MEGATRON: Reproducing the Diversity of High-Redshift Galaxy Spectra with Cosmological Radiation Hydrodynamics Simulations

TL;DR

MEGATRON tackles how to reproduce the spectral diversity of high-redshift galaxies by simulating a Milky Way–mass region with on-the-fly non-equilibrium thermochemistry and multi-frequency radiation transfer. The suite resolves parsec-scale ISM physics starting from zero metallicity and follows galaxy evolution from Pop III to Cosmic Noon, generating a library of over $1.75\times10^{5}$ intrinsic spectra. The results show that much of JWST-observed spectral variety—Pop III signatures, EELGs, star-forming disks, mini-quenched systems, and Balmer-break galaxies—emerges naturally within a $\Lambda$CDM framework, with clear dependencies on feedback, IMF, and non-equilibrium chemistry. This forward-modeling approach enables direct comparisons to JWST data and lays the groundwork for constraining ISM/CGM physics in the early universe and for near-field cosmology at $z\approx 0$.

Abstract

We present the MEGATRON suite of cosmological radiation hydrodynamics simulations following the formation of Milky Way-mass galaxies from the earliest cosmic epochs when Population III stars form to Cosmic Noon. The suite represents the first set of cosmological simulations that couples a vast non-equilibrium thermochemistry network of primordial species, metals, and molecules to multifrequency, on-the-fly radiation transport, allowing us to directly predict the spectral properties of early galaxies. By initializing the simulations at zero metallicity, resolving haloes well below the atomic cooling threshold, reaching parsec-scale resolution, and modeling a Milky Way-mass environment, we aim to address four key science themes: 1) Star formation at cosmic dawn, 2) Galaxy formation and the interstellar medium in the epoch of reionization, 3) The circumgalactic medium towards cosmic noon, and 4) Reionization in a local volume environment and near-field cosmology. In this introductory work, we present an overview of the physical characteristics of high-redshift MEGATRON galaxies and their environment at $z>8$. We present a library of $>175,000$ simulated galaxy spectra and demonstrate how the diversity of galaxy spectra seen by JWST is naturally reproduced in the context of a $Λ$CDM cosmology. This project represents a step towards making more direct comparisons between simulations and observations and will enable future work to both optimize methods for inferring galaxy properties from observations and to elucidate the physics that governs galaxy formation in the early Universe.

Figures (20)

• Figure 1: Summary of the four key science drivers of the MEGATRON simulations. Top left: A $z\sim15$ density map with H$\beta$ (white) and [O III] $\lambda$5007 (yellow) shows radiation-driven H$\beta$ emission extending into the IGM, while metals remain near galaxies. Top right: A rotationally-supported $z=10$ galaxy reveals early H$_2$ and CO formation and complex emission line morphology. Bottom left: The CGM of a massive $z\sim4$ star-forming galaxy highlights how different ions trace distinct gas phases. Bottom right: The $z=0$ dark matter distribution connects high- and low-redshift structure formation.
• Figure 2: Demonstration of how MEGATRON galaxies can be mock observed in multiple JWST observing modes. The example galaxy is a dusty spiral at $z=4$ from the Cosmic Noon suite. The background RGB image combines JWST NIRCam filters F115W and F150W, in the blue channel, F200W and F277W, in the green channel, and F356W, F410M, and F444W in the red channel. Prominent dust lanes are visible as are large plumes of stars that are a remnant of a merger. To mock the NIRSpec micro shutter array (MSA), we overlay slitlets and show the spectrum of the central region of the galaxy. The object has a very red UV slope from dust attenuation and a clear Balmer break due to an aging stellar population, and some weak emission lines that are remnants of previous and ongoing star formation. Finally, we mock the central region of the NIRSpec IFU focusing on the velocity-resolved H$\alpha$ line. We split the velocity channels in to bins of $v<-400~{\rm km/s}$, $-400<v<-200~{\rm km/s}$, $-200<v<200~{\rm km/s}$, $200<v<400~{\rm km/s}$, and $v>400~{\rm km/s}$. Note how the morphology of the H$\alpha$ line strongly depends on velocity. All images, spectra, and IFU data were created with the Monte Carlo radiation transfer code RASCAS Leo2020.
• Figure 3: Mass growth histories of the main progenitor in the three sets of initial conditions compared to the typical mass growth histories of 28,475 similar mass haloes from IllustrisTNG-300. The dark and light shaded regions represent the $1\sigma$ and $2\sigma$ results from IllustrisTNG-300 with the black line representing the median relation. The early forming halo (purple) is used for the high-redshift suite as it results in more numerous massive progenitors at high redshift rather than a single dominant object. The reference and late forming initial conditions (dashed blue and green lines, respectively) are only used for the CGM suite.
• Figure 4: Histograms of halo mass (top) and stellar mass (bottom) across all snapshots for each simulation in the high-redshift suite. In the top panel we indicate the halo mass that corresponds to 1,000 DM particles, which represents the minimum required to be included in our spectroscopic sample.
• Figure 5: Star formation histories of Pop. II stars (solid) and Pop. III stars (dotted) across the entire Lagrange volume for each of the high-redshift suite simulations.