H. Dejonghe

Simulations of the formation and evolution of isolated dwarf galaxies

We present new fully self-consistent models of the formation and evolution of isolated dwarf galaxies (DGs). We have used the publicly available N-body/smoothed particle hydrodynamics (SPH) code HYDRA, to which we have added a set of star formation criteria, and prescriptions for chemical enrichment [taking into account contributions from both Type Ia supernova (SN Ia) and Type II supernova (SN II)], supernova feedback, and gas cooling. We extensively tested the soundness of these prescriptions and the numerical convergence of the models. The models follow the evolution of an initially homogeneous gas cloud collapsing in a pre-existing dark matter (DM) halo. These simplified initial conditions are supported by the merger trees of isolated DGs extracted from the milli-Millennium Simulation. The star formation histories (SFHs) of the model galaxies exhibit burst-like behaviour. These bursts are a consequence of the blow-out and subsequent in-fall of gas. The amount of gas that leaves the galaxy for good is found to be small, in absolute numbers, ranging between 3 x 10(7) and 6 x 10(7) M(circle dot). For the least massive models, however, this is over 80 per cent of their initial gas mass. The local fluctuations in gas density are strong enough to trigger starbursts in the massive models, or to inhibit anything more than small residual star formation (SF) for the less massive models. Between these starbursts there can be time intervals of several gigayears. nThe models surface brightness profiles are well fitted by Sersic profiles and the correlations between the models Sersic parameters and luminosity agree with the observations. We have also compared model predictions for the half-light radius R(e), central velocity dispersion sigma(c), broad-band colour B - v, metallicity [Z/Z(circle dot)] versus luminosity relations and for the location relative to the fundamental plane with the available data. The properties of the model DGs agree quite well with those of observed DGs. However, the properties of the most massive models deviate from those of observed galaxies. This most likely signals that galaxy mergers are starting to affect the galaxies SFHs in this mass regime (M greater than or similar to 10(9) M(circle dot)). nWe found that a good way to assess the soundness of models is provided by the combination of R(e) and sigma(c). The demand that these are reproduced simultaneously places a stringent constraint on the spatial distribution of SF and on the shape and extent of the DM halo relative to that of the stars.

Full lensing analysis of Abell 1703: comparison of independent lens-modelling techniques

The inner mass-prole of the relaxed cluster Abell 1703 is analysed by two very dierent strong-lensing techniques applied to deep ACS and WFC3 imaging. Our parametric method has the accuracy required to reproduce the many sets of multiple images, based on the assumption that mass approximately traces light. We test this assumption with a fully non-parametric, adaptive grid method, with no knowledge of the galaxy distribution. Dierences between the methods are seen on

Simulations of the formation and evolution of isolated dwarf galaxies – II. Angular momentum as a second parameter

We show results based on a large suite ofu2002N-body/smoothed particle hydrodynamics simulations of isolated, flat dwarf galaxies, both rotating and non-rotating. The main goal is to investigate possible mechanisms to explain the observed dichotomy in radial stellar metallicity profiles of dwarf galaxies: dwarf irregulars (dIrrs) and flat, rotating dwarf ellipticals (dEs) generally possess flat metallicity profiles, while rounder and non-rotating dEs show strong negative metallicity gradients. n n n nThese simulations show that flattening by rotation is key to reproducing the observed characteristics of flat dwarf galaxies, proving particularly efficient in erasing metallicity gradients. We propose a ‘centrifugal barrier mechanism’ as an alternative to the previously suggested ‘fountain mechanism’ for explaining the flat metallicity profiles of dIrrs and flat, rotating dEs. While only flattening the dark matter halo has little influence, the addition of angular momentum slows down the infall of gas, so that star formation (SF) and the ensuing feedback are less centrally concentrated, occurring galaxy-wide. Additionally, this leads to more continuous star formation histories by preventing large-scale oscillations in the star formation rate (‘breathing’), and creates low-density holes in the interstellar medium, in agreement with observations of dIrrs. n n n nOur general conclusion is that rotation has a significant influence on the evolution and appearance of dwarf galaxies, and we suggest angular momentum as au2002second parameteru2002(after galaxy mass as the dominant parameter) in dwarf galaxy evolution. Angular momentum differentiates between SF modes, making our fast rotating models qualitatively resemble dIrrs, which does not seem possible without rotation.

Kelvin-Helmholtz instabilities in Smoothed Particle Hydrodynamics

In this paper we investigate whether smoothed particle hydrodynamics (SPH), equipped with artificial conductivity (AC), is able to capture the physics of density/energy discontinuities in the case of the so-called shearing layers test, a test for examining Kelvin–Helmholtz (KH) instabilities. We can trace back each failure of SPH to show KH rolls to two causes: (i) shock waves travelling in the simulation box and (ii) particle clumping, or more generally, particle noise. The probable cause of shock waves is the local mixing instability, previously identified in the literature. Particle noise on the other hand is a problem because it introduces a large error in the SPH momentum equation. n n n nThe shocks are hard to avoid in SPH simulations with initial density gradients because the most straightforward way of removing them, i.e. relaxing the initial conditions, is not viable. Indeed, by the time sufficient relaxing has taken place the density and energy gradients have become prohibitively wide. The particle disorder introduced by the relaxation is also a problem. We show that setting up initial conditions with a suitably smoothed density gradient dramatically improves results: shock waves are reduced whilst retaining relatively sharp gradients and avoiding unnecessary particle disorder. Particle clumping is easy to overcome, the most straightforward method being the use of a suitable smoothing kernel with non-zero first central derivative. We present results to that effect using a new smoothing kernel: the linear quartic kernel. n n n nWe also investigate the role of AC. Although AC is necessary in the simulations to avoid ‘oily’ features in the gas due to artificial surface tension, we fail to find any relation between using AC and the appearance of seeded KH rolls. Including AC is necessary for the long-term behaviour of the simulation (e.g. to get λ= 1/2, 1 KH rolls). In sensitive hydrodynamical simulations great care is however needed in selecting the AC signal velocity, with the default formulation leading to too much energy diffusion. We present new signal velocities that lead to less diffusion. n n n nThe effects of the shock waves and of particle disorder become less important as the time-scale of the physical problem (for the shearing layers problem: lower density contrast and higher Mach numbers) decreases. At the resolution of current galaxy formation simulations mixing is probably not important. However, mixing could become crucial for next-generation simulations.

Non-parametric strong lens inversion of SDSS J1004+4112

In this article, we study the well-known strong lensing system SDSS J1004+4112. Not only does it host a large-separation lensed quasar with measured time-delay information, but several other lensed galaxies have been identified as well. A previously developed strong lens inversion procedure that is designed to handle a wide variety of constraints is applied to this lensing system and compared to results reported in other works. Without the inclusion of a tentative central image of one of the galaxies as a constraint, we find that the model recovered by the other constraints indeed predicts an image at that location. An inversion which includes the central image provides tighter constraints on the shape of the central part of the mass map. The resulting model also predicts a central image of a second galaxy where indeed an object is visible in the available Advanced Camera for Surveys images. We find masses of 2.5 x 10 13 and 6.1 × 10 13 M⊙) within a radius of 60 and 110 kpc, respectively, confirming the results from other authors. The resulting mass map is compatible with an elliptical generalization of a projected NFW profile, with r s = 58 +21 -13 arcsec and c vir = 3.91 ± 0.74. The orientation of the elliptical NFW profile closely follows the orientation of the central cluster galaxy and the overall distribution of cluster members.

ON THE UNIVERSALITY OF THE GLOBAL DENSITY SLOPE-ANISOTROPY INEQUALITY

Recently, some intriguing results have led to speculations whether the central density slope-velocity dispersion anisotropy inequality (An & Evans) actually holds at all radii for spherical dynamical systems. We extend these studies by providing a complete analysis of the global slope-anisotropy inequality for all spherical systems in which the augmented density is a separable function of radius and potential. We prove that these systems indeed satisfy the global inequality if their central anisotropy is β0 ≤ 1/2. Furthermore, we present several systems with β0>1/2 for which the inequality does not hold, thus demonstrating that the global density slope-anisotropy inequality is not a universal property. This analysis is a significant step toward an understanding of the relation for general spherical systems.

THE DYNAMICAL STRUCTURE OF DARK MATTER HALOS WITH UNIVERSAL PROPERTIES

N-body simulations have unveiled several apparently universal properties of dark matter halos, including a cusped density profile, a power-law pseudo-phase-space density ?/?3 r , and a linear ?-? relation between the density slope and the velocity anisotropy. We present a family of self-consistent phase-space distribution functions (DFs) F(E, L), based on the Dehnen-McLaughlin Jeans models, that incorporate these universal properties very accurately. These DFs, derived using a quadratic programming technique, are analytical, positive, and smooth over the entire phase space and are able to generate four-parameter velocity anisotropy profiles ?(r) with arbitrary asymptotic values ?0 and ??. We discuss the orbital structure of six radially anisotropic systems in detail and argue that, apart from its use for generating initial conditions for N-body studies, our dynamical modeling provides a valuable complementary approach to understand the processes involved in the formation of dark matter halos.

Global m= 1 instabilities and lopsidedness in disc galaxies

Lopsidedness is common in spiral galaxies. Often, there is no obvious external cause, such as an interaction with a nearby galaxy, for such features. Alternatively, the lopsidedness may have an internal cause, such as a dynamical instability. In order to explore this idea, we have developed a computer code that searches for self-consistent perturbations in razor-thin disc galaxies and performed a thorough mode-analysis of a suite of dynamical models for disc galaxies embedded in an inert dark-matter halo with varying amounts of rotation and radial anisotropy. Models with two equal-mass counter-rotating discs and fully rotating models both show growing lopsided modes. For the counter-rotating models, this is the well-known counter-rotating instability, becoming weaker as the net rotation increases. The m = 1 mode of the maximally rotating models, on the other hand, becomes stronger with increasing net rotation. This rotating m = 1 mode is reminiscent of the eccentricity instability in near-Keplerian discs. To unravel the physical origin of these two dierent m = 1 instabilities, we studied the individual stellar orbits in the perturbed potential and found that the presence of the perturbation gives rise to a very rich orbital behaviour. In the linear regime, both instabilities are supported by aligned loop orbits. In the non-linear regime, other orbit families exist that can help support the modes. In terms of density waves, the counterrotating m = 1 mode is due to a purely growing Jeans-type instability. The rotating m = 1 mode, on the other hand, grows as a result of the swing amplier working inside the resonance cavity that extends from the disc center out to the radius where non-rotating waves are stabilized by the model’s outwardly rising Q-prole.