., et al. (2026). Electron temperature relations and the direct N, O, Ne, S, and Ar abundances of 49,959 star-forming galaxies in DESI data release 2. Monthly Notices of the Royal Astronomical Society, DOI 10.1093/mnras/stag1381. https://doi.org/10.1093/mnras/stag1381
18. Trussler, J. A. A., Eisenstein, D. J., Bunker, A. J., Cameron, A. J., Carniani, S., Charlot, S., Chevallard, J., Conselice, C. J., Curti, M., Curtis-Lake, E., et al. (2026). Out of oxygen: extremely metal-poor galaxy candidates identified at 2.5<z<6.5 with deep JADES medium-band imaging. Monthly Notices of the Royal Astronomical Society, DOI 10.1093/mnras/stag1360. https://doi.org/10.1093/mnras/stag1360
19. Koller, M., Maiolino, R., Übler, H., Duan, Q., Scholtz, J., Arribas, S., Baker, W. M., Carniani, S., Charlot, S., Curti, M., et al. (2026). Metal Mayhem at z∼7-10: Diversity and Evolution of Gas-Phase Metallicity Gradients. arXiv:2604.07076. https://arxiv.org/abs/2604.07076
20. Choustikov, N., Katz, H., Cameron, A. J., Saxena, A., Devriendt, J., Slyz, A., Rey, M. P., Cadiou, C., Blaizot, J., Kimm, T., et al. (2025). MEGATRON: Disentangling Physical Processes and Observational Bias in the Multi-Phase ISM of High-Redshift Galaxies. arXiv:2510.06347. https://arxiv.org/abs/2510.06347