# Modeling Dust Production, Growth, and Destruction in Reionization-Era Galaxies with the CROC Simulations II: Predicting the Dust Content of High-Redshift Galaxies

**Authors:** Clarke J. Esmerian, Nickolay Y. Gnedin  
**arXiv:** [2308.11723](https://arxiv.org/abs/2308.11723)  
**Date:** 2023-08-22  
**Summary page:** https://www.summarizepaper.com/en/paper/2308.11723v1/

## Key points

- Investigate interstellar dust content of galaxies during the reionization era
- Focus on galaxies with stellar masses ranging from $10^5$ to $10^9$ times the mass of our Sun within the first 1.2 billion years of the universe's existence
- Employ a post-processing method to model dust creation and destruction processes
- Determine if observable quantities related to dust in early galaxies can provide constraints on dust physics
- Use simulations to forward model observable properties and compare them to existing observational data
- None of the models can simultaneously match all existing observational constraints
- Models predicting larger amounts of dust with a ratio of dust mass to metallicity ($D/Z$) greater than or equal to 0.1 at redshift $z=5$ are preferred based on constraints related to total dust mass and infrared luminosities
- These models produce excessive extinction in the ultraviolet (UV), preventing them from matching observations of $\beta_{\rm UV}$
- Simulated distributions of stars and dust appear overly symmetric compared to existing data, likely due to limitations in their stellar feedback model
- Observable properties related to interstellar dust serve as a strong test for stellar feedback mechanisms in high-redshift galaxies
- Current simulations cannot reproduce observed dust masses and optical depths, suggesting a need for more dynamic modeling approaches

## Summary

In this study, the authors investigate the interstellar dust content of galaxies during the reionization era using cosmological, fluid-dynamical simulations. They focus on galaxies with stellar masses ranging from approximately $10^5$ to $10^9$ times the mass of our Sun within the first 1.2 billion years of the universe's existence. To model the dust content, they employ a post-processing method that takes into account dust creation and destruction processes. By systematically varying these parameters, they aim to determine whether observable quantities related to dust in galaxies at these early epochs can provide constraints on dust physics. The authors then use their simulations to forward model observable properties of these high-redshift galaxies and compare them to existing observational data. However, they find that none of their models can simultaneously match all existing observational constraints. Specifically, models predicting larger amounts of dust with a ratio of dust mass to metallicity ($D/Z$) greater than or equal to 0.1 at redshift $z=5$ are preferred based on constraints related to total dust mass and infrared luminosities; however, these models produce excessive extinction in the ultraviolet (UV), preventing them from matching observations of $\beta_{\rm UV}$. To further investigate this discrepancy, the authors analyze the relative spatial distribution of stars and dust as probed by infrared (IR) and UV emission and find that their simulated distributions appear overly symmetric compared to existing data - likely due to limitations in their stellar feedback model. Overall, this study highlights that observable properties related to interstellar dust in high-redshift galaxies serve as a strong test for stellar feedback mechanisms. The inability of current simulations to simultaneously reproduce observed dust masses and optical depths suggests a need for more dynamic modeling approaches that capture a wider range of column densities in order to better understand how star formation affects interstellar medium evolution over cosmic time scales.