Simulated vs. observed UV emission at high redshift: a hint for a clumpy ISM?

TL;DR

This study uses a clumpy dust attenuation model in cosmological simulations to better match observed UV luminosity functions and slopes at high redshift, suggesting a highly obscuring, inhomogeneous interstellar medium.

Contribution

It introduces a dust attenuation model with age-dependent extinction that accounts for a clumpy ISM, improving agreement with observations at high redshift.

Findings

The model fits observed UV luminosity functions and slopes well.

Optical depths in star-forming clouds are 30-100 times higher than the mean.

High-redshift ISM is more clumpy and obscuring than local galaxies.

Abstract

We discuss the rest-frame UV emission between 5< z < 7 from the MareNostrum High-z Universe, a SPH simulation done with more than 2 billion particles. Cosmological simulations of galaxy formation generally overpredict the UV restframe luminosity function at high redshift, both at the bright and faint ends. In this Letter we explore a dust attenuation model where a larger extinction is applied to star populations younger than a given age, mimicking the effect of a clumpy interstellar medium. We show that this scenario fits reasonably well both the UV luminosity functions and the UV-continuum slopes derived from observations. The model assumes a large obscuration for stars younger than 25 Myr from the gas clouds where they should be embedded at their formation time. We find that the optical depth in these clouds should be between 30 and 100 times larger than the mean optical depth for the homogeneous part of the interstellar medium. These values are one order of magnitude larger than those estimated in local galaxies. Therefore, we conclude that LambdaCDM predictions for the high-z UV emission can accommodate the current observations if we consider a dust extinction model based on the assumption of a clumpy environment at high redshift.