Calibrated, cosmological hydrodynamical simulations with variable IMFs I: Method and effect on global galaxy scaling relations

Abstract

The recently inferred variations in the stellar initial mass function (IMF) among local high-mass early-type galaxies may require a reinterpretation of observations of galaxy populations and may have important consequences for the predictions of models of galaxy formation and evolution. We present a new pair of cosmological, hydrodynamical simulations based on the EAGLE model that self-consistently adopt an IMF that respectively becomes bottom-or top-heavy in high-pressure environments for individual star-forming gas particles. In such models, the excess stellar mass-to-light (M/L) ratio with respect to a reference IMF is increased due to an overabundance of low-mass dwarf stars or stellar remnants, respectively. Crucially, both pressuredependent IMFs have been calibrated to reproduce the observed trends of increasing excess M/L with central stellar velocity dispersion (σ e ) in early-type galaxies, while maintaining agreement with the observables used to calibrate the EAGLE model, namely the galaxy luminosity function, half-light radii of late-type galaxies, and black hole masses. We find that while the M/L excess is a good measure of the IMF for low-mass slope variations, it depends strongly on the age of the stellar population for high-mass slope variations. The normalization of the [Mg/Fe]−σ e relation is decreased (increased) for bottom-(top-)heavy IMF variations, while the slope is not strongly affected. Bottom-heavy variations have little impact on galaxy metallicities, half-light radii of early-type galaxies, or star formation rates, while top-heavy variations significantly increase these quantities for high-mass galaxies, leading to tension with observations.

SDSS-IV MaNGA: stellar initial mass function variation inferred from Bayesian analysis of the integral field spectroscopy of early-type galaxies

“…Clauwens et al 2016) and cosmological simulations (e.g. Gutcke & Springel 2019;Barber et al 2018Barber et al , 2019a.…”

The evolution of the UV-to-mm extragalactic background light: evidence for a top-heavy initial mass function?

“…We note that Somerville et al (2012), using a semi-analytical galaxy formation model assuming a universal IMF, found a reasonable match to the EBL, cosmic star formation history and present-day K-band luminosity function, but under-predicted the number counts of galaxies at 850 µm. Other galaxy formation studies have considered IMF variations, such as Gargiulo et al (2015) and Fontanot et al (2017), who invoked a top-heavy IMF in regions of high star formation in semi-analytical models, and Barber et al (2018), who imposed a pressure dependent IMF in an EAGLE hydrodynamical simulation; these studies did not consider the EBL. This paper is structured as follows: in Section 2 we introduce the theoretical model, which incorporates a semianalytical model of galaxy formation implemented within halo merger trees derived from a Millennium-style dark matter only N-body simulation (Springel et al 2005;Baugh et al 2019) and the radiative transfer code, grasil (Silva et al 1998), for computing the absorption and re-emission of stellar radiation by interstellar dust.…”

Calibrated, cosmological hydrodynamical simulations with variable IMFs III: spatially resolved properties and evolution

“…These apparently contradictory results could be evidence of a time dependence of IMF variations. Indeed, different forms of IMF parametrization can lead to very different predictions of IMF-related observational diagnostics at the present day, even for a fixed mass-to-light ratio (Barber et al 2018a, hereafter Paper I). These differences may be even stronger at high redshift when the stars are actually forming.…”