# Dust attenuation, dust emission, and dust temperature in galaxies at z>=5: a view from the FIRE-2 simulations

## TL;DR

This study uses high-resolution cosmological simulations to analyze dust properties, emission, and temperature in galaxies at redshifts z>=5, revealing key relations and predictions relevant for understanding early galaxy evolution.

## Contribution

The paper introduces a comprehensive simulation-based analysis of dust effects in high-redshift galaxies, highlighting the IRX-beta_UV relation and dust temperature evolution.

## Key findings

- IRX-beta_UV relation mainly depends on extinction curve shape
- Dust temperatures at z>=5 are higher than at z=0, driven by higher SFRs
- Predicted UV and IR luminosity functions align with observations

## Abstract

We present a suite of 34 high-resolution cosmological zoom-in simulations consisting of thousands of halos up to M_halo~10^12 M_sun (M_star~10^10.5 M_sun) at z>=5 from the Feedback in Realistic Environments project. We post-process our simulations with a three-dimensional Monte Carlo dust radiative transfer code to study dust extinction, dust emission, and dust temperature within these simulated z>=5 galaxies. Our sample forms a tight correlation between infrared excess (IRX=F_IR/F_UV) and ultraviolet (UV)-continuum slope (beta_UV), despite the patchy, clumpy dust geometry shown in our simulations. We find that the IRX-beta_UV relation is mainly determined by the shape of the extinction curve and is independent of its normalization (set by the dust-to-gas ratio). The bolometric IR luminosity (L_IR) correlates with the intrinsic UV luminosity and the star formation rate (SFR) averaged over the past 10 Myr. We predict that at a given L_IR, the peak wavelength of the dust spectral energy distributions for z>=5 galaxies is smaller by a factor of 2 (due to higher dust temperatures on average) than at z=0. The higher dust temperatures are driven by higher specific SFRs and SFR surface densities with increasing redshift. We derive the galaxy UV luminosity functions (LFs) at z=5-10 from our simulations and confirm that a heavy attenuation is required to reproduce the observed bright-end UVLFs. We also predict the IRLFs and UV luminosity densities at z=5-10. We discuss the implications of our results on current and future observations probing dust attenuation and emission in z>=5 galaxies.