Reconstruction of cosmological initial conditions from galaxy redshift catalogues

We present and test a new method for the reconstruction of cosmological initial conditions from a full-sky galaxy catalogue. This method, called ZTRACE, is based on a self-consistent solution of the growing mode of gravitational instabilities according to the Zel'dovich approximation and higher order in Lagrangian perturbation theory. Given the evolved redshift-space density field, smoothed on some scale, ZTRACE finds, via an iterative procedure, an approximation to the initial density field for any given set of cosmological parameters; real-space densities and peculiar velocities are also reconstructed. The method is tested by applying it to N -body simulations of an Einstein–de Sitter and an open cold dark matter universe. It is shown that errors in the estimate of the density contrast dominate the noise of the reconstruction. As a consequence, the reconstruction of real-space density and peculiar velocity fields using non-linear algorithms is little improved over those based on linear theory. The use of a mass-preserving adaptive smoothing, equivalent to a smoothing in Lagrangian space, allows an unbiased (although noisy) reconstruction of initial conditions, as long as the (linearly extrapolated) density contrast does not exceed unity. The probability distribution function of the initial conditions is recovered to high precision, even for Gaussian smoothing scales of ∼5  h ⁻¹ Mpc, except for the tail at δ ≥1. This result is insensitive to the assumptions of the background cosmology.     

The one-point PDF of the initial conditions of our local Universe from the IRAS PSC redshift catalogue

The algorithm ztrace of Monaco & Efstathiou is applied to the IRAS PSCz catalogue to reconstruct the initial conditions of our local Universe with a resolution down to ~5  h ¹ Mpc. The one-point probability distribution function (PDF) of the reconstructed initial conditions is consistent with the assumptions that: (i) IRAS galaxies trace mass on scales of ~5  h ¹ Mpc and (ii) the statistics of the primordial density fluctuations are Gaussian. We use simulated PSCz catalogues, constructed from N -body simulations with Gaussian initial conditions, to show that local non-linear bias can cause the recovered initial PDF (assuming no bias) to be non-Gaussian. However, for plausible bias models, the distortions of the recovered PDF would be difficult to detect using the volume finely sampled by the PSCz catalogue. So, for Gaussian initial conditions, a range of bias models remain compatible with our PSCz reconstruction results.     

Cosmological implications of the PSCz PDF and its moments

We compare the probability density function (PDF) and its low-order moments (variance and skewness) of the smoothed IRAS Point Source Catalogue Redshift Survey (PSC z ) galaxy density field and of the corresponding simulated PSC z look-alikes, generated from N -body simulations of six different dark matter models: four structure-normalized with     and     , one COBE -normalized, and the old standard cold dark matter model. The galaxy distributions are smoothed with a Gaussian window at three different smoothing scales,     , 10 and 15  h ⁻¹ Mpc. We find that the simulation PSC z look-alike PDFs are sensitive only to the normalization of the power spectrum, probably owing to the shape similarity of the simulated galaxy power spectrum on the relevant scales. We find that the only models that are consistent, at a high significance level, with the observed PSC z PDF are models with a relatively low power spectrum normalization     . From the phenomenologically derived σ ₈–moments relation, fitted from the simulation data, we find that the PSC z moments suggest     .     

Cold dark matter models with a step-like initial power spectrum

We investigate the properties of clusters of galaxies in the ΛCDM models with a step-like initial power spectrum. We examine the mass function, the peculiar velocities and the power spectrum of clusters in models with different values of the density parameter Ω₀, the normalized Hubble constant h and the spectral parameter p that describes the shape of the initial power spectrum. The results are compared with observations. We also investigate the rms bulk velocity in the models, where the properties of clusters are consistent with the observed data. We find that the power spectrum of clusters is in good agreement with the observed power spectrum of the Abell–ACO clusters if the spectral parameter p is in the range p =0.6–0.8. The power spectrum and the rms peculiar velocity of clusters are consistent with observations only if Ω₀<0.4 . The models with Ω₀=0.3 are consistent with the observed properties of clusters if h =0.50–0.63. For h =0.65, we find that Ω₀=0.20–0.27.     

The clustering of barred galaxies in the local Universe

We study the clustering properties of barred galaxies using data from the Sloan Digital Sky Survey. We compute projected redshift-space two-point cross-correlation functions   w _p( r _p)  for a sample of nearly 1000 galaxies for which we have performed detailed structural decompositions using the methods described in Gadotti. The sample includes 286 barred galaxies. The clustering of barred and unbarred galaxies of similar stellar mass is indistinguishable over all the scales probed (∼20 kpc–30 Mpc). This result also holds even if the sample is restricted to bars with bluer   g − i   colours (and hence younger ages). Our result also does not change if we split our sample of barred galaxies according to bar-to-total luminosity ratio, bar boxyness, effective surface brightness, length or the shape of the surface density profile within the bar. There is a hint that red, elliptical bars are more strongly clustered than red and less elliptical bars, on scales  ≳1 Mpc  , although the statistical significance is not high. We conclude that there is no significant evidence that bars are a product of mergers or interactions. We tentatively interpret the stronger clustering of the more elliptical bars as evidence that they are located in older galaxies, which reside in more massive haloes.     

Redshift-space distortions in the PSCz galaxy catalogue

We apply a spherical harmonic analysis to the Point Source Redshift Survey (PSC z ), to compute the real-space galaxy power spectrum and the degree of redshift distortion caused by peculiar velocities. We employ new parameter eigenvector and hierarchical data compression techniques, allowing a much larger number of harmonic modes to be included, and correspondingly smaller error bars. Using 4644 harmonic modes, compressed to 2278, we find that the IRAS redshift-space distortion parameter is     and the amplitude of galaxy clustering on a scale of     is     . Combining these we find the amplitude of mass perturbations is     . While this is compatible with results from the cosmic microwave background (CMB), with a small degree of tilt, it disagrees with the amplitude of matter perturbations estimated from the abundance of clusters by a factor of 2, independent of cosmology. A preliminary model fitting analysis combining the CMB with either the PSC z or cluster abundances shows that the cosmological matter density parameter     , and the IRAS bias parameter     . However, the cluster abundances suggest that     and     , while the PSC z requires     and     . Given the physics of galaxy formation is poorly constrained, we conclude that IRAS galaxies and mass are only partially correlated.     

The limits of bound structures in the accelerating Universe

According to the latest evidence, the Universe is entering an era of exponential expansion, where gravitationally bound structures will get disconnected from each other, forming isolated 'island universes'. In this scenario, we present a theoretical criterion to determine the boundaries of gravitationally bound structures and a physically motivated definition of superclusters as the largest bound structures in the Universe. We use the spherical collapse model self-consistently to obtain an analytical condition for the mean density enclosed by the last bound shell of the structure (2.36 times the critical density in the present Universe, assumed to be flat, with 30 per cent matter and 70 per cent cosmological constant, in agreement with the previous, numerical result of Chiueh & He). N -body simulations extended to the future show that this criterion, applied at the present cosmological epoch, defines a sphere that encloses ≈99.7 per cent of the particles that will remain bound to the structure at least until the scale parameter of the Universe is 100 times its present value. On the other hand, (28 ± 13) per cent of the enclosed particles are in fact not bound, so the enclosed mass overestimates the bound mass, in contrast with the previous, less rigorous criterion of, e.g. Busha and collaborators, which gave a more precise mass estimate. We also verify that the spherical collapse model estimate for the radial infall velocity of a shell enclosing a given mean density gives an accurate prediction for the velocity profile of infalling particles, down to very near the centre of the virialized core.     

Recovering the topology of the initial density fluctuations using the IRAS Point Source Catalogue Redshift Survey

We apply the reconstruction technique of Nusser & Dekel to the recently available Point Source Catalogue Redshift Survey (PSCz) in order to subtract the non-Gaussianities that are expected to develop in the mild non-linear regime of gravitational evolution. We study the evolution of isodensity contours defined using an adaptive smoothing algorithm, in order to minimize the problems derived from the non-commutativity of the smoothing operator and the time-evolution operator. We study the topology of these isodensity contours and concentrate on the evolution of the amplitude drop of the genus compared to a Gaussian field with an identical power spectrum, in order to quantify the level of phase-correlation present in the field. In order to test the method and to quantify the level of statistical uncertainty, we apply the method to a set of mock PSCz catalogues derived from the N -body simulations of two standard cold dark matter (CDM) models, kindly granted to us by the Virgo consortium. We find the method to be reliable in recovering the correct amplitude drops. When applied to PSCz, the level of phase correlations observed is very low on all scales ranging from 5 to 60  h ⁻¹ Mpc, providing support to the theory that structure originated from Gaussian initial conditions.     

Large-scale bias in the Universe — II. Redshift-space bispectrum

The determination of the density parameter Ω₀ from the large-scale distribution of galaxies is one of the major goals of modern cosmology. However, if galaxies are biased tracers of the underlying mass distribution, linear perturbation theory leads to a degeneracy between Ω₀ and the linear bias parameter b , and the density parameter cannot be estimated. In Matarrese, Verde &38; Heavens we developed a method based on second-order perturbation theory to use the bispectrum to lift this degeneracy by measuring the bias parameter in an Ω₀-independent way. The formalism was developed assuming that one has perfect information on the positions of galaxies in three dimensions. In galaxy redshift surveys, the three-dimensional information is imperfect, because of the contaminating effects of peculiar velocities, and the resulting clustering pattern in redshift space is distorted. In this paper we combine second-order perturbation theory with a model for collapsed, virialized structures, to extend the method to redshift space, and demonstrate that the method should be successful in determining with reasonable accuracy the bias parameter from state-of-the-art surveys such as the Anglo-Australian 2 degree Field Survey and the Sloan Digital Sky Survey.     

The PSCz catalogue

We present the catalogue, mask, redshift data and selection function for the PSC z survey of 15 411 IRAS galaxies across 84 per cent of the sky. Most of the IRAS data are taken from the Point Source Catalog, but this has been supplemented and corrected in various ways to improve the completeness and uniformity. We quantify the known imperfections in the catalogue, and we assess the overall uniformity, completeness and data quality. We find that overall the catalogue is complete and uniform to within a few per cent at high latitudes and 10 per cent at low latitudes. Ancillary information, access details, guidelines and caveats for using the catalogue are given.     

Excursion set approach to the clustering of dark matter haloes in Lagrangian space

We present a stochastic approach to the spatial clustering of dark matter haloes in Lagrangian space. Our formalism is based on a local formulation of the 'excursion set' approach by Bond et al., which automatically accounts for the 'cloud-in-cloud' problem in the identification of bound systems. Our method allows us to calculate correlation functions of haloes in Lagrangian space using either a multidimensional Fokker–Planck equation with suitable boundary conditions, or an array of Langevin equations with spatially correlated random forces. We compare the results of our method with theoretical predictions for the halo autocorrelation function considered in the literature, and find good agreement with the results recently obtained within a treatment of halo clustering in terms of 'counting fields' by Catelan et al. Finally, the possible effect of spatial correlations on numerical simulations of halo merger trees is discussed.     

PSCz superclusters: detection, shapes and cosmological implications

We study the possibility of correctly identifying, from the smooth galaxy density field of the PSC z flux-limited catalogue, high-density regions (superclusters) and recovering their true shapes in the presence of a bias introduced by the coupling between the selection function and the constant radius smoothing. We quantify such systematic biases in the smoothed PSC z density field and after applying the necessary corrections we study supercluster multiplicity and morphologies using a differential geometry definition of shape. Our results strongly suggest that filamentary morphology is the dominant feature of PSC z superclusters. Finally, we compare our results with those expected in three different cosmological models and find that the Λ cold dark matter (CDM) model (Ω_Λ=1−Ω_m=0.7) performs better than Ω_m=1 CDM models.     

Power-spectrum normalization from the local abundance of rich clusters of galaxies

The number density of rich galaxy clusters still provides the most robust way of normalizing the power spectrum of dark matter perturbations on scales relevant to large-scale structure. We revisit this constraint in the light of several recent developments: (1) the availability of well-defined samples of local clusters with relatively accurate X-ray temperatures; (2) new theoretical mass functions for dark matter haloes, which provide a good fit to large numerical simulations; (3) more accurate mass–temperature relations from larger catalogues of hydrodynamical simulations; (4) the requirement to consider closed as well as open and flat cosmologies to obtain full multiparameter likelihood constraints for CMB and SNe studies. We present a new sample of clusters drawn from the literature and use this sample to obtain improved results on σ ₈, the normalization of the matter power spectrum on scales of 8  h ⁻¹ Mpc, as a function of the matter density and cosmological constant in a universe with general curvature. We discuss our differences with previous work, and the remaining major sources of uncertainty. Final results on the normalization, approximately independent of power spectrum shape, can be expressed as constraints on σ at an appropriate cluster normalization scale R _Cl. We provide fitting formulas for R _Cl and σ ( R _Cl) for general cosmologies, as well as for σ ₈ as a function of cosmology and shape parameter Γ. For flat models we find approximately σ ₈≃(0.495_−0.037^+0.034)Ω_M^−0.60 for Γ=0.23, where the error bar is dominated by uncertainty in the mass–temperature relation.     

The cluster distribution as a test of dark matter models — IV. Topology and geometry

We study the geometry and topology of the large-scale structure traced by galaxy clusters in numerical simulations of a box of side 320 h ⁻¹ Mpc, and compare them with available data on real clusters. The simulations we use are generated by the Zel'dovich approximation, using the same methods as we have used in the first three papers in this series. We consider the following models to see if there are measurable differences in the topology and geometry of the superclustering they produce: (i) the standard cold dark matter model (SCDM); (ii) a CDM model with Ω₀ = 0.2 (OCDM); (iii) a CDM model with a 'tilted' power spectrum having n  = 0.7 (TCDM); (iv) a CDM model with a very low Hubble constant, h  = 0.3 (LOWH); (v) a model with mixed CDM and HDM (CHDM); (vi) a flat low-density CDM model with Ω₀ = 0.2 and a non-zero cosmological Λ term (ΛCDM). We analyse these models using a variety of statistical tests based on the analysis of: (i) the Euler–Poincaré characteristic; (ii) percolation properties; (iii) the minimal spanning tree construction. Taking all these tests together we find that the best-fitting model is ΛCDM and, indeed, the others do not appear to be consistent with the data. Our results demonstrate that despite their biased and extremely sparse sampling of the cosmological density field, it is possible to use clusters to probe subtle statistical diagnostics of models, which go far beyond the low-order correlation functions usually applied to study superclustering.