Marcel Zemp

Clumps and streams in the local dark matter distribution

In cold dark matter cosmological models, structures form and grow through the merging of smaller units. Numerical simulations have shown that such merging is incomplete; the inner cores of haloes survive and orbit as ‘subhaloes’ within their hosts. Here we report a simulation that resolves such substructure even in the very inner regions of the Galactic halo. We find hundreds of very concentrated dark matter clumps surviving near the solar circle, as well as numerous cold streams. The simulation also reveals the fractal nature of dark matter clustering: isolated haloes and subhaloes contain the same relative amount of substructure and both have cusped inner density profiles. The inner mass and phase-space densities of subhaloes match those of recently discovered faint, dark-matter-dominated dwarf satellite galaxies, and the overall amount of substructure can explain the anomalous flux ratios seen in strong gravitational lenses. Subhaloes boost γ-ray production from dark matter annihilation by factors of 4 to 15 relative to smooth galactic models. Local cosmic ray production is also enhanced, typically by a factor of 1.4 but by a factor of more than 10 in one per cent of locations lying sufficiently close to a large subhalo. (These estimates assume that the gravitational effects of baryons on dark matter substructure are small.)

Globular clusters, satellite galaxies and stellar haloes from early dark matter peaks

The Milky Way contains several distinct old stellar components that provide a fossil record of its formation. We can understand their spatial distribution and kinematics in a hierarchical formation scenario by associating the proto-galactic fragments envisaged by Searle and Zinn (1978) with the rare peaks able to cool gas in the cold dark matter density field collapsing at redshift z > 10. We use hierarchical structure formation simulations to explore the kinematics and spatial distribution of these early star-forming structures in galaxy haloes today. Most of the proto-galaxies rapidly merge, their stellar contents and dark matter becoming smoothly distributed and forming the inner Galactic halo. The metal-poor globular clusters and old halo stars become tracers of this early evolutionary phase, centrally biased and naturally reproducing the observed steep fall off with radius. The most outlying peaks fall in late and survive to the present day as satellite galaxies. The observed radial velocity dispersion profile and the local radial velocity anisotropy of Milky Way halo stars are successfully reproduced in this model. If this epoch of structure formation coincides with a suppression of further cooling into lower sigma peaks then we can reproduce the rarity, kinematics and spatial distribution of satellite galaxies as suggested by Bullock et al. (2000). Reionisation at z = 12±2 provides a natural solution to the missing satellites problem. Measuring the distribution of globular clusters and halo light on scales from galaxies to clusters could be used to constrain global versus local reionisation models. If reionisation occurs contemporary, our model predicts a constant frequency of blue globulars relative to the host halo mass, except for dwarf galaxies where the average relative frequencies become smaller.

Haloes gone MAD: The Halo-Finder Comparison Project

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

Does the Fornax dwarf spheroidal have a central cusp or core

ABSTRACT The dark matter dominated Fornax dwarf spheroidal has five globular clusters orbitingat ∼ 1kpc from its centre. In a cuspy CDM halo the globulars would sink to thecentre from their current positions within a few Gyrs, presenting a puzzle as to whythey survive undigested at the present epoch. We show that a solution to this timingproblem is to adopt a cored dark matter halo. We use numerical simulations andanalytic calculations to show that, under these conditions, the sinking time becomesmany Hubble times; the globulars effectively stall at the dark matter core radius. Weconclude that the Fornax dwarf spheroidal has a shallow inner density profile with acore radius constrained by the observed positions of its globular clusters. If the phasespace density of the core is primordial then it implies a warm dark matter particleand gives an upper limit to its mass of ∼ 0.5keV, consistent with that required tosignificantly alleviate the substructure problem.Keywords: galaxies:starclusters — galaxies:dwarfs— galaxies:individual (Fornax)methods: N-body simulations

Cusps in cold dark matter haloes

We resolve the inner region of a massive cluster forming in a cosmological ΛCDM simulation with a mass resolution of 2 × 106M⊙ and before z=4.4 even 3 × 10 5M⊙. This is a billion times less than the clusters final virial mass and a substantial increase over current ΛCDM simulations. We achieve this resolution using a new multi-mass refinement procedure and are now able to probe a dark matter halo density profile down to 0.1 percent of the virial radius. The inner density profile of this cluster halo is well fitted by a power-law ρ ∝ r down to the smallest resolved scale. An inner region with roughly constant logarithmic slope is now resolved, which suggests that cuspy profiles describe the inner profile better than recently proposed profiles with a core. The cluster studied here is one out of a sample of six high resolution cluster simulations of Diemand et al. (2004b) and its inner slope of about γ = 1.2 lies close to the sample average.

Dark matter direct detection with non-Maxwellian velocity structure

The velocity distribution function of dark matter particles is expected to show significant departures from a Maxwell-Boltzmann distribution. This can have profound effects on the predicted dark matter - nucleon scattering rates in direct detection experiments, especially for dark matter models in which the scattering is sensitive to the high velocity tail of the distribution, such as inelastic dark matter (iDM) or light (few GeV) dark matter (LDM), and for experiments that require high energy recoil events, such as many directionally sensitive experiments. Here we determine the velocity distribution functions from two of the highest resolution numerical simulations of Galactic dark matter structure (Via Lactea II and GHALO), and study the effects for these scenarios. For directional detection, we find that the observed departures from Maxwell-Boltzmann increase the contrast of the signal and change the typical direction of incoming DM particles. For iDM, the expected signals at direct detection experiments are changed dramatically: the annual modulation can be enhanced by more than a factor two, and the relative rates of DAMA compared to CDMS can change by an order of magnitude, while those compared to CRESST can change by a factor of two. The spectrum of the signal can also change dramatically, with many features arising due to substructure. For LDM the spectral effects are smaller, but changes do arise that improve the compatibility with existing experiments. We find that the phase of the modulation can depend upon energy, which would help discriminate against background should it be found.

The graininess of dark matter haloes

We use the recently completed one billion particle Via Lactea II A cold dark matter simulation to investigate local properties like density, mean velocity, velocity dispersion, anisotropy, orientation and shape of the velocity dispersion ellipsoid, as well as the structure in velocity space of dark matter haloes. We show that at the same radial distance from the halo centre, these properties can deviate by orders of magnitude from the canonical, spherically averaged values, a variation that can only be partly explained by triaxiality and the presence of subhaloes. The mass density appears smooth in the central relaxed regions but spans four orders of magnitude in the outskirts, both because of the presence of subhaloes as well as of underdense regions and holes in the matter distribution. In the inner regions, the local velocity dispersion ellipsoid is aligned with the shape ellipsoid of the halo. This is not true in the outer parts where the orientation becomes more isotropic. The clumpy structure in local velocity space of the outer halo cannot be well described by a smooth multivariate normal distribution. Via Lactea II also shows the presence of cold streams made visible by their high 6D phase space density. Generally, the structure of dark matter haloes shows a high degree of graininess in phase space that cannot be described by a smooth distribution function.

ON DETERMINING THE SHAPE OF MATTER DISTRIBUTIONS

A basic property of objects, such as galaxies and halos that form in cosmological structure formation simulations, is their shape. Here, we critically investigate shape determination methods that are commonly used in the literature. It is found that using an enclosed integration volume and weight factors r −2 and r −2 ell (elliptical radius) for the contribution of each particle or volume element in the shape tensor leads to biased axis ratios and smoothing of details when calculating the local shape as a function of distance from the center. To determine the local shape of matter distributions as a function of distance for well-resolved objects (typically more than O(10 4 ) particles), we advocate a method that (1) uses an ellipsoidal shell (homoeoid) as an integration volume without any weight factors in the shape tensor and (2) removes subhalos.

Fossil Remnants of Reionization in the Halo of the Milky Way

Our recently completed one billion particle Via Lactea II simulation of a Milky Way-sized dark matter halo resolves over 50,000 gravitationally bound clumps orbiting today within the virialized region of the main host. About 2300 of these subhalos have one or more progenitors with -->M > 106 M? at redshift -->z = 11, i.e., massive enough for their gas to have cooled via excitation of H2 and fragmented prior to the epoch of cosmic reionization. We count 4500 such progenitors: if these were able to convert a fraction of their gas content into very metal-poor stars with a Salpeter initial mass function (IMF), they would be shining today with a visual magnitude -->MV = 6.7 per solar mass in stars. Assuming a universal baryon fraction, we show that mean star formation efficiencies as low as 0.1% in progenitors 108 M? would overproduce the abundance of the faint Galactic dwarf spheroidals observed by the Sloan Digital Sky Survey. Star formation at first light must either have occurred with an IMF lacking stars below 0.9 M?, or was intrinsically very inefficient in small dark matter halos. If the latter, our results may be viewed as another hint that there is a minimum scale in galaxy formation.

A universal velocity distribution of relaxed collisionless structures

Several general trends have been identified for equilibrated, self-gravitating collisionless systems, such as density or anisotropy profiles. These are integrated quantities which naturally depend on the underlying velocity distribution function (VDF) of the system. We study this VDF through a set of numerical simulations, which allow us to extract both the radial and the tangential VDFs. We find that the shape of the VDF is universal, in the sense that it depends only on two things: namely the dispersion (radial or tangential) and the local slope of the density. Both the radial and the tangential VDFs are universal for a collection of simulations, including controlled collisions with very different initial conditions, radial infall simulation, and structures formed in cosmological simulations.