Comparing E-MOSAICS predictions of high-redshift proto-globular clusters with JWST observations in lensed galaxies

TL;DR

This study compares high-redshift star cluster predictions from the E-MOSAICS simulation with JWST observations, finding general agreement but noting some discrepancies in brightness and mass that require further investigation.

Contribution

It provides a direct comparison between simulated star cluster properties and actual high-redshift observations, testing the validity of the E-MOSAICS model in early universe conditions.

Findings

Brightest cluster ages match well between model and observations

Observed clusters tend to be brighter and more massive than predictions

Results support common formation mechanisms for star clusters across cosmic time

Abstract

High-resolution imaging and strong gravitational lensing of high-redshift galaxies have enabled the detection of compact sources with properties similar to nearby massive star clusters. Often found to be very young, these sources may be globular clusters detected in their earliest stages. In this work, we compare predictions of high-redshift ($z \sim 1$--$10$) star cluster properties from the E-MOSAICS simulation of galaxy and star cluster formation with those of the star cluster candidates in strongly lensed galaxies from James Webb (JWST) and Hubble Space Telescope (HST) imaging. We select galaxies in the simulation that match the luminosities of the majority of lensed galaxies with star cluster candidates observed with JWST. We find that the luminosities, ages and masses of the brightest star cluster candidates in the high-redshift galaxies are consistent with the E-MOSAICS model. In particular, the brightest cluster ages are in excellent agreement. The results suggest that star clusters in both low- and high-redshift galaxies may form via common mechanisms. However, the brightest clusters in the lensed galaxies tend to be $\approx 1$--$1.5$ mag brighter and $\approx 0.5$ dex more massive than the median E-MOSAICS predictions. We discuss the large number of effects that could explain the discrepancy, including simulation and observational limitations, stellar population models, cluster detection biases and nuclear star clusters. Understanding these limitations would enable stronger tests of globular cluster formation models.