Joerg Colberg - Academia.edu (original) (raw)

Papers by Joerg Colberg

The Astrophysical Journal, 2002

We use giga-particle N-body simulations to study galaxy cluster populations in Hubble volumes of ... more We use giga-particle N-body simulations to study galaxy cluster populations in Hubble volumes of ΛCDM (Ω m = 0.3, Ω Λ = 0.7) and τ CDM (Ω m = 1) world models. Mapping past light-cones of locations in the computational space, we create mock sky surveys of dark matter structure to z ≃ 1.4 over 10, 000 sq deg and to z ≃ 0.5 over two full spheres. Calibrating the Jenkins mass function at z = 0 with samples of ∼ 1.5 million clusters, we show that the fit describes the sky survey counts to ∼ < 20% acccuracy over all redshifts for systems more massive than poor galaxy groups (5 × 10 13 h −1 M ⊙ ).

Monthly Notices of the Royal Astronomical Society, 2002

... Colberg JM, White SDM, MacFarland TJ, Jenkins A., Frenk CS, Pearce FR, Evrard AE, Couchman HM... more ... Colberg JM, White SDM, MacFarland TJ, Jenkins A., Frenk CS, Pearce FR, Evrard AE, Couchman HMP, Efstathiou G., Peacock JA &amp;amp;amp;amp;amp;amp;amp; Thomas PA, 1998, in MellierY., ed., Wide field Surveys in Astronomy. Editions Frontieres, Paris, p. 247; Cole S. &amp;amp;amp;amp;amp;amp;amp; Kaiser N., 1989, MNRAS, 319 ...

Monthly Notices of the Royal Astronomical Society, 2008

Despite a history that dates back at least a quarter of a century studies of voids in the large-s... more Despite a history that dates back at least a quarter of a century studies of voids in the large-scale structure of the Universe are bedevilled by a major problem: there exist a large number of quite different void-finding algorithms, a fact that has so far got in the way of groups comparing their results without worrying about whether such a comparison in fact makes sense. Because of the recent increased interest in voids, both in very large galaxy surveys and in detailed simulations of cosmic structure formation, this situation is very unfortunate. We here present the first systematic comparison study of thirteen different void finders constructed using particles, haloes, and semianalytical model galaxies extracted from a subvolume of the Millennium simulation. The study includes many groups that have studied voids over the past decade. We show their results and discuss their differences and agreements. As it turns out, the basic results of the various methods agree very well with each other in that they all locate a major void near the centre of our volume. Voids have very underdense centres, reaching below 10 percent of the mean cosmic density. In addition, those void finders that allow for void galaxies show that those galaxies follow similar trends. For example, the overdensity of void galaxies brighter than m B = −20 is found to be smaller than about −0.8 by all our void finding algorithms.

Monthly Notices of the Royal Astronomical Society, 2001

We combine data from a number of N-body simulations to predict the abundance of dark halos in Col... more We combine data from a number of N-body simulations to predict the abundance of dark halos in Cold Dark Matter universes over more than 4 orders of magnitude in mass. A comparison of different simulations suggests that the dominant uncertainty in our results is systematic and is smaller than 10-30% at all masses, depending on the halo definition used. In particular, our "Hubble Volume" simulations of τ CDM and ΛCDM cosmologies allow the abundance of massive clusters to be predicted with uncertainties well below those expected in all currently planned observational surveys. We show that for a range of CDM cosmologies and for a suitable halo definition, the simulated mass function is almost independent of epoch, of cosmological parameters, and of initial power spectrum when expressed in appropriate variables. This universality is of exactly the kind predicted by the familiar Press-Schechter model, although this model predicts a mass function shape which differs from our numerical results, overestimating the abundance of "typical" halos and underestimating that of massive systems.

Monthly Notices of the Royal Astronomical Society, 2001

We create mock pencil-beam redshift surveys from very large cosmological N -body simulations of t... more We create mock pencil-beam redshift surveys from very large cosmological N -body simulations of two Cold Dark Matter cosmogonies, an Einstein-de Sitter model (τ CDM) and a flat model with Ω 0 = 0.3 and a cosmological constant (ΛCDM). We use these to assess the significance of the apparent periodicity discovered by . Simulation particles are tagged as 'galaxies' so as to reproduce observed present-day correlations. They are then identified along the past light-cones of hypothetical observers to create mock catalogues with the geometry and the distance distribution of the Broadhurst et al. data. We produce 1936 (2625) quasi-independent catalogues from our τ CDM (ΛCDM) simulation. A couple of large clumps in a catalogue can produce a high peak at low wavenumbers in the corresponding one-dimensional power spectrum, without any apparent large-scale periodicity in the original redshift histogram. Although the simulated redshift histograms frequently display regularly spaced clumps, the spacing of these clumps varies between catalogues and there is no 'preferred' period over our many realisations. We find only a 0.72 (0.49) per cent chance that the highest peak in the power spectrum of a τ CDM (ΛCDM) catalogue has a peak-to-noise ratio higher than that in the Broadhurst et al. data. None of the simulated catalogues with such high peaks shows coherently spaced clumps with a significance as high as that of the real data. We conclude that in CDM universes, the regularity on a scale of ∼ 130h −1 Mpc observed by Broadhurst et al. has a priori probability well below 10 −3 .

The Astrophysical Journal, 1998

We present an analysis of the clustering evolution of dark matter in four cold dark matter (CDM) ... more We present an analysis of the clustering evolution of dark matter in four cold dark matter (CDM) cosmologies. We use a suite of high resolution, 17-million particle, N-body simulations which sample volumes large enough to give clustering statistics with unprecedented accuracy. We investigate a flat model with Ω 0 = 0.3, an open model also with Ω 0 = 0.3, and two models with Ω = 1, one with the standard CDM power spectrum and the other with the same power spectrum as the Ω 0 = 0.3 models. In all cases, the amplitude of primordial fluctuations is set so that the models reproduce the observed abundance of rich galaxy clusters by the present day. We compute mass two-point correlation functions and power spectra over three orders of magnitude in spatial scale and find that in all our simulations they differ significantly from those of the observed galaxy distribution, in both shape and amplitude. Thus, for any of these models to provide an acceptable representation of reality, the distribution of galaxies must be biased relative to the mass in a non-trivial, scale-dependent, fashion. In the Ω = 1 models the required bias is always greater than unity, but in the Ω 0 = 0.3 models an "antibias" is required on scales smaller than ∼ 5h −1 Mpc. The mass correlation functions in the simulations are well fit by recently published analytic models. The velocity fields are remarkably similar in all the models, whether they be characterised as bulk flows, single-particle or pairwise velocity dispersions. This similarity is a direct consequence of our adopted normalisation and runs contrary to the common belief that the amplitude of the observed galaxy velocity fields can be used to constrain the value of Ω 0 . The small-scale pairwise velocity dispersion of the dark matter is somewhat larger than recent determinations from galaxy redshift surveys, but the bulk flows predicted by our models are broadly in agreement with most available data.

The Astrophysical Journal, 2002

We use giga-particle N-body simulations to study galaxy cluster populations in Hubble volumes of ... more We use giga-particle N-body simulations to study galaxy cluster populations in Hubble volumes of ΛCDM (Ω m = 0.3, Ω Λ = 0.7) and τ CDM (Ω m = 1) world models. Mapping past light-cones of locations in the computational space, we create mock sky surveys of dark matter structure to z ≃ 1.4 over 10, 000 sq deg and to z ≃ 0.5 over two full spheres. Calibrating the Jenkins mass function at z = 0 with samples of ∼ 1.5 million clusters, we show that the fit describes the sky survey counts to ∼ < 20% acccuracy over all redshifts for systems more massive than poor galaxy groups (5 × 10 13 h −1 M ⊙ ).

Monthly Notices of the Royal Astronomical Society, 2002

... Colberg JM, White SDM, MacFarland TJ, Jenkins A., Frenk CS, Pearce FR, Evrard AE, Couchman HM... more ... Colberg JM, White SDM, MacFarland TJ, Jenkins A., Frenk CS, Pearce FR, Evrard AE, Couchman HMP, Efstathiou G., Peacock JA &amp;amp;amp;amp;amp;amp;amp; Thomas PA, 1998, in MellierY., ed., Wide field Surveys in Astronomy. Editions Frontieres, Paris, p. 247; Cole S. &amp;amp;amp;amp;amp;amp;amp; Kaiser N., 1989, MNRAS, 319 ...

Monthly Notices of the Royal Astronomical Society, 2008

We investigate the properties of one-dimensional flux "voids" (connected regions in the flux dist... more We investigate the properties of one-dimensional flux "voids" (connected regions in the flux distribution above the mean flux level) by comparing hydrodynamical simulations of large cosmological volumes with a set of observed high-resolution spectra at z ∼ 2. After addressing the effects of box size and resolution, we study how the void distribution changes when the most significant cosmological and astrophysical parameters are varied. We find that the void distribution in the flux is in excellent agreement with predictions of the standard ΛCDM cosmology, which also fits other flux statistics remarkably well. We then model the relation between flux voids and the corresponding one-dimensional gas density field along the line-of-sight and make a preliminary attempt to connect the one-dimensional properties of the gas density field to the three-dimensional dark matter distribution at the same redshift. This provides a framework that allows statistical interpretations of the void population at high redshift using observed quasar spectra, and eventually it will enable linking the void properties of the high-redshift universe with those at lower redshifts, which are better known.

Monthly Notices of the Royal Astronomical Society, 2008

Despite a history that dates back at least a quarter of a century studies of voids in the large-s... more Despite a history that dates back at least a quarter of a century studies of voids in the large-scale structure of the Universe are bedevilled by a major problem: there exist a large number of quite different void-finding algorithms, a fact that has so far got in the way of groups comparing their results without worrying about whether such a comparison in fact makes sense. Because of the recent increased interest in voids, both in very large galaxy surveys and in detailed simulations of cosmic structure formation, this situation is very unfortunate. We here present the first systematic comparison study of thirteen different void finders constructed using particles, haloes, and semianalytical model galaxies extracted from a subvolume of the Millennium simulation. The study includes many groups that have studied voids over the past decade. We show their results and discuss their differences and agreements. As it turns out, the basic results of the various methods agree very well with each other in that they all locate a major void near the centre of our volume. Voids have very underdense centres, reaching below 10 percent of the mean cosmic density. In addition, those void finders that allow for void galaxies show that those galaxies follow similar trends. For example, the overdensity of void galaxies brighter than m B = −20 is found to be smaller than about −0.8 by all our void finding algorithms.

Monthly Notices of the Royal Astronomical Society, 2008

We make use of the first high-resolution hydrodynamic simulations of structure formation which se... more We make use of the first high-resolution hydrodynamic simulations of structure formation which self-consistently follows the build up of supermassive black holes introduced in Di Matteo et al. (2007) to investigate the relation between black holes (BH), host halo and large-scale environment. There are well-defined relations between halo and black hole masses and between the activities of galactic nuclei and halo masses at low redshifts. A large fraction of black holes forms anti-hierarchically, with a higher ratio of black hole to halo mass at high than at low redshifts. At z = 1, we predict group environments (regions of enhanced local density) to contain the highest mass and most active (albeit with a large scatter) BHs while the rest of the BH population to be spread over all densities from groups to filaments and voids. Density dependencies are more pronounced at high rather than low redshift. These results are consistent with the idea that gas rich mergers are likely the main regulator of quasar activity. We find star formation to be a somewhat stronger and tighter function of local density than BH activity, indicating some difference in the triggering of the latter versus the former. There exists a large number of low-mass black holes, growing slowly predominantly through accretion, which extends all the way into the most underdense regions, i.e. in voids.

Monthly Notices of the Royal Astronomical Society, 2000

We investigate the peculiar velocities predicted for galaxy clusters by theories in the cold dark... more We investigate the peculiar velocities predicted for galaxy clusters by theories in the cold dark matter family. A widely used hypothesis identi es rich clusters with high peaks of a suitably smoothed version of the linear density uctuation eld. Their peculiar velocities are then obtained by extrapolating the similarly smoothed linear peculiar velocities at the positions of these peaks. We test these ideas using large high resolution N{body simulations carried out within the Virgo supercomputing consortium. We nd that at early times the barycentre of the material which ends up in a rich cluster is generally very close to a high peak of the initial density eld. Furthermore the mean peculiar velocity of this material agrees well with the linear value at the peak. The late-time growth of peculiar velocities is, however, systematically underestimated by linear theory. At the time clusters are identi ed we nd their rms peculiar velocity to be about 40% larger than predicted. Nonlinear e ects are particularly important in superclusters. These systematics must be borne in mind when using cluster peculiar velocities to estimate the parameter combination 8 0:6 . c 0000 RAS 2 J.M. Colberg et al.

Monthly Notices of the Royal Astronomical Society, 1999

We introduce a new technique for following the formation and evolution of galaxies in cosmologica... more We introduce a new technique for following the formation and evolution of galaxies in cosmological N-body simulations. Dissipationless simulations are used to track the formation and merging of dark matter halos as a function of redshift. Simple prescriptions, taken directly from semi-analytic models of galaxy formation, are adopted for gas cooling, star formation, supernova feedback and the merging of galaxies within the halos. This scheme enables us to explore the clustering properties of galaxies and to investigate how selection by luminosity, colour or type influences the results. In this paper, we study the properties of the galaxy distribution at z = 0. These include B and K-band luminosity functions, two-point correlation functions, pairwise peculiar velocities, cluster mass-to-light ratios, B − V colours and star formation rates. We focus on two variants of a cold dark matter (CDM) cosmology: a high-density (Ω = 1) model with shape-parameter Γ = 0.21 (τ CDM), and a low-density model with Ω = 0.3 and Λ = 0.7 (ΛCDM). Both models are normalized to reproduce the I-band Tully-Fisher relation of near a circular velocity of 220 km s −1 . Our results depend strongly both on this normalization and on the adopted prescriptions for star formation and feedback. Very different assumptions are required to obtain an acceptable model in the two cases. For τ CDM, efficient feedback is required to suppress the growth of galaxies, particularly in low-mass field haloes. Without it, there are too many galaxies and the correlation function exhibits a strong turnover on scales below 1 Mpc. For ΛCDM, feedback must be weaker, otherwise, too few L * galaxies are produced and the correlation function is too steep. Although neither model is perfect both come close to reproducing most of the data. Given the uncertainties in modelling some of the critical physical processes, we conclude that it is not yet possible to draw firm conclusions about the values of cosmological parameters from studies of this kind. Further observational work on global star formation and feedback effects is required to narrow the range of possibilities.

Monthly Notices of the Royal Astronomical Society, 1998

We show that the equations of motion governing the evolution of a collisionless gravitating syste... more We show that the equations of motion governing the evolution of a collisionless gravitating system of particles in an expanding universe can be cast in a form which is almost independent of the cosmological density parameter, Ω, and the cosmological constant, Λ. The new equations are expressed in terms of a time variable τ ≡ lnD, where D is the linear rate of growth of density fluctuations. The weak dependence on the density parameter is proportional to ǫ = Ω −0.2 −1 times the difference between the peculiar velocity (with respect to τ ) of particles and the gravity field (minus the gradient of the potential), or, before shell-crossing, times the sum of the density contrast and the velocity divergence. In a 1-dimensional collapse or expansion, the equations are fully independent of Ω and Λ before shell-crossing. In the general case, the effect of this weak Ω dependence is to enhance the rate of evolution of density perturbations in dense regions. In a flat universe with Λ = 0, this enhancement is less pronounced than in an open universe with Λ = 0 and the same Ω. Using the spherical collapse model, we find that the increase of the rms density fluctuations in a low Ω universe relative to that in a flat universe with the same linear normalization is ∼ 0.01ǫ(Ω) < δ 3 >, where δ is the density field in the flat universe. The equations predict that the smooth average velocity field scales like Ω 0.6 while the local velocity dispersion (rms value) scales, approximately, like Ω 0.5 . High resolution N-body simulations confirm these results and show that density fields, when smoothed on scales slightly larger than clusters, are insensitive to the cosmological model. Halos in an open model simulation are more concentrated than halos of the same M/Ω in a flat model simulation.

Monthly Notices of the Royal Astronomical Society, 2001

We combine data from a number of N-body simulations to predict the abundance of dark halos in Col... more We combine data from a number of N-body simulations to predict the abundance of dark halos in Cold Dark Matter universes over more than 4 orders of magnitude in mass. A comparison of different simulations suggests that the dominant uncertainty in our results is systematic and is smaller than 10-30% at all masses, depending on the halo definition used. In particular, our "Hubble Volume" simulations of τ CDM and ΛCDM cosmologies allow the abundance of massive clusters to be predicted with uncertainties well below those expected in all currently planned observational surveys. We show that for a range of CDM cosmologies and for a suitable halo definition, the simulated mass function is almost independent of epoch, of cosmological parameters, and of initial power spectrum when expressed in appropriate variables. This universality is of exactly the kind predicted by the familiar Press-Schechter model, although this model predicts a mass function shape which differs from our numerical results, overestimating the abundance of "typical" halos and underestimating that of massive systems.

Monthly Notices of the Royal Astronomical Society, 2001

We create mock pencil-beam redshift surveys from very large cosmological N -body simulations of t... more We create mock pencil-beam redshift surveys from very large cosmological N -body simulations of two Cold Dark Matter cosmogonies, an Einstein-de Sitter model (τ CDM) and a flat model with Ω 0 = 0.3 and a cosmological constant (ΛCDM). We use these to assess the significance of the apparent periodicity discovered by . Simulation particles are tagged as 'galaxies' so as to reproduce observed present-day correlations. They are then identified along the past light-cones of hypothetical observers to create mock catalogues with the geometry and the distance distribution of the Broadhurst et al. data. We produce 1936 (2625) quasi-independent catalogues from our τ CDM (ΛCDM) simulation. A couple of large clumps in a catalogue can produce a high peak at low wavenumbers in the corresponding one-dimensional power spectrum, without any apparent large-scale periodicity in the original redshift histogram. Although the simulated redshift histograms frequently display regularly spaced clumps, the spacing of these clumps varies between catalogues and there is no 'preferred' period over our many realisations. We find only a 0.72 (0.49) per cent chance that the highest peak in the power spectrum of a τ CDM (ΛCDM) catalogue has a peak-to-noise ratio higher than that in the Broadhurst et al. data. None of the simulated catalogues with such high peaks shows coherently spaced clumps with a significance as high as that of the real data. We conclude that in CDM universes, the regularity on a scale of ∼ 130h −1 Mpc observed by Broadhurst et al. has a priori probability well below 10 −3 .

We report on analyses of cluster samples obtained from the Hubble Volume Simulations. These simul... more We report on analyses of cluster samples obtained from the Hubble Volume Simulations. These simulations, an Ω = 1 model named τ CDM and a flat low Ω model with a cosmological constant (ΛCDM), comprise the largest computational efforts to date in numerical cosmology. We investigate the presence of massive galaxy clusters at z ≈ 0.8. The τ CDM model fails to form clusters at such a redshift. However, due to the small number of observed clusters around z ≈ 0.8 and the uncertainties in the determinations of their masses, this conclusion still is somewhat preliminary. We produce cluster catalogs at z = 0 for both cosmologies and investigate their two-point correlation function ξ. We show that the relationship between the mean density of subsamples of clusters, expressed via their mean separation dc, and the correlation length r0, defined through ξ(r0) = 1, is not linear but turns over gently for large dc. An analytic prediction by Mo & White [13] overpredicts r0. The results from the analysis of the APM cluster data by Croft et al. [7] are nicely matched by the ΛCDM model.

We are making available on the WWW a selection of the archived data from N-body simulations carri... more We are making available on the WWW a selection of the archived data from N-body simulations carried out by the Virgo consortium and related groups. This currently includes: (i) time-slice, lightcone and cluster data from the two 10910^9109-particle Hubble volume simulations described by Evrard 1998; (ii) time-slice data from simulations of 4 different cold dark matter cosmological models with 2563256^32563

The Astrophysical Journal, 2002

We use giga-particle N-body simulations to study galaxy cluster populations in Hubble volumes of ... more We use giga-particle N-body simulations to study galaxy cluster populations in Hubble volumes of ΛCDM (Ω m = 0.3, Ω Λ = 0.7) and τ CDM (Ω m = 1) world models. Mapping past light-cones of locations in the computational space, we create mock sky surveys of dark matter structure to z ≃ 1.4 over 10, 000 sq deg and to z ≃ 0.5 over two full spheres. Calibrating the Jenkins mass function at z = 0 with samples of ∼ 1.5 million clusters, we show that the fit describes the sky survey counts to ∼ < 20% acccuracy over all redshifts for systems more massive than poor galaxy groups (5 × 10 13 h −1 M ⊙ ).

Monthly Notices of the Royal Astronomical Society, 2002

... Colberg JM, White SDM, MacFarland TJ, Jenkins A., Frenk CS, Pearce FR, Evrard AE, Couchman HM... more ... Colberg JM, White SDM, MacFarland TJ, Jenkins A., Frenk CS, Pearce FR, Evrard AE, Couchman HMP, Efstathiou G., Peacock JA &amp;amp;amp;amp;amp;amp;amp; Thomas PA, 1998, in MellierY., ed., Wide field Surveys in Astronomy. Editions Frontieres, Paris, p. 247; Cole S. &amp;amp;amp;amp;amp;amp;amp; Kaiser N., 1989, MNRAS, 319 ...

Monthly Notices of the Royal Astronomical Society, 2008

Despite a history that dates back at least a quarter of a century studies of voids in the large-s... more Despite a history that dates back at least a quarter of a century studies of voids in the large-scale structure of the Universe are bedevilled by a major problem: there exist a large number of quite different void-finding algorithms, a fact that has so far got in the way of groups comparing their results without worrying about whether such a comparison in fact makes sense. Because of the recent increased interest in voids, both in very large galaxy surveys and in detailed simulations of cosmic structure formation, this situation is very unfortunate. We here present the first systematic comparison study of thirteen different void finders constructed using particles, haloes, and semianalytical model galaxies extracted from a subvolume of the Millennium simulation. The study includes many groups that have studied voids over the past decade. We show their results and discuss their differences and agreements. As it turns out, the basic results of the various methods agree very well with each other in that they all locate a major void near the centre of our volume. Voids have very underdense centres, reaching below 10 percent of the mean cosmic density. In addition, those void finders that allow for void galaxies show that those galaxies follow similar trends. For example, the overdensity of void galaxies brighter than m B = −20 is found to be smaller than about −0.8 by all our void finding algorithms.

Monthly Notices of the Royal Astronomical Society, 2001

We combine data from a number of N-body simulations to predict the abundance of dark halos in Col... more We combine data from a number of N-body simulations to predict the abundance of dark halos in Cold Dark Matter universes over more than 4 orders of magnitude in mass. A comparison of different simulations suggests that the dominant uncertainty in our results is systematic and is smaller than 10-30% at all masses, depending on the halo definition used. In particular, our "Hubble Volume" simulations of τ CDM and ΛCDM cosmologies allow the abundance of massive clusters to be predicted with uncertainties well below those expected in all currently planned observational surveys. We show that for a range of CDM cosmologies and for a suitable halo definition, the simulated mass function is almost independent of epoch, of cosmological parameters, and of initial power spectrum when expressed in appropriate variables. This universality is of exactly the kind predicted by the familiar Press-Schechter model, although this model predicts a mass function shape which differs from our numerical results, overestimating the abundance of "typical" halos and underestimating that of massive systems.

Monthly Notices of the Royal Astronomical Society, 2001

We create mock pencil-beam redshift surveys from very large cosmological N -body simulations of t... more We create mock pencil-beam redshift surveys from very large cosmological N -body simulations of two Cold Dark Matter cosmogonies, an Einstein-de Sitter model (τ CDM) and a flat model with Ω 0 = 0.3 and a cosmological constant (ΛCDM). We use these to assess the significance of the apparent periodicity discovered by . Simulation particles are tagged as 'galaxies' so as to reproduce observed present-day correlations. They are then identified along the past light-cones of hypothetical observers to create mock catalogues with the geometry and the distance distribution of the Broadhurst et al. data. We produce 1936 (2625) quasi-independent catalogues from our τ CDM (ΛCDM) simulation. A couple of large clumps in a catalogue can produce a high peak at low wavenumbers in the corresponding one-dimensional power spectrum, without any apparent large-scale periodicity in the original redshift histogram. Although the simulated redshift histograms frequently display regularly spaced clumps, the spacing of these clumps varies between catalogues and there is no 'preferred' period over our many realisations. We find only a 0.72 (0.49) per cent chance that the highest peak in the power spectrum of a τ CDM (ΛCDM) catalogue has a peak-to-noise ratio higher than that in the Broadhurst et al. data. None of the simulated catalogues with such high peaks shows coherently spaced clumps with a significance as high as that of the real data. We conclude that in CDM universes, the regularity on a scale of ∼ 130h −1 Mpc observed by Broadhurst et al. has a priori probability well below 10 −3 .

The Astrophysical Journal, 1998

We present an analysis of the clustering evolution of dark matter in four cold dark matter (CDM) ... more We present an analysis of the clustering evolution of dark matter in four cold dark matter (CDM) cosmologies. We use a suite of high resolution, 17-million particle, N-body simulations which sample volumes large enough to give clustering statistics with unprecedented accuracy. We investigate a flat model with Ω 0 = 0.3, an open model also with Ω 0 = 0.3, and two models with Ω = 1, one with the standard CDM power spectrum and the other with the same power spectrum as the Ω 0 = 0.3 models. In all cases, the amplitude of primordial fluctuations is set so that the models reproduce the observed abundance of rich galaxy clusters by the present day. We compute mass two-point correlation functions and power spectra over three orders of magnitude in spatial scale and find that in all our simulations they differ significantly from those of the observed galaxy distribution, in both shape and amplitude. Thus, for any of these models to provide an acceptable representation of reality, the distribution of galaxies must be biased relative to the mass in a non-trivial, scale-dependent, fashion. In the Ω = 1 models the required bias is always greater than unity, but in the Ω 0 = 0.3 models an "antibias" is required on scales smaller than ∼ 5h −1 Mpc. The mass correlation functions in the simulations are well fit by recently published analytic models. The velocity fields are remarkably similar in all the models, whether they be characterised as bulk flows, single-particle or pairwise velocity dispersions. This similarity is a direct consequence of our adopted normalisation and runs contrary to the common belief that the amplitude of the observed galaxy velocity fields can be used to constrain the value of Ω 0 . The small-scale pairwise velocity dispersion of the dark matter is somewhat larger than recent determinations from galaxy redshift surveys, but the bulk flows predicted by our models are broadly in agreement with most available data.

The Astrophysical Journal, 2002

We use giga-particle N-body simulations to study galaxy cluster populations in Hubble volumes of ... more We use giga-particle N-body simulations to study galaxy cluster populations in Hubble volumes of ΛCDM (Ω m = 0.3, Ω Λ = 0.7) and τ CDM (Ω m = 1) world models. Mapping past light-cones of locations in the computational space, we create mock sky surveys of dark matter structure to z ≃ 1.4 over 10, 000 sq deg and to z ≃ 0.5 over two full spheres. Calibrating the Jenkins mass function at z = 0 with samples of ∼ 1.5 million clusters, we show that the fit describes the sky survey counts to ∼ < 20% acccuracy over all redshifts for systems more massive than poor galaxy groups (5 × 10 13 h −1 M ⊙ ).

Monthly Notices of the Royal Astronomical Society, 2002

... Colberg JM, White SDM, MacFarland TJ, Jenkins A., Frenk CS, Pearce FR, Evrard AE, Couchman HM... more ... Colberg JM, White SDM, MacFarland TJ, Jenkins A., Frenk CS, Pearce FR, Evrard AE, Couchman HMP, Efstathiou G., Peacock JA &amp;amp;amp;amp;amp;amp;amp; Thomas PA, 1998, in MellierY., ed., Wide field Surveys in Astronomy. Editions Frontieres, Paris, p. 247; Cole S. &amp;amp;amp;amp;amp;amp;amp; Kaiser N., 1989, MNRAS, 319 ...

Monthly Notices of the Royal Astronomical Society, 2008

We investigate the properties of one-dimensional flux "voids" (connected regions in the flux dist... more We investigate the properties of one-dimensional flux "voids" (connected regions in the flux distribution above the mean flux level) by comparing hydrodynamical simulations of large cosmological volumes with a set of observed high-resolution spectra at z ∼ 2. After addressing the effects of box size and resolution, we study how the void distribution changes when the most significant cosmological and astrophysical parameters are varied. We find that the void distribution in the flux is in excellent agreement with predictions of the standard ΛCDM cosmology, which also fits other flux statistics remarkably well. We then model the relation between flux voids and the corresponding one-dimensional gas density field along the line-of-sight and make a preliminary attempt to connect the one-dimensional properties of the gas density field to the three-dimensional dark matter distribution at the same redshift. This provides a framework that allows statistical interpretations of the void population at high redshift using observed quasar spectra, and eventually it will enable linking the void properties of the high-redshift universe with those at lower redshifts, which are better known.

Monthly Notices of the Royal Astronomical Society, 2008

Despite a history that dates back at least a quarter of a century studies of voids in the large-s... more Despite a history that dates back at least a quarter of a century studies of voids in the large-scale structure of the Universe are bedevilled by a major problem: there exist a large number of quite different void-finding algorithms, a fact that has so far got in the way of groups comparing their results without worrying about whether such a comparison in fact makes sense. Because of the recent increased interest in voids, both in very large galaxy surveys and in detailed simulations of cosmic structure formation, this situation is very unfortunate. We here present the first systematic comparison study of thirteen different void finders constructed using particles, haloes, and semianalytical model galaxies extracted from a subvolume of the Millennium simulation. The study includes many groups that have studied voids over the past decade. We show their results and discuss their differences and agreements. As it turns out, the basic results of the various methods agree very well with each other in that they all locate a major void near the centre of our volume. Voids have very underdense centres, reaching below 10 percent of the mean cosmic density. In addition, those void finders that allow for void galaxies show that those galaxies follow similar trends. For example, the overdensity of void galaxies brighter than m B = −20 is found to be smaller than about −0.8 by all our void finding algorithms.

Monthly Notices of the Royal Astronomical Society, 2008

We make use of the first high-resolution hydrodynamic simulations of structure formation which se... more We make use of the first high-resolution hydrodynamic simulations of structure formation which self-consistently follows the build up of supermassive black holes introduced in Di Matteo et al. (2007) to investigate the relation between black holes (BH), host halo and large-scale environment. There are well-defined relations between halo and black hole masses and between the activities of galactic nuclei and halo masses at low redshifts. A large fraction of black holes forms anti-hierarchically, with a higher ratio of black hole to halo mass at high than at low redshifts. At z = 1, we predict group environments (regions of enhanced local density) to contain the highest mass and most active (albeit with a large scatter) BHs while the rest of the BH population to be spread over all densities from groups to filaments and voids. Density dependencies are more pronounced at high rather than low redshift. These results are consistent with the idea that gas rich mergers are likely the main regulator of quasar activity. We find star formation to be a somewhat stronger and tighter function of local density than BH activity, indicating some difference in the triggering of the latter versus the former. There exists a large number of low-mass black holes, growing slowly predominantly through accretion, which extends all the way into the most underdense regions, i.e. in voids.

Monthly Notices of the Royal Astronomical Society, 2000

We investigate the peculiar velocities predicted for galaxy clusters by theories in the cold dark... more We investigate the peculiar velocities predicted for galaxy clusters by theories in the cold dark matter family. A widely used hypothesis identi es rich clusters with high peaks of a suitably smoothed version of the linear density uctuation eld. Their peculiar velocities are then obtained by extrapolating the similarly smoothed linear peculiar velocities at the positions of these peaks. We test these ideas using large high resolution N{body simulations carried out within the Virgo supercomputing consortium. We nd that at early times the barycentre of the material which ends up in a rich cluster is generally very close to a high peak of the initial density eld. Furthermore the mean peculiar velocity of this material agrees well with the linear value at the peak. The late-time growth of peculiar velocities is, however, systematically underestimated by linear theory. At the time clusters are identi ed we nd their rms peculiar velocity to be about 40% larger than predicted. Nonlinear e ects are particularly important in superclusters. These systematics must be borne in mind when using cluster peculiar velocities to estimate the parameter combination 8 0:6 . c 0000 RAS 2 J.M. Colberg et al.

Monthly Notices of the Royal Astronomical Society, 1999

We introduce a new technique for following the formation and evolution of galaxies in cosmologica... more We introduce a new technique for following the formation and evolution of galaxies in cosmological N-body simulations. Dissipationless simulations are used to track the formation and merging of dark matter halos as a function of redshift. Simple prescriptions, taken directly from semi-analytic models of galaxy formation, are adopted for gas cooling, star formation, supernova feedback and the merging of galaxies within the halos. This scheme enables us to explore the clustering properties of galaxies and to investigate how selection by luminosity, colour or type influences the results. In this paper, we study the properties of the galaxy distribution at z = 0. These include B and K-band luminosity functions, two-point correlation functions, pairwise peculiar velocities, cluster mass-to-light ratios, B − V colours and star formation rates. We focus on two variants of a cold dark matter (CDM) cosmology: a high-density (Ω = 1) model with shape-parameter Γ = 0.21 (τ CDM), and a low-density model with Ω = 0.3 and Λ = 0.7 (ΛCDM). Both models are normalized to reproduce the I-band Tully-Fisher relation of near a circular velocity of 220 km s −1 . Our results depend strongly both on this normalization and on the adopted prescriptions for star formation and feedback. Very different assumptions are required to obtain an acceptable model in the two cases. For τ CDM, efficient feedback is required to suppress the growth of galaxies, particularly in low-mass field haloes. Without it, there are too many galaxies and the correlation function exhibits a strong turnover on scales below 1 Mpc. For ΛCDM, feedback must be weaker, otherwise, too few L * galaxies are produced and the correlation function is too steep. Although neither model is perfect both come close to reproducing most of the data. Given the uncertainties in modelling some of the critical physical processes, we conclude that it is not yet possible to draw firm conclusions about the values of cosmological parameters from studies of this kind. Further observational work on global star formation and feedback effects is required to narrow the range of possibilities.

Monthly Notices of the Royal Astronomical Society, 1998

We show that the equations of motion governing the evolution of a collisionless gravitating syste... more We show that the equations of motion governing the evolution of a collisionless gravitating system of particles in an expanding universe can be cast in a form which is almost independent of the cosmological density parameter, Ω, and the cosmological constant, Λ. The new equations are expressed in terms of a time variable τ ≡ lnD, where D is the linear rate of growth of density fluctuations. The weak dependence on the density parameter is proportional to ǫ = Ω −0.2 −1 times the difference between the peculiar velocity (with respect to τ ) of particles and the gravity field (minus the gradient of the potential), or, before shell-crossing, times the sum of the density contrast and the velocity divergence. In a 1-dimensional collapse or expansion, the equations are fully independent of Ω and Λ before shell-crossing. In the general case, the effect of this weak Ω dependence is to enhance the rate of evolution of density perturbations in dense regions. In a flat universe with Λ = 0, this enhancement is less pronounced than in an open universe with Λ = 0 and the same Ω. Using the spherical collapse model, we find that the increase of the rms density fluctuations in a low Ω universe relative to that in a flat universe with the same linear normalization is ∼ 0.01ǫ(Ω) < δ 3 >, where δ is the density field in the flat universe. The equations predict that the smooth average velocity field scales like Ω 0.6 while the local velocity dispersion (rms value) scales, approximately, like Ω 0.5 . High resolution N-body simulations confirm these results and show that density fields, when smoothed on scales slightly larger than clusters, are insensitive to the cosmological model. Halos in an open model simulation are more concentrated than halos of the same M/Ω in a flat model simulation.

Monthly Notices of the Royal Astronomical Society, 2001

We combine data from a number of N-body simulations to predict the abundance of dark halos in Col... more We combine data from a number of N-body simulations to predict the abundance of dark halos in Cold Dark Matter universes over more than 4 orders of magnitude in mass. A comparison of different simulations suggests that the dominant uncertainty in our results is systematic and is smaller than 10-30% at all masses, depending on the halo definition used. In particular, our "Hubble Volume" simulations of τ CDM and ΛCDM cosmologies allow the abundance of massive clusters to be predicted with uncertainties well below those expected in all currently planned observational surveys. We show that for a range of CDM cosmologies and for a suitable halo definition, the simulated mass function is almost independent of epoch, of cosmological parameters, and of initial power spectrum when expressed in appropriate variables. This universality is of exactly the kind predicted by the familiar Press-Schechter model, although this model predicts a mass function shape which differs from our numerical results, overestimating the abundance of "typical" halos and underestimating that of massive systems.

Monthly Notices of the Royal Astronomical Society, 2001

We create mock pencil-beam redshift surveys from very large cosmological N -body simulations of t... more We create mock pencil-beam redshift surveys from very large cosmological N -body simulations of two Cold Dark Matter cosmogonies, an Einstein-de Sitter model (τ CDM) and a flat model with Ω 0 = 0.3 and a cosmological constant (ΛCDM). We use these to assess the significance of the apparent periodicity discovered by . Simulation particles are tagged as 'galaxies' so as to reproduce observed present-day correlations. They are then identified along the past light-cones of hypothetical observers to create mock catalogues with the geometry and the distance distribution of the Broadhurst et al. data. We produce 1936 (2625) quasi-independent catalogues from our τ CDM (ΛCDM) simulation. A couple of large clumps in a catalogue can produce a high peak at low wavenumbers in the corresponding one-dimensional power spectrum, without any apparent large-scale periodicity in the original redshift histogram. Although the simulated redshift histograms frequently display regularly spaced clumps, the spacing of these clumps varies between catalogues and there is no 'preferred' period over our many realisations. We find only a 0.72 (0.49) per cent chance that the highest peak in the power spectrum of a τ CDM (ΛCDM) catalogue has a peak-to-noise ratio higher than that in the Broadhurst et al. data. None of the simulated catalogues with such high peaks shows coherently spaced clumps with a significance as high as that of the real data. We conclude that in CDM universes, the regularity on a scale of ∼ 130h −1 Mpc observed by Broadhurst et al. has a priori probability well below 10 −3 .

We report on analyses of cluster samples obtained from the Hubble Volume Simulations. These simul... more We report on analyses of cluster samples obtained from the Hubble Volume Simulations. These simulations, an Ω = 1 model named τ CDM and a flat low Ω model with a cosmological constant (ΛCDM), comprise the largest computational efforts to date in numerical cosmology. We investigate the presence of massive galaxy clusters at z ≈ 0.8. The τ CDM model fails to form clusters at such a redshift. However, due to the small number of observed clusters around z ≈ 0.8 and the uncertainties in the determinations of their masses, this conclusion still is somewhat preliminary. We produce cluster catalogs at z = 0 for both cosmologies and investigate their two-point correlation function ξ. We show that the relationship between the mean density of subsamples of clusters, expressed via their mean separation dc, and the correlation length r0, defined through ξ(r0) = 1, is not linear but turns over gently for large dc. An analytic prediction by Mo & White [13] overpredicts r0. The results from the analysis of the APM cluster data by Croft et al. [7] are nicely matched by the ΛCDM model.

We are making available on the WWW a selection of the archived data from N-body simulations carri... more We are making available on the WWW a selection of the archived data from N-body simulations carried out by the Virgo consortium and related groups. This currently includes: (i) time-slice, lightcone and cluster data from the two 10910^9109-particle Hubble volume simulations described by Evrard 1998; (ii) time-slice data from simulations of 4 different cold dark matter cosmological models with 2563256^32563