The mass and structural evolution of massive galaxies is one of the hottest topics in galaxy formation. This is because it may reveal invaluable insights into the still debated evolutionary processes governing the growth and assembly of spheroids. However, direct comparison between models and observations is usually prevented by the so-called progenitor bias, i.e., new galaxies entering the observational selection at later epochs, thus eluding a precise study of how pre-existing galaxies actually evolve in size. To limit this effect, we here gather data on high-redshift brightest group and cluster galaxies, evolve their (mean) host halo masses down to z = 0 along their main progenitors, and assign as their "descendants" local Sloan Digital Sky Survey central galaxies matched in host halo mass. At face value, the comparison between high redshift and local data suggests a noticeable increase in stellar mass of a factor of 2 since z ∼ 1, and of 2.5 in mean effective radius. We then compare the inferred stellar mass and size growth with those predicted by hierarchical models for central galaxies, selected at high redshifts to closely match the halo and stellar mass bins as in the data. Only hierarchical models characterized by very limited satellite stellar stripping and parabolic orbits are capable of broadly reproducing the stellar mass and size increase of a factor of ∼2-4 observed in cluster galaxies since z ∼ 1. The predicted, average (major) merger rate since z ∼ 1 is in good agreement with the latest observational estimates.

The stellar mass-halo mass relation is a key constraint in all semi-analytic, numerical, and semi-empirical models of galaxy formation and evolution. However, its exact shape and redshift dependence remain debated. Several recent works support a relation in the local Universe steeper than previously thought. Based on the comparisons with a variety of data on massive central galaxies, we show that this steepening holds up to z ∼ 1, for stellar masses M star 2 × 10 11 M ⊙ . Specifically, we find significant evidence for a high-mass end slope of β 0.35 − 0.70, instead of the usual β 0.20 − 0.30 reported by a number of previous results. When including the independent constraints from the recent BOSS clustering measurements, the data, independent of any systematic errors in stellar masses, tend to favor a model with a very small scatter ( 0.15 dex) in stellar mass at fixed halo mass, in the redshift range z < 0.8 and for M star > 3 × 10 11 M ⊙ , suggesting a close connection between massive galaxies and host halos even at relatively recent epochs. We discuss the implications of our results with respect to the evolution of the most massive galaxies since z ∼ 1.

We investigate the non-linear evolution of Baryon Acoustic Oscillations (BAO) in the low-redshift matter power spectrum from the DEUS-FUR ΛCDM model simulation. This is the first cosmological N-body simulation encompassing the full observable cosmic volume, thus allowing cosmic variance limited predictions at BAO scales. We control the effect of numerical systematic errors using a series of large volume high-resolution simulations. The combined analysis allows us to measure the matter power spectrum between z = 0 and 1 to 1% over the entire BAO range, 0.03 < k [h Mpc −1 ] < 0.3, in bins of size ∆k/k 1%. We define the BAO with respect to a non-linearly evolved wiggle-free spectrum and determine the characteristics of the BAO without recurring to extrapolation from global fitting functions. We quantify the effects of non-linearities on the position and amplitude of the BAO extrema, and the coupling to the broadband slope of the power spectrum. We use these estimates to test non-linear predictions from semi-analytical models. Quite remarkably from the analysis of the redshift evolution of BAO we find that the second dip and third peak remains unaltered by non-linear effects. Furthermore, we find that the square of the damping factor and the shift of the position of BAO extrema scale to good approximation as the square of the growth factor, in agreement with expectations from perturbation theory. This confirms the idea that, besides cosmic distances, an accurate measurement of BAO at different redshifts can directly probe the growth of cosmic structures.

The triggering mechanisms for Active Galactic Nuclei (AGN) are still debated. Some of the most popular ones include galaxy interactions (IT) and disk instabilities (DI). Using an advanced semi analytic model (SAM) of galaxy formation, coupled to accurate halo occupation distribution modeling, we investigate the imprint left by each separate triggering process on the clustering strength of AGN at small and large scales. Our main results are as follows: i) DIs, irrespective of their exact implementation in the SAM, tend to fall short in triggering AGN activity in galaxies at the center of halos with M h > 10 13.5 h −1 M . On the contrary, the IT scenario predicts abundance of active, central galaxies that generally agrees well with observations at every halo mass.ii) The relative number of satellite AGN in DIs at intermediate-to-low luminosities is always significantly higher than in IT models, especially in groups and clusters. The low AGN satellite fraction predicted for the IT scenario might suggest that different feeding modes could simultaneously contribute to the triggering of satellite AGN. iii) Both scenarios are quite degenerate in matching large-scale clustering measurements, suggesting that the sole average bias might not be an effective observational constraint. iv) Our analysis suggests the presence of both a mild luminosity and a more consistent redshift dependence in the AGN clustering, with AGN inhabiting progressively less massive dark matter halos as the redshift increases. We also discuss the impact of different observational selection cuts in measuring AGN clustering, including possible discrepancies between optical and X-ray surveys.On the other hand, there is mounting observational evidence based on the star-forming properties and on the mor-

Observations of colliding galaxy clusters with high relative velocity probe the tail of the halo pairwise velocity distribution with the potential of providing a powerful test of cosmology. As an example it has been argued that the discovery of the Bullet Cluster challenges standard ΛCDM model predictions. Halo catalogs from N-body simulations have been used to estimate the probability of Bullet-like clusters. However, due to simulation volume effects previous studies had to rely on a Gaussian extrapolation of the pairwise velocity distribution to high velocities. Here, we perform a detail analysis using the halo catalogs from the Dark Energy Universe Simulation Full Universe Runs (DEUS-FUR), which enables us to resolve the high-velocity tail of the distribution and study its dependence on the halo mass definition, redshift and cosmology. Building upon these results we estimate the probability of Bullet-like systems in the framework of Extreme Value Statistics. We show that the tail of extreme pairwise velocities significantly deviates from that of a Gaussian, moreover it carries an imprint of the underlying cosmology. We find the Bullet Cluster probability to be two orders of magnitude larger than previous estimates, thus easing the tension with the ΛCDM model. Finally, the comparison of the inferred probabilities for the different DEUS-FUR cosmologies suggests that observations of extreme interacting clusters can provide constraints on dark energy models complementary to standard cosmological tests.