The hydrogen Lyα forest is an important probe of the z > 2 Universe that is otherwise challenging to observe with galaxy redshift surveys, but this technique has traditionally been limited to 1D studies in front of bright quasars. However, by pushing to faint magnitudes (g > 23) with 8-10m large telescopes it becomes possible to exploit the high area density of high-redshift star-forming galaxies to create 3D tomographic maps of large-scale structure in the foreground. I describe the first pilot observations using this technique, as well discuss future surveys and the resulting science possibilities for galaxy evolution and cosmology.

The intergalactic medium (IGM) accounts for ≳ 90% of baryons at all epochs and yet its three dimensional distribution in the cosmic web remains mostly unknown. This is so because the only feasible way to observe the bulk of the IGM is through intervening absorption line systems in the spectra of bright background sources, which limits its characterization to being one-dimensional. Still, an averaged three dimensional picture can be obtained by combining and cross-matching multiple one-dimensional IGM information with three-dimensional galaxy surveys. Here, we present our recent and current efforts to map and characterize the IGM in the cosmic web using galaxies as tracers of the underlying mass distribution. In particular, we summarize our results on: (i) IGM around star-forming and non-star-forming galaxies; (ii) IGM within and around galaxy voids; and (iii) IGM in intercluster filaments. With these datasets, we can directly test the modern paradigm of structure formation and evolution of baryonic matter in the Universe.

We have developed a new method to approach the missing baryons problem. We assume that the missing baryons reside in a form of Warm Hot Intergalactic Medium, i.e. the WHIM. Our method consists of (a) detecting the coherent large scale structure in the spatial distribution of galaxies that traces the Cosmic Web and that in hydrodynamical simulations is associated to the WHIM, (b) mapping its luminosity into a galaxy luminosity density field, (c) using numerical simulations to relate the luminosity density to the density of the WHIM, (d) applying this relation to real data to trace the WHIM using the observed galaxy luminosities in the Sloan Digital Sky Survey and 2dF redshift surveys. In our application we find evidence for the WHIM along the line of sight to the Sculptor Wall, at redshifts consistent with the recently reported X-ray absorption line detections. Our indirect WHIM detection technique complements the standard method based on the detection of characteristic X-ray absorption lines, showing that the galaxy luminosity density is a reliable signpost for the WHIM. For this reason, our method could be applied to current galaxy surveys to optimise the observational strategies for detecting and studying the WHIM and its properties. Our estimates of the WHIM hydrogen column density NH in Sculptor agree with those obtained via the X-ray analysis. Due to the additional NH estimate, our method has potential for improving the constrains of the physical parameters of the WHIM as derived with X-ray absorption, and thus for improving the understanding of the missing baryons problem.

We present coupled radiation hydrodynamical simulations of the epoch of reionization, aimed at probing self-feedback on galactic scales. Unlike previous works, which assume a (quasi) homogeneous UV background, we self-consistently evolve both the radiation field and the gas to model the impact of previously unresolved processes such as spectral hardening and self-shielding. We find that the characteristic halo mass with a gas fraction half the cosmic mean, Mc(z), a quantity frequently used in semi-analytical models of galaxy formation, is significantly larger than previously assumed. While this results in an increased suppression of star formation in the early Universe, our results are consistent with the extrapolated stellar abundance matching models from Moster et al. 2013.

We present a mechanism for generating cosmological magnetic fields during the Epoch of Reionization, based on the photoionization of intergalactic hydrogen. A general formula is presented, together with an example numerical application which yields magnetic field strengths between 10−23 to 10−19 G on intersource scales. This mechanism, which operates all along Reionization around any ionizing source, participates to the premagnetization of the whole intergalactic medium. Also, the spatial configuration of these fields may help discriminate them from those produced by other mechanisms in future observations.

Central to Zeldovich's attempts to understand the origin of cosmological structure was his exploration of the fluid dynamical effects in the primordial gas, and how the baryonic dark matter formed. Unfortunately microlensing searches for condensed objects in the foreground of the Magellanic Clouds were flawed by the assumption that the objects would be uniformly (Gaussian) distributed, and because the cadence of daily observations strongly disfavored detection of planet mass microlenses. But quasar microlensing showed them to exist at planetary mass at the same time that a hydro-gravitational theory predicted the planet-mass population as fossils of turbulence at the time of recombination (z = 1100; Gibson 1996, 2001). Where the population has now been detected from MACHO searches to the LMC (Sumi et al. 2011) we compare the quasar microlensing results to the recent determination of the mass distribution function measured for the planetary mass function, and show that the population can account for the baryonic dark matter.

The star formation history of galaxies is a complex process usually considered to be stochastic in nature, for which we can only give average descriptions such as the color-density relation. In this work we follow star-forming gas particles in a hydrodynamical N-body simulation back in time in order to study their initial spatial configuration. By keeping record of the time when a gas particle started forming stars we can produce Lagrangian gas-star isochrone surfaces delineating the surfaces of accreting gas that begin producing stars at different times. These surfaces form a complex a network of filaments in Eulerian space from which galaxies accrete cold gas. Lagrangian accretion surfaces are closely packed inside dense regions, intersecting each other, and as a result galaxies inside proto-clusters stop accreting gas early, naturally explaining the color dependence on density. The process described here has a purely gravitational / geometrical origin, arguably operating at a more fundamental level than complex processes such as AGN and supernovae, and providing a conceptual origin for the color-density relation.

Numerical simulations predict that gas accretion from the cosmic web drives star formation in disks galaxies. The process is important in low mass haloes (< 1012 M⊙), therefore, in the early universe when galaxies were low mass, but also in dwarf galaxies of the local universe. The gas that falls in is predicted to be tenuous, patchy, partly ionized, multi-temperature, and large-scale; therefore, hard to show in a single observation. One of the most compelling cases for gas accretion at work in the local universe comes from the extremely metal poor (XMP) galaxies. They show metallicity inhomogeneities associated with star-forming regions, so that large starbursts have lower metallicity than the underlying galaxy. Here we put forward the case for gas accretion from the web posed by XMP galaxies. Two other observational results are discussed too, namely, the fact that the gas consumption time-scale is shorter than most stellar ages, and the systematic morphological distortions of the HI around galaxies.

We study the evolution of the gaseous components in massive simulated galaxies and show that their early formation is fuelled by cold, low entropy gas streams. At lower redshifts of z ≲ 3 the simulated galaxies are massive enough to support stable virial shocks resulting in a transition from cold to hot gas accretion. The gas accretion history of early-type galaxies is directly linked to the formation of their stellar component in the two phased formation scenario, in which the central parts of the galaxy assemble rapidly through in situ star formation and the later assembly is dominated primarily by minor stellar mergers.

Cosmological hydrodynamical simulations show that the baryonic elements that at z = 0 form the stellar populations of late-type galaxies (LTGs), display, at high z, a gaseous web-like organization, where different singular structures (walls, filaments, nodes) show up. The analysis also shows that the spheroid-to-be elements are the first to be involved in the singular structures, while thick and thin disk-to-be elements chronologically follow them. We discuss how these differences at high z can explain the differences among these three components at z = 0.

We study the role of environment in the evolution of galaxies up to z = 0.7 using the final zCOSMOS-bright data set. We use the colour as a proxy for the quenched population, and measure the dependence of the red fraction of galaxies on stellar mass and two environmental indicators: the local overdensity of galaxies δ and a demarcation of galaxies to centrals and satellites. The analysis is carried out by quantifying the role of different quenching processes. We find that the measured dependence of the red fraction of galaxies on stellar mass and environment can be well described by two quenching processes: one related only to stellar mass (mass quenching) and the other related to the local environment (environment quenching). Within the errors, these processes are independent of each other, and consistent with the z ~ 0 measurement. Moreover, the red fraction of centrals fr,cen (both singleton centrals and centrals in the groups) does not show any trend with δ and more than 95% of fr,cen is consistent with being produced through the mass quenching alone. The satellite galaxies are redder than the centrals at the same stellar mass and δ, requiring additional environment quenching. Given the observed fractional distribution of satellites at different overdensities, the normalized excess in the red fraction of satellites with respect to the red fraction of centrals is consistent with a scenario in which the satellites account for most of the δ-dependences observed in the overall population of galaxies covering 0.1 < z < 0.7.

We take a closer look at the dependence of the galactic colour histogram on the environment density using a volume-limited sample of SDSS galaxies. We find that the strongest changes with environment are taking place with spiral galaxies. In dense environment, discs become considerably redder, apparently due to the shortage of gas, and less concentrated. Contrary to expectation, the mean Sérsic index of luminous elliptical galaxies decreases in denser environments.

The properties of galaxies depend on their environment: red, passive elliptical galaxies are usually located in denser environments than blue, star-forming spiral galaxies. This difference in galaxy populations can be detected at all scales from groups of galaxies to superclusters. In this paper, we will discuss the effect of the large-scale environment on galaxies. Our results suggest that galaxies in superclusters are more likely to be passive than galaxies in voids even when they belong to groups with the same richness. In addition, the galaxies in superclusters are also affected by the morphology of the supercluster: filament-type superclusters contain relatively more red, passive galaxies than spider-type superclusters. These results suggest that the evolution of a galaxy is not determined by its local environment alone, but the large-scale environment also affects.

The smallest dark matter halos are formed first in the early universe. According to recent studies, the central density cusp is much steeper in these halos than in larger halos and scales as ρ ∝ r−(1.5−1.3). We present results of very large cosmological N-body simulations of the hierarchical formation and evolution of halos over a wide mass range, beginning from the formation of the smallest halos. We confirmed early studies that the inner density cusps are steeper in halos at the free streaming scale. The cusp slope gradually becomes shallower as the halo mass increases. The slope of halos 50 times more massive than the smallest halo is approximately -1.3. The concentration parameter is nearly independent of halo mass, and ruling out simple power law mass-concentration relations. The steeper inner cusps of halos near the free streaming scale enhance the annihilation luminosity of a Milky Way sized halo between 12 to 67%.

Filaments of the cosmic web drive spin acquisition of disc galaxies. The point process of filament-type saddle represent best this environment and can be used to revisit the Tidal Torque Theory in the context of an anisotropic peak (saddle) background split. The constrained misalignment between the tidal tensor and the Hessian of the density field generated in the vicinity of filament saddle points simply explains the corresponding transverse and longitudinal point-reflection symmetric geometry of spin distribution. It predicts in particular an azimuthal orientation of the spins of more massive galaxies and spin alignment with the filament for less massive galaxies. Its scale dependence also allows us to relate the transition mass corresponding to the alignment of dark matter halos' spin relative to the direction of their neighboring filament to this geometry, and to predict accordingly it's scaling with the mass of non linearity, as was measured in simulations.

Using the Horizon-AGN simulation we find a mass dependent spin orientation trend for galaxies: the spin of low-mass, rotation-dominated, blue, star-forming galaxies are preferentially aligned with their closest filament, whereas high-mass, velocity dispersion- supported, red quiescent galaxies tend to possess a spin perpendicular to these filaments. We explore the physical mechanisms driving galactic spin swings and quantify how much mergers and smooth accretion re-orient them relative to their host filaments.

Intrinsic alignments are believed to be a major source of systematics for future generation of weak gravitational lensing surveys like Euclid or LSST. Direct measurements of the alignment of the projected light distribution of galaxies in wide field imaging data seem to agree on a contamination at a level of a few per cent of the shear correlation functions, although the amplitude of the effect depends on the population of galaxies considered. Given this dependency, it is difficult to use dark matter-only simulations as the sole resource to predict and control intrinsic alignments. We report here estimates on the level of intrinsic alignment in the cosmological hydrodynamical simulation Horizon-AGN that could be a major source of systematic errors in weak gravitational lensing measurements. In particular, assuming that the spin of galaxies is a good proxy for their ellipticity, we show how those spins are spatially correlated and how they couple to the tidal field in which they are embedded. We will also present theoretical calculations that illustrate and qualitatively explain the observed signals.

Tidal torque theory and simulations of large scale structure predict spin vectors of massive galaxies should be coplanar with sheets in the cosmic web. Recently demonstrated, the giants (Ks ⩽ -22.5 mag) in the Local Volume beyond the Local Sheet have spin vectors directed close to the plane of the Local Supercluster, supporting the predictions of Tidal Torque Theory. However, the giants in the Local Sheet encircling the Local Group display a distinctly different arrangement, suggesting that the mass asymmetry of the Local Group or its progenitor torqued them from their primordial spin directions. To investigate the origin of the spin alignment of giants locally, analogues of the Local Sheet were identified in the SDSS DR9. Similar to the Local Sheet, analogues have an interacting pair of disk galaxies isolated from the remaining sheet members. Modified sheets in which there is no interacting pair of disk galaxies were identified as a control sample.

Galaxies in face-on control sheets do not display axis ratios predominantly weighted toward low values, contrary to the expectation of tidal torque theory. For face-on and edge-on sheets, the distribution of axis ratios for galaxies in analogues is distinct from that in controls with a confidence of 97.6% & 96.9%, respectively. This corroborates the hypothesis that an interacting pair can affect spin directions of neighbouring galaxies.

Galaxies are not randomly distributed across the universe but showing different kinds of alignment on different scales. On small scales satellite galaxies have a tendency to distribute along the major axis of the central galaxy, with dependence on galaxy properties that both red satellites and centrals have stronger alignment than their blue counterparts. On large scales, it is found that the major axes of Luminous Red Galaxies (LRGs) have correlation up to 30Mpc/h. Using hydro-dynamical simulation with star formation, we investigate the origin of galaxy alignment on different scales. It is found that most red satellite galaxies stay in the inner region of dark matter halo inside which the shape of central galaxy is well aligned with the dark matter distribution. Red centrals have stronger alignment than blue ones as they live in massive haloes and the central galaxy-halo alignment increases with halo mass. On large scales, the alignment of LRGs is also from the galaxy-halo shape correlation, but with some extent of mis-alignment. The massive haloes have stronger alignment than haloes in filament which connect massive haloes. This is contrary to the naive expectation that cosmic filament is the cause of halo alignment.

Coherent alignments of galaxy shapes, often called“intrinsic alignments” (IA), are the most significant source of astrophysical uncertainty in weak lensing measurements. We develop the tidal alignment model of IA and demonstrate its success in describing observational data. We also describe a technique to separate IA from galaxy-galaxy lensing measurements. Applying this technique to luminous red galaxy lenses in the Sloan Digital Sky Survey, we constrain potential IA contamination from associated sources to be below a few percent.