Abstract
We present a new power spectrum emulator named EuclidEmulator that estimates the nonlinear correction to the linear dark matter power spectrum depending on the six cosmological parameters ...ωb, ωm, ns, h, $w$0, and σ8. It is constructed using the uncertainty quantification software UQLab using a spectral decomposition method called polynomial chaos expansion. All steps in its construction have been tested and optimized: the large high-resolution N-body simulations carried out with PKDGRAV3 were validated using a simulation from the Euclid Flagship campaign and demonstrated to have converged up to wavenumbers $k\approx 5\, h\, {\rm Mpc}^{-1}$ for redshifts $z$ ≤ 5. The emulator is based on 100 input cosmologies simulated in boxes of (1250 Mpc/h)3 using 20483 particles. We show that by creating mock emulators it is possible to successfully predict and optimize the performance of the final emulator prior to performing any N-body simulations. The absolute accuracy of the final nonlinear power spectrum is as good as one obtained with N-body simulations, conservatively, ${\sim } 1$ per cent for $k\lesssim 1\, h\, {\rm Mpc}^{-1}$ and $z$ ≲ 1. This enables efficient forward modelling in the nonlinear regime, allowing for estimation of cosmological parameters using Markov Chain Monte Carlo methods. EuclidEmulator has been compared to HALOFIT, CosmicEmu, and NGenHalofit, and shown to be more accurate than these other approaches. This work paves a new way for optimal construction of future emulators that also consider other cosmological observables, use higher resolution input simulations, and investigate higher dimensional cosmological parameter spaces.
ABSTRACT We present initial results from the "Ponos" zoom-in numerical simulations of dark matter substructures in massive ellipticals. Two very highly resolved dark matter halos with Mvir = 1.2 × ...1013 and Mvir = 6.5 × 1012 and different ("violent" versus "quiescent") assembly histories have been simulated down to z = 0 in a ΛCDM cosmology with a total of 921,651,914 and 408,377,544 particles, respectively. Within the virial radius, the total mass fraction in self-bound Msub > 106 subhalos at the present epoch is 15% for the violent host and 16.5% for the quiescent one. At z = 0.7, these fractions increase to 19% and 33%, respectively, as more recently accreted satellites are less prone to tidal destruction. In projection, the average fraction of surface mass density in substructure at a distance of R/Rvir = 0.02 (∼5-10 kpc) from the two halo centers ranges from 0.6% to 2%, significantly higher than that measured in simulations of Milky Way-sized halos. The contribution of subhalos with Msub < 109 to the projected mass fraction is between one-fifth and one-third of the total, with the smallest share found in the quiescent host. We assess the impact of baryonic effects via twin, lower-resolution hydrodynamical simulations that include metallicity-dependent gas cooling, star formation, and a delayed-radiative-cooling scheme for supernova feedback. Baryonic contraction produces a super-isothermal total density profile and increases the number of massive subhalos in the inner regions of the main host. The host density profiles and projected subhalo mass fractions appear to be broadly consistent with observations of gravitational lenses.
We use a series of cosmological N-body simulations for a flat Λ cold dark matter (ΛCDM) cosmology to investigate the structural properties of dark matter haloes, at redshift zero, in the mass range 3 ...× 109 h−1 ≲Mvir ≲ 3 × 1013 h−1 M⊙. These properties include the concentration parameter, c, the spin parameter, λ, and the mean axis ratio, . For the concentration–mass relation we find c∝ in agreement with the model proposed by Bullock et al., but inconsistent with the alternative model of Eke et al. The normalization of the concentration–mass relation, however, is 15 per cent lower than suggested by Bullock et al. The results for λ and are in good agreement with previous studies, when extrapolated to the lower halo masses probed here, while c and λ are anticorrelated, in that high-spin haloes have, on average, lower concentrations. In an attempt to remove unrelaxed haloes from the sample, we compute for each halo the offset parameter, xoff, defined as the distance between the most bound particle and the centre of mass, in units of the virial radius. Removing haloes with large xoff increases the mean concentration by ∼10 per cent, lowers the mean spin parameter by ∼15 per cent, and removes the most prolate haloes. In addition, it largely removes the anticorrelation between c and λ, though not entirely. We also investigate the relation between halo properties and their large-scale environment density. For low-mass haloes we find that more concentrated haloes live in denser environments than their less concentrated counterparts of the same mass, consistent with recent correlation function analyses. Note, however, that the trend is weak compared to the scatter. For the halo spin parameters we find no environment dependence, while there is a weak indication that the most spherical haloes reside in slightly denser environments. Finally, using a simple model for disc galaxy formation we show that haloes that host low surface brightness galaxies are expected to be hosted by a biased subset of haloes. Not only do these haloes have spin parameters that are larger than average, they also have concentration parameters that are ∼15 per cent lower than the average at a given halo mass. We discuss the implications of all these findings for the claimed disagreement between halo concentrations inferred from low surface brightness rotation curves, and those expected for a ΛCDM cosmology.
We have carried out a comparison study of hydrodynamical codes by investigating their performance in modelling interacting multiphase fluids. The two commonly used techniques of grid and smoothed ...particle hydrodynamics (SPH) show striking differences in their ability to model processes that are fundamentally important across many areas of astrophysics. Whilst Eulerian grid based methods are able to resolve and treat important dynamical instabilities, such as Kelvin–Helmholtz or Rayleigh–Taylor, these processes are poorly or not at all resolved by existing SPH techniques. We show that the reason for this is that SPH, at least in its standard implementation, introduces spurious pressure forces on particles in regions where there are steep density gradients. This results in a boundary gap of the size of an SPH smoothing kernel radius over which interactions are severely damped.
Haloes gone MAD: The Halo-Finder Comparison Project Knebe, Alexander; Knollmann, Steffen R.; Muldrew, Stuart I. ...
Monthly Notices of the Royal Astronomical Society,
08/2011, Volume:
415, Issue:
3
Journal Article
Peer reviewed
Open access
ABSTRACT
We present a detailed comparison of fundamental dark matter halo properties retrieved by a substantial number of different halo finders. These codes span a wide range of techniques including ...friends‐of‐friends, spherical‐overdensity and phase‐space‐based algorithms. We further introduce a robust (and publicly available) suite of test scenarios that allow halo finder developers to compare the performance of their codes against those presented here. This set includes mock haloes containing various levels and distributions of substructure at a range of resolutions as well as a cosmological simulation of the large‐scale structure of the universe.
All the halo‐finding codes tested could successfully recover the spatial location of our mock haloes. They further returned lists of particles (potentially) belonging to the object that led to coinciding values for the maximum of the circular velocity profile and the radius where it is reached. All the finders based in configuration space struggled to recover substructure that was located close to the centre of the host halo, and the radial dependence of the mass recovered varies from finder to finder. Those finders based in phase space could resolve central substructure although they found difficulties in accurately recovering its properties. Through a resolution study we found that most of the finders could not reliably recover substructure containing fewer than 30–40 particles. However, also here the phase‐space finders excelled by resolving substructure down to 10–20 particles. By comparing the halo finders using a high‐resolution cosmological volume, we found that they agree remarkably well on fundamental properties of astrophysical significance (e.g. mass, position, velocity and peak of the rotation curve).
We further suggest to utilize the peak of the rotation curve, vmax, as a proxy for mass, given the arbitrariness in defining a proper halo edge.
Merging haloes with similar masses (i.e. major mergers) pose significant challenges for halo finders. We compare five halo-finding algorithms’ (ahf, hbt, rockstar, subfind, and velociraptor) recovery ...of halo properties for both isolated and cosmological major mergers. We find that halo positions and velocities are often robust, but mass biases exist for every technique. The algorithms also show strong disagreement in the prevalence and duration of major mergers, especially at high redshifts (z > 1). This raises significant uncertainties for theoretical models that require major mergers for, e.g. galaxy morphology changes, size changes, or black hole growth, as well as for finding Bullet Cluster analogues. All finders not using temporal information also show host halo and subhalo relationship swaps over successive timesteps, requiring careful merger tree construction to avoid problematic mass accretion histories. We suggest that future algorithms should combine phase-space and temporal information to avoid the issues presented.
We explore fundamental properties of the distribution of low-mass dark matter haloes within the cosmic web using warm dark matter (WDM) and cold dark matter (CDM) cosmological simulations. Using ...self-abundance-matched mock galaxy catalogues, we show that the distribution of dwarf galaxies in a WDM universe, wherein low-mass halo formation is heavily suppressed, is nearly indistinguishable to that of a CDM universe whose low-mass haloes are not seen because galaxy formation is suppressed below some threshold halo mass. However, if the scatter between dwarf galaxy luminosity and halo properties is large enough, low-mass CDM haloes would sometimes host relatively bright galaxies thereby populating CDM voids with the occasional isolated galaxy and reducing the numbers of completely empty voids. Otherwise, without high mass to light scatter, all mock galaxy clustering statistics that we consider – the auto-correlation function, the numbers and radial profiles of satellites, the numbers of isolated galaxies, and the probability distribution function of small voids – are nearly identical in CDM and WDM. WDM voids are neither larger nor emptier than CDM voids, when constructed from abundance-matched halo catalogues. It is thus a challenge to determine whether the CDM problem of the overabundance of small haloes with respect to the number density of observed dwarf galaxies has a cosmological solution or an astrophysical solution. However, some clues about the dark matter particle and the scatter between the properties of dwarf galaxies and their dark matter halo hosts might be found in the cosmic web of galaxies in future surveys of the local volume.
Cosmological surveys aim to use the evolution of the abundance of galaxy clusters to accurately constrain the cosmological model. In the context of Λcold dark matter, we show that it is possible to ...achieve the required per cent level accuracy in the halo mass function with gravity-only cosmological simulations, and we provide simulation start and run parameter guidelines for doing so. Some previous works have had sufficient statistical precision, but lacked robust verification of absolute accuracy. Convergence tests of the mass function with, for example, simulation start redshift can exhibit false convergence of the mass function due to counteracting errors, potentially misleading one to infer overly optimistic estimations of simulation accuracy. Per cent level accuracy is possible if initial condition particle mapping uses second-order Lagrangian perturbation theory, and if the start epoch is between 10 and 50 expansion factors before the epoch of halo formation of interest. The mass function for haloes with fewer than ∼1000 particles is highly sensitive to simulation parameters and start redshift, implying a practical minimum mass resolution limit due to mass discreteness. The narrow range in converged start redshift suggests that it is not presently possible for a single simulation to capture accurately the cluster mass function while also starting early enough to model accurately the numbers of reionization era galaxies, whose baryon feedback processes may affect later cluster properties. Ultimately, to fully exploit current and future cosmological surveys will require accurate modelling of baryon physics and observable properties, a formidable challenge for which accurate gravity-only simulations are just an initial step.
Future large-scale structure surveys of the Universe will aim to constrain the cosmological model and the true nature of dark energy with unprecedented accuracy. In order for these surveys to achieve ...their designed goals, they will require predictions for the non-linear matter power spectrum to sub-percent accuracy. Through the use of a large ensemble of cosmological N-body simulations, we demonstrate that if we do not understand the uncertainties associated with simulating structure formation, i.e. knowledge of the 'optimal' simulation parameters, and simply seek to marginalize over them, then the constraining power of such future surveys can be significantly reduced. However, for the parameters {n
s
, h, Ω
b
, Ω
m
}, this effect can be largely mitigated by adding the information from a cosmic microwave background experiment, like Planck. In contrast, for the amplitude of fluctuations σ8 and the time-evolving equation of state of dark energy {w
0, w
a
}, the mitigation is mild. On marginalizing over the simulation parameters, we find that the dark-energy figure of merit can be degraded by ∼2. This is likely an optimistic assessment, since we do not take into account other important simulation parameters. A caveat is our assumption that the Hessian of the likelihood function does not vary significantly when moving from our adopted to the optimal simulation parameter set. This paper therefore provides strong motivation for rigorous convergence testing of N-body codes to meet the future challenges of precision cosmology.
PDF
