Investigating the star formation histories of galaxies from Cosmic Dawn to the Epoch of Reionization with the Santa Cruz SAM

Abstract: The James Webb Space Telescope (JWST) has opened a new window onto galaxy evolution in the very early Universe. In this work, we leverage halo merger trees extracted from the GUREFT dark-matter-only cosmological simulation suite together with the Santa Cruz semi-analytic model (SAM) for galaxy formation to investigate the predicted star formation histories (SFHs) of galaxies from cosmic dawn (z ~ 14) to the end of the Epoch of Reionization (EoR; z~6). While we find that on average, median SFHs of galaxies across all masses are uniformly and rapidly rising over time from 14 < z < 6 as expected, individual galaxy SFHs show a range of diverse SFHs, even for a fixed terminal mass or redshift, with bursts and mini-quenching episodes in agreement with SFHs inferred from observations. The median lookback time to form the youngest 50% (t_50) and 90% (t_90) of galaxies' stars decreases weakly with increasing stellar mass, and strongly with the redshift of observation. For galaxies at z>12, we find typical values of t_50 < 30 Myr and t_90 < 70 Myr, a factor of ~3 to 4 shorter than for comparable galaxies near the end of EoR (z ~ 6). The young-star dominated nature of stellar populations in ultra-high-z galaxies implies that careful modelling of young stellar populations is crucial for obtaining accurate synthetic photometry. In addition, our results have important implications for interpreting observational indicators of star formation histories and timescales.

• The paper presents a detailed Santa Cruz SAM analysis showing rapidly rising, bursty star formation histories in galaxies from z≈14 to z≈6.
• It leverages high-resolution GUREFT merger trees and refined age-binning to reveal short stellar assembly timescales and significant UV luminosity enhancements.
• The study underscores the need for accurate SFH modeling in JWST SED analysis, cautioning against oversimplified star formation prescriptions.

Star Formation Histories from Cosmic Dawn to the Epoch of Reionization: Santa Cruz SAM Predictions

Introduction

This paper presents a comprehensive investigation into the star formation histories (SFHs) of galaxies during the early Universe, spanning from Cosmic Dawn ($z\sim14$) to the end of the Epoch of Reionization (EoR; $z\sim6$), leveraging the Santa Cruz semi-analytic model (SAM) for galaxy formation. The analysis is powered by dark-matter halo merger trees from the ultra-high-resolution GUREFT $N$-body simulation suite, designed specifically to capture the rapid and stochastic assembly of halos at $z>6$. The primary goals are to produce physically-motivated predictions on the timescales, diversity, and burstiness of star formation at $z\gtrsim6$, and to assess the implications of these SFHs for interpreting JWST observations and theoretical galaxy formation models.

Model Framework: Santa Cruz SAM Coupled to GUREFT

The Santa Cruz SAM encodes a suite of physics modules representing gas inflow, cooling, star formation, chemical evolution, feedback, and merging, evolving baryonic components along dark-matter halo merger trees. For this work, the SAM is configured with an $\mathrm{H}_2$-regulated star formation law and a multi-phase gas partitioning prescription, yielding SFRs as a function of stellar age and metallicity. The GUREFT suite provides tiered simulation volumes with mass and time resolution capable of resolving the assembly of dwarf-mass halos and capturing the high merger and formation rates at $z>10$.

A significant technical improvement is implemented in the SFH-to-synthetic SED mapping: the model rebinning of SFHs from native 10 Myr age bins to $\sim$0.67 Myr bins allows accurate quantification of the UV luminosity and spectral contributions from the youngest stellar populations.

Star Formation Histories: Characteristics and Drivers

The ensemble of predicted SFHs from $6

• Rapidly-Rising Star Formation: Median SFHs in all mass bins at $z\gtrsim6$ exhibit strong, monotonically rising SFRs as a function of time, leading to stellar populations dominated by extremely young stars.
• Diversity at Fixed Mass: Individual SFHs, even at fixed observed mass and redshift, display significant diversity, including stochastic bursts and mini-quenching episodes driven by mergers and feedback cycles, consistent with short-timescale SFR fluctuations inferred observationally.
• Assembly Timescales: For galaxies at $z\sim6$0, the median time required to form the youngest half of the stellar mass ($z\sim6$1) is $z\sim6$2 Myr, with $z\sim6$3 Myr. These are factors of $z\sim6$4–4 shorter than for galaxies of similar mass at $z\sim6$5. The median $z\sim6$6 and $z\sim6$7 weakly decrease with increasing mass but depend strongly on redshift.
• Star Formation Efficiency Trends: Both instantaneous and integrated star formation efficiencies (SFE) are higher in more massive systems, and increase along the evolutionary tracks to lower redshift, reflecting both increasing halo mass and evolution of galaxy-scale feedback regulation.
• UV Luminosity Function Sensitivity: Incorporating the improved age-binning reveals that UV magnitudes computed with $z\sim6$8 Myr resolution are $z\sim6$9–2 mag brighter compared to 10 Myr binning at $N$0. This adjustment brings model UV luminosity functions (UVLFs) into quantitative agreement with JWST-detected bright galaxies up to $N$1 without invoking changes in the underlying physics prescriptions.

Burstiness, Timescale Diagnostics, and SED Modelling Implications

Analyses of SFR ratios (e.g., $N$2, where the indices indicate SFR averaged over the last 20 and 100 Myr) demonstrate that:

• At $N$3, most galaxies are strongly dominated by ongoing, extremely young star formation.
• The median $N$4 ratio declines towards lower redshift, closely tracking the increase in stellar assembly timescales rather than any fundamental shift to less stochastic SFH. Thus, high values of this ratio at ultra-high redshift predominantly reflect globally rapid mass assembly, not increased stochasticity.
• The SF timescales for UV-bright galaxies at $N$5 are substantially shorter than the standard timescales ($N$6 Myr) over which UV luminosity is often assumed to trace SFR, challenging the interpretation of single "conversion factors" in SED fitting at early times.

This dominance of young stellar populations has immediate consequences for both forward modeling and the interpretation of high-$N$7 galaxy photometry:

• Failing to resolve the youngest stellar populations with sufficiently fine age bins in SFH-to-SED mapping underestimates UV and nebular line luminosities, biasing inferred stellar masses and SFRs.
• SED fitting utilizing parametric SFHs (e.g., $N$8-models, lognormals) or coarse time bins can systematically overestimate stellar mass and mischaracterize SFH in ultra-high-$N$9 samples, since real galaxies rarely exhibit smooth, slowly declining SFRs during this epoch.
• Theoretical priors informed by these results—favoring rising, short-duration SFHs when fitting early-universe SEDs—should result in more physically accurate recovery of galaxy properties.

Physical Origin and Broader Implications

The ubiquity of rapidly rising, young-star-dominated SFHs fundamentally arises from the cosmological context of high-$z>6$0 structure formation: short cosmic timescales, high merger rates, and high average gas surface densities in halos above the atomic cooling threshold. "Downsizing" is evident in the sense that more massive galaxies at a given redshift assemble their stellar mass on shorter timescales, but, at these redshifts, the driver is the higher SFE and earlier progenitor mass growth for high-mass halos (a consequence of merger bias and feedback regulation), not early quenching.

Limitations remain in the treatment of very short timescale (<10 Myr) star formation stochasticity, as sub-GMC scale physics is necessarily parametrized. The GUREFT simulation's dynamic range imposes uncertainties in the earliest star formation in the rarest, most massive halos, but these limitations have a modest effect on the median SFH properties at $z>6$1. The analysis calls for complementing SAMs and simulations with physically informed SFH priors in observational modeling, and motivates further investigation of burstiness diagnostics jointly with clustering and emission-line properties.

Future developments may include integration of nebular emission models and generative models of merger trees to further extend the dynamic range and capture the earliest galaxy assembly phases.

Conclusion

The study presents a self-consistent theoretical framework, based on the Santa Cruz SAM and high-fidelity GUREFT merger trees, capable of predicting the star formation histories, luminosities, and assembly timescales of galaxies from Cosmic Dawn to the end of Reionization. The work highlights the necessity of high temporal resolution in SFH modeling for both predictive accuracy and comparison with JWST-era observations. The results demonstrate that the observed abundance and properties of ultra-high-$z>6$2 galaxies can be reconciled with standard cosmological and galaxy formation models once the impact of extremely young stellar populations is treated accurately in the synthetic photometry. The implications span galaxy evolution theory, the analysis of deep-field surveys, and the prescription of SFH priors for SED fitting at the redshift frontier.

---

Reference: "Investigating the star formation histories of galaxies from Cosmic Dawn to the Epoch of Reionization with the Santa Cruz SAM" (2607.02650)