The Impact of Star Formation and Feedback Recipes on the Stellar Mass and Interstellar Medium of High-Redshift Galaxies

TL;DR

This paper introduces MEGATRON, a high-resolution, non-equilibrium chemistry and on-the-fly radiative transfer galaxy formation model implemented in RAMSES-RTZ to simulate a $10^9\,M_{\odot}$ halo at $z=6$ with pc-scale resolution. It systematically varies subgrid prescriptions for star formation and feedback to assess their impact on stellar mass growth and ISM properties, finding that the total feedback energy budget predominantly governs high-redshift star formation regulation, while many other numerical choices have only modest effects on the final stellar mass. Importantly, even models that yield similar final masses can produce markedly different ISM conditions and emission-line ratios, illustrating that nebular diagnostics like O32, C43, and the [O III] 4363/5007 ratio provide powerful, self-consistent constraints on high-redshift physics with JWST data. The work demonstrates that combining non-equilibrium thermochemistry with RT enables direct forward modeling of nebular emission, offering a path to tightly constrain star formation and feedback processes in the early universe using ISM observations.

Abstract

We introduce MEGATRON, a new galaxy formation model for cosmological radiation hydrodynamics simulations of high-redshift galaxies. The model accounts for the non-equilibrium chemistry and heating/cooling processes of $\geq 80$ atoms, ions, and molecules, coupled to on-the-fly radiation transfer. We apply the model in a cosmological setting to the formation of a $10^9\ {\rm M_{\odot}}$ halo at $z=6$, and run 25 realizations at pc-scale resolution, varying numerous parameters associated with our state-of-the-art star formation, stellar feedback, and chemical enrichment models. We show that the overall budget of feedback energy is the key parameter that controls star formation regulation at high redshift, with other numerical parameters (e.g. supernova clustering, star formation conditions) having a more limited impact. As a similar feedback model has been shown to produce realistic $z=0$ galaxies, our work demonstrates that calibration at $z=0$ does not guarantee strong regulation of star formation at high-redshift. Interestingly, we find that subgrid model variations that have little impact on the final $z=6$ stellar mass can lead to substantial changes on the observable properties of high-redshift galaxies. For example, different star formation models based on, e.g. density thresholds or turbulence inspired criteria, lead to fundamentally distinct nebular emission line ratios across the interstellar medium (ISM). These results highlight the ISM as an important resource for constraining models of star formation, feedback, and galaxy formation in the JWST era, where emission line measurements for $>1,000$ high-redshift galaxies are now available.

Figures (16)

• Figure 1: (Top) Equilibrium temperature-density (top) and pressure-density (bottom) relations for the thermodynamic model used in MEGATRON, as a function of metallicity (solid lines). For all models, we assume a local ISM radiation field such that $G_0=1$ and a background cosmic ray hydrogen ionization rate of $10^{-16}\ {\rm s^{-1}}$. At solar metallicities (green), the thermodynamic model is consistent with other ISM models (Bialy2019, Koyama2000, Kim2023; red dashed, dotted, and solid, respectively) and data in the Milky Way and the LMC/SMC (Gerin2015, Jenkins2011, Herrera2017, Welty2016; orange points, purple line, red dashed, solid, dotted lines for cold-neutral fractions of 0.7, 0.5, and 0.3, and grey triangles, respectively). Changes in atomic data and reaction rates lead to small variations with the earlier version of the model (PRISM, dashed blue lines).
• Figure 2: Emission line luminosities from a Strömgren sphere around an O4V star in gas with varying density at 0.1 metallicity. Results from the MEGATRON non-equilibrium, radiative transfer calculation (points) show good agreement with CLOUDY calculations (lines).
• Figure 3: Fraction of the gas cell mass that needs to be converted into zero-age main-sequence star particles at different metallicities (different colored lines) such that the Strömgren sphere is resolved. For this figure, we assume a spatial resolution of 5.34 pc, which is the maximum level of refinement at $z=13.3$ for the fiducial simulations presented below. In the code this will change depending on redshift to correspond to the physical resolution of the cells.
• Figure 4: Expected terminal momentum from SN explosions (dashed black line) as a function of gas density compared to that measured in idealized MEGATRON simulations (red circles). The discontinuity at $n_{\rm H}=10^{2.3}\,{\rm cm^{-3}}$ represents the transition point between a resolved and unresolved cooling radius for the SN remnant. The expected terminal momentum follows the results of Kim2015.
• Figure 5: Cumulative stellar mass formed (top) or 1Myr-averaged star formation rate (bottom) as a function of time for all stars in the simulation. Pop. III star formation (cyan) begins first and continues until $z\approx8$, but is rapidly overtaken by Pop. II star formation (purple). Rapid increases in stellar mass (vertical red dashed) track vigorous assembly through major mergers.