Simulating galaxy clusters - I. Thermal and chemical properties of the intracluster medium

Abstract

We have performed a series of N‐body/hydrodynamical (treesph) simulations of clusters and groups of galaxies, selected from a cosmological volume within a Lambda cold dark matter (ΛCDM) framework: these objects have been resimulated at higher resolution to z= 0, in order to follow also the dynamical, thermal and chemical input on to the intracluster medium (ICM) from stellar populations within galaxies. The simulations include metallicity‐dependent radiative cooling, star formation according to different initial mass functions (IMFs), energy feedback as strong starburst‐driven galactic superwinds, chemical evolution with non‐instantaneous recycling of gas and heavy elements, effects of a metagalactic ultraviolet (UV) field and thermal conduction in the ICM. In this paper, the first in a series of three, we derive results, mainly at z= 0, on the temperature and entropy profiles of the ICM, its X‐ray luminosity, the cluster cold components [cold fraction as well as mass‐to‐light ratio (MLR)] and the metal distribution between ICM and stars. In general, models with efficient superwinds (produced by the action of supernovae and, in some simulations, of active galactic nuclei (AGNs), along with a top‐heavy stellar IMF, are able to reproduce fairly well the observed LX–T relation, the entropy profiles and the cold fraction: both features are found to be needed in order to remove high‐density and low‐entropy cold gas at core scales, although additional alternative feedback mechanisms would still be required to prevent late‐time central cooling flows, and subsequent overproduction of stars and heavy elements at the centre. Observed radial ICM temperature profiles can be matched, except for the gradual decline in temperature inside r∼ 0.1Rvir. Metal enrichment of the ICM gives rise to somewhat steep inner iron gradients; yet, the global level of enrichment compares well to observational estimates when a top‐heavy IMF is adopted, and after correcting for the stars formed at late times at the base of the cooling flows, the metal partition between stars and ICM gets into good agreement with observations. The overall abundance and profile of iron in the ICM is found essentially unchanged from z= 1 to present time. Finally, the α/Fe of the gas is found to increase steadily with radius, decreasing over time.

The Chemical Enrichment of the ICM from Hydrodynamical Simulations

“…Valdarnini (2003) performed an extended set of cluster simulations and showed that profiles of the iron abundance are steeper than the observed ones. A similar analysis has been presented by Romeo et al (2006), who also considered the effect of varying the IMF and the feedback efficiency on the enrichment pattern of the ICM. Scannapieco et al (2005) presented an implementation of a model of chemical enrichment in the GADGET-2 code, coupled to a self-consistent model for star formation and feedback.…”

“…As an example, we show in the left panel of Fig. 8 the comparison between the iron profiles measured by De Grandi et al (2004) and results from the chemodynamical SPH simulations by Romeo et al (2006). These authors performed simulations for two clusters having sizes comparable to those of the Virgo and of the Coma cluster, respectively.…”

“…In principle, a number of processes, acting at different scales, are expected to play a role in determining the spatial distribution of metals: galactic winds (e.g. Tornatore et al 2004;Romeo et al 2006), transport by buoyant bubbles (Roediger et al 2007;Sijacki et al 2007), ram-pressure stripping (e.g., Kapferer et al 2007b), diffusion by stochastic gas motions , sinking of highly enriched low-entropy gas (Cora 2006), etc. It is clear that modelling properly all such processes represents a non-trivial task for numerical simulations, which are required both to cover a wide dynamical range and to provide a reliable description of a number of complex astrophysical processes (Schindler and Diaferio 2008-Chap.…”

Rhapsody-G simulations – II. Baryonic growth and metal enrichment in massive galaxy clusters

“…Cosmological hydrodynamical simulations have been used with similar goals (Valdarnini 2003;Romeo et al 2006;Tornatore et al 2007;Davé et al 2008;Wiersma et al 2011;Crain et al 2013;Skory et al 2013): the results by Tornatore et al (2007) showed good agreement of metal abundances in simulated clusters in absence of AGN feedback at the price of significant over-production of galaxy stellar masses. More recent papers highlighted that much better agreement between stellar masses and cluster metallicities with observational results can be achieved when AGN feedback is included (Wiersma et al 2011;Crain et al 2013;Skory et al 2013).…”

“…The history of theoretical studies of galaxy clusters is long and one of the approaches that has been followed by several authors to model these systems and to understand the origin of their metal content is through semi-analytical models coupled to N-body simulations (Cora et al 2008;Kapferer et al 2009;Arieli et al 2010;Short et al 2013) or cosmological hydrodynamical simulations (Valdarnini 2003;Romeo et al 2006;Tornatore et al 2007;Davé et al 2008;Wiersma et al 2011;Crain et al 2013;Skory et al 2013). Most state-of-the-art cosmological zoom-in simulations of galaxy clusters reach a spatial resolution of "kpc (Puchwein et al 2008;Teyssier et al 2011;Ettori et al 2012;Martizzi et al 2014b;Le Brun et al 2014;Lau et al 2015;Sembolini et al 2015).…”