No Arabic abstract
We examine two extreme models for the build-up of the stellar component of luminous elliptical galaxies. In one case, we assume the build-up of stars is dissipational, with centrally accreted gas radiating away its orbital and thermal energy; the dark matter halo will undergo adiabatic contraction and the central dark matter density profile will steepen. For the second model, we assume the central galaxy is assembled by a series of dissipationless mergers of stellar clumps that have formed far from the nascent galaxy. In order to be accreted, these clumps lose their orbital energy to the dark matter halo via dynamical friction, thereby heating the central dark matter and smoothing the dark matter density cusp. The central dark matter density profiles differ drastically between these models. For the isolated elliptical galaxy, NGC 4494, the central dark matter densities follow the power-laws r^(-0.2) and r^(-1.7) for the dissipational and dissipationless models, respectively. By matching the dissipational and dissipationless models to observations of the stellar component of elliptical galaxies, we examine the relative contributions of dissipational and dissipationless mergers to the formation of elliptical galaxies and look for observational tests that will distinguish between these models. Comparisons to strong lensing brightest cluster galaxies yield median M*/L_B ratios of 2.1+/-0.8 and 5.2+/-1.7 at z=0.39 for the dissipational and dissipationless models, respectively. For NGC 4494, the best-fit dissipational and dissipationless models have M*/L_B=2.97 and 3.96. Comparisons to expected stellar mass-to-light ratios from passive evolution and population syntheses appear to rule out a purely dissipational formation mechanism for the central stellar regions of giant elliptical galaxies.
We present recalibrations of the GALFORM semi-analytical model of galaxy formation in a new N-body simulation with the Planck cosmology. The Planck Millennium simulation uses more than 128 billion particles to resolve the matter distribution in a cube of $800$ Mpc on a side, which contains more than 77 million dark matter haloes with mass greater than $2.12 times 10^{9} h^{-1} {rm M_{odot}}$ at the present day. Only minor changes to a very small number of model parameters are required in the recalibration. We present predictions for the atomic hydrogen content (HI) of dark matter halos, which is a key input into the calculation of the HI intensity mapping signal expected from the large-scale structure of the Universe. We find that the HI mass $-$ halo mass relation displays a clear break at the halo mass above which AGN heating suppresses gas cooling, $approx 3 times 10^{11} h^{-1} M_{rm odot}$. Below this halo mass, the HI content of haloes is dominated by the central galaxy; above this mass it is the combined HI content of satellites that prevails. We find that the HI mass - halo mass relation changes little with redshift up to $z=3$. The bias of HI sources shows a scale dependence that gets more pronounced with increasing redshift.
We select a sample of galaxies from the Sloan Digital Sky Survey Data Release 7 (SDSS-DR7) where galaxies are classified, through visual inspection, as hosting strong bars, weak bars or as unbarred galaxies, and make use of HI mass and kinematic information from the Arecibo Legacy Fast ALFA (ALFALFA) survey catalog, to study the stellar, atomic gas and dark matter content of barred disk galaxies. We find, in agreement with previous studies, that the bar fraction increases with increasing stellar mass. A similar trend is found with total baryonic mass, although the dependence is not as strong as with stellar mass, this due to the contribution of gas. The bar fraction shows a decrease with increasing gas mass fraction. This anticorrelation between the likelihood of a galaxy hosting a bar with the gas richness of the galaxy results from the inhibiting effect the gas has in the formation of bars. We also find that for massive galaxies with stellar masses larger than 10$^{10} M_{odot}$, at fixed stellar mass, the bar fraction decreases with increasing global halo mass (i.e. halo mass measured up to a radius of the order of the HI disk extent).
We use the halo occupation distribution (HOD) framework to characterise the predictions from two independent galaxy formation models for the galactic content of dark matter haloes and its evolution with redshift. Our galaxy samples correspond to a range of fixed number densities defined by stellar mass and span $0 le z le 3$. We find remarkable similarities between the model predictions. Differences arise at low galaxy number densities which are sensitive to the treatment of heating of the hot halo by active galactic nuclei. The evolution of the form of the HOD can be described in a relatively simple way, and we model each HOD parameter using its value at $z=0$ and an additional evolutionary parameter. In particular, we find that the ratio between the characteristic halo masses for hosting central and satellite galaxies can serve as a sensitive diagnostic for galaxy evolution models. Our results can be used to test and develop empirical studies of galaxy evolution and can facilitate the construction of mock galaxy catalogues for future surveys.
Galaxies are the basic structural element of the universe; galaxy formation theory seeks to explain how these structures came to be. I trace some of the foundational ideas in galaxy formation, with emphasis on the need for non-baryonic cold dark matter. Many elements of early theory did not survive contact with observations of low surface brightness galaxies, leading to the need for auxiliary hypotheses like feedback. The failure points often trace to the surprising predictive successes of an alternative to dark matter, the Modified Newtonian Dynamics (MOND). While dark matter models are flexible in accommodating observations, they do not provide the predictive capacity of MOND. If the universe is made of cold dark matter, why does MOND get any predictions right?
This paper presents further results from our spectroscopic study of the globular cluster (GC) system of the group elliptical NGC 3923. From observations made with the GMOS instrument on the Gemini South telescope, an additional 50 GC and Ultra Compact Dwarf (UCD) candidates have been spectroscopically confirmed as members of the NGC 3923 system. When the recessional velocities of these GCs are combined with the 29 GC velocities reported previously, a total sample of 79 GC/UCD velocities is produced. This sample extends to over 6 arcmin (>6 Re sim30 kpc) from the centre of NGC 3923, and is used to study the dynamics of the GC system and the dark matter content of NGC 3923. It is found that the GC system of NGC 3923 displays no appreciable rotation, and that the projected velocity dispersion is constant with radius within the uncertainties. The velocity dispersion profiles of the integrated light and GC system of NGC 3923 are indistinguishable over the region in which they overlap. We find some evidence that the diffuse light and GCs of NGC 3923 have radially biased orbits within sim130. The application of axisymmetric orbit-based models to the GC and integrated light velocity dispersion profiles demonstrates that a significant increase in the mass-to-light ratio (from M/Lv = 8 to 26) at large galactocentric radii is required to explain these observations. We therefore confirm the presence of a dark matter halo in NGC 3923. We find that dark matter comprises 17.5% of the mass within 1 Re, 41.2% within 2 Re, and 75.6% within the radius of our last kinematic tracer at 6.9 Re. The total dynamical mass within this radius is found to be 1.5 x 10^12 solar masses. In common with other studies of large ellipticals, we find that our derived dynamical mass profile is consistently higher than that derived by X-ray observations, by a factor of around 2.