The internal metallicity distributions of simulated galaxies from EAGLE, Illustris, and IllustrisTNG at z=1.8-4 as probed by Gamma Ray Burst hosts

TL;DR

This study compares simulated galaxy metallicity distributions from three cosmological models to observed GRB host data, revealing differences that can help refine galaxy evolution theories and progenitor bias understanding.

Contribution

It predicts the relationship between absorption and emission metallicities in simulated galaxies and compares these predictions with observational data to evaluate model realism.

Findings

Different simulations show distinct Zabs-Zem relationships.

Marginal support for Illustris as most realistic model.

Future JWST observations will help discriminate models.

Abstract

Massive stars are thought to be progenitors of Long Gamma Ray Bursts (GRBs), most likely with a bias favouring low metallicity progenitors. Because galaxies do not have a constant metallicity throughout, the combination of line-of-sight absorption metallicity inferred from GRB afterglow spectroscopy and of host galaxy global metallicity derived from emission lines diagnostics represents a powerful way to probe both the bias function for GRB progenitors, and the chemical inhomogeneities across star forming regions. In this study, we predict the relationship between Zabs and Zem using three different hydrodynamical cosmological simulations: Illustris, EAGLE, and IllustrisTNG. We find that while the qualitative shape of the curve relating emission versus absorption metallicity remains the same, the predicted relationship between these two observables is significantly different between the simulations. Using data for the host galaxy of GRB121024A for which both Zabs and Zem have been measured, we find marginal support for the Illustris simulation as producing the most-realistic internal metallicity distributions within star-forming galaxies at cosmic noon. Overall, all simulations predict similar properties for the bulk of the GRB host galaxy population, but each has distinct features in the tail of the Zabs-Zem distribution that in principle allow to discriminate between models if a sufficiently large sample of observations are available (i.e. N>11 on average). Substantial progress is expected in the near future, with upcoming JWST/NIRspec observations of 10 GRB host galaxies for which absorption metallicity from the afterglow spectra exists.