The cold dark matter model has become the leading theoretical picture for the formation of structure in the Universe. This model, together with the theory of cosmic inflation, makes a clear prediction for the initial conditions for structure formation and predicts that structures grow hierarchically through gravitational instability. Testing this model requires that the precise measurements delivered by galaxy surveys can be compared to robust and equally precise theoretical calculations. Here we present a simulation of the growth of dark matter structure using 2,160(3) particles, following them from redshift z = 127 to the present in a cube-shaped region 2.230 billion lightyears on a side. In postprocessing, we also follow the formation and evolution of the galaxies and quasars. We show that baryon-induced features in the initial conditions of the Universe are reflected in distorted form in the low-redshift galaxy distribution, an effect that can be used to constrain the nature of dark energy with future generations of observational surveys of galaxies.

Aims. We examine radial and vertical metallicity gradients using a suite of disk galaxy hydrodynamical simulations, supplemented with two classic chemical evolution approaches. We determine the rate of change of gradient slope and reconcile the differences existing between extant models and observations within the canonical "inside-out" disk growth paradigm. Methods. A suite of 25 cosmological disks is used to examine the evolution of metallicity gradients; this consists of 19 galaxies selected from the RaDES (Ramses Disk Environment Study) sample, realised with the adaptive mesh refinement code ramses, including eight drawn from the "field" and six from "loose group" environments. Four disks are selected from the MUGS (McMaster Unbiased Galaxy Simulations) sample, generated with the smoothed particle hydrodynamics (SPH) code gasoline. Two chemical evolution models of inside-out disk growth were employed to contrast the temporal evolution of their radial gradients with those of the simulations. Results. We first show that generically flatter gradients are observed at redshift zero when comparing older stars with those forming today, consistent with expectations of kinematically hot simulations, but counter to that observed in the Milky Way. The vertical abundance gradients at ∼1−3 disk scalelengths are comparable to those observed in the thick disk of the Milky Way, but significantly shallower than those seen in the thin disk. Most importantly, we find that systematic differences exist between the predicted evolution of radial abundance gradients in the RaDES and chemical evolution models, compared with the MUGS sample; specifically, the MUGS simulations are systematically steeper at high-redshift, and present much more rapid evolution in their gradients. Conclusions. We find that the majority of the models predict radial gradients today which are consistent with those observed in late-type disks, but they evolve to this self-similarity in different fashions, despite each adhering to classical "inside-out" growth. We find that radial dependence of the efficiency with which stars form as a function of time drives the differences seen in the gradients; systematic differences in the sub-grid physics between the various codes are responsible for setting these gradients. Recent, albeit limited, data at redshift z ∼ 1.5 are consistent with the steeper gradients seen in our SPH sample, suggesting a modest revision of the classical chemical evolution models may be required.

We present a detailed investigation of a number of different approaches to modelling feedback in simulations of galaxy formation. Gas-dynamic forces are evaluated using Smoothed Particle Hydrodynamics (SPH). Star formation and supernova feedback are included using a three parameter model which determines the star formation rate (SFR) normalization, feedback energy and lifetime of feedback regions. The star formation rate is calculated for all gas particles which fall within prescribed temperature, density and convergent flow criteria, and for cosmological simulations we also include a self-gravity criterion for gas particles to prevent star formation at high redshifts. A Lagrangian Schmidt law is used to calculate the star formation rate from the SPH density. Conversion of gas to stars is performed when the star mass for a gas particle exceeds a certain limit, typically half that of the gas particle. Feedback is incorporated by returning a precalculated amount of energy to the ISM as thermal heating. We compare the effects of distributing this energy over the smoothing scale or depositing it on a single particle. Radiative losses are prevented from heated particles by adjusting the density used in radiative cooling so that the decay of energy occurs over a set half-life, or by turning off cooling completely and allowing feedback regions a brief period of adiabatic expansion. We test the models on the formation of galaxies from cosmological initial conditions and also on isolated disk galaxies. For isolated prototypes of the Milky Way and the dwarf galaxy NGC 6503 we find feedback has a significant effect, with some algorithms being capable of unbinding gas from the dark matter halo ('blow-away'). As expected feedback has a stronger effect on the dwarf galaxy, producing significant disk evaporation and also larger feedback 'bubbles' for the same parameters. In the critical-density CDM cosmological simulations, evolved to a redshift z = 1, we find the reverse to be true. Further, feedback only manages to produce a disk with a specific angular momentum value approximately twice that of the run with no feedback, the disk thus has an specific angular momentum value that is characteristic of observed elliptical galaxies. We argue that this is a result of the extreme central concentration of the dark halos in the standard CDM model and the pervasiveness of the core-halo angular momentum transport mechanism (even in light of feedback). A simulation with extremely violent feedback, relative to our fiducial models, leads to a disk that resembles the other simulations at z = 1 and has a specific angular momentum value that is more typical of observed disk galaxies. At later times, z = 0.5, a large amount of halo gas which does not suffer an angular momentum deficit is present, however the cooling time is too long to accrete on to the disk. We further point out that the disks formed in hierarchical simulations are partially a numerical artifact produced by the minimum mass scale of the simulation acting as a highly efficient...