Blue galaxies: modelling nebular He II emission in high redshift galaxies

Using cosmological simulations to make useful, scientifically relevant emission line predictions is a relatively new and rapidly evolving field. However, nebular emission lines have been particularly challenging to model because they are extremely sensitive to the local photoionization balance, which can be driven by a spatially dispersed distribution of stars amidst an inhomogeneous absorbing medium of dust and gas. As such, several unmodelled mysteries in observed emission line patterns exist in the literature. For example, there is some question as to why He II lambda 4686/H beta ratios in observations of lower metallicity dwarf galaxies tend to be higher than model predictions. Since hydrodynamic cosmological simulations are best suited to this mass and metallicity regime, this question presents a good test case for the development of a robust emission line modelling pipeline. The pipeline described in this work can model a process that produces high He II lambda 4686/H beta ratios and eliminate some of the modelling discrepancy for ratios below 3 per cent without including AGNs, X-ray binaries, high mass binaries, or a top-heavy stellar initial mass function. These ratios are found to be more sensitive to the presence of 15 Myr or longer gaps in the star formation histories than to extraordinary ionization parameters or specific star formation rates. They also closely correspond to the WR phase of massive stars. In addition to the investigation into He II lambda 4686/H beta ratios, this work charts a general path forward for the next generation of nebular emission line modelling studies.