Xi Kang

We revisit the Milky Way satellite problem using a semi-analytical model of galaxy formation and compare the predicted luminosity function to recent result from the SDSS. With cosmic photoionization, the luminosity function can be brought into broad agreement with the data between $-15< M_{V} <-2$. This improvement over previous semi-analytical model results (e.g., Benson et al.2002) is from our adoption of improved models for galaxy merger history and galaxy merging time-scales. The very faint satellites ($M_{v} > -5$) formed in halos with virial temperature over $10^{4}K$ (mass around $10^{9} M_{\odot}$ before accretion), but their baryon content are strongly suppressed by photoionization. We model the mass evolution of the subhalos, and compare the predicted mass-to-light ratio with the data. We find that the measured total mass inside the luminous radii of satellites are about 5% of their present total dark matter mass.

Recent studies have shown that the distribution of satellite galaxies is preferentially aligned with the major axis of their central galaxy. The strength of this alignment has been found to depend strongly on the colours of the satellite and central galaxies, and only weakly on the mass of the halo in which the galaxies reside. In this paper we study whether these alignment signals, and their dependence on galaxy and halo properties, can be reproduced in a hierarchical structure formation model of a $\Lambda$CDM concordance cosmology. To that extent we use a large $N$-body simulation which we populate with galaxies following a semi-analytical model for galaxy formation. We find that if the orientation of the central galaxy is perfectly aligned with that of its dark matter halo, then the predicted central-satellite alignment signal is much stronger than observed. If, however, the minor axis of a central galaxy is perfectly aligned with the angular momentum vector of its dark matter halo, we can accurately reproduce the observed alignment strength as function of halo mass and galaxy color. Although this suggests that the orientation of central galaxies is governed by the angular momentum of their dark matter haloes, we emphasize that any other scenario in which the minor axes of central galaxy and halo are misaligned by $\sim 40^{\circ}$ (on average) will match the data equally well. Finally, we show that dependence of the alignment strength on the color of the central galaxy is most likely an artefact due to interlopers in the group catalogue. The dependence on the color of the satellite galaxies, on the other hand, is real and owes to the fact that red satellites are associated with subhaloes that were more massive at their time of accretion.

Using a semi-analytical model for galaxy formation, combined with a large N-body simulation, we investigate the origin of the dichotomy among early-type galaxies. We find that boxy galaxies originate from mergers with a progenitor mass ratio $n < 2$ and with a combined cold gas mass fraction $F_{\rm cold} < 0.1$. Our model accurately reproduces the observed fraction of boxy systems as a function of luminosity and halo mass, for both central galaxies and satellites. After correcting for the stellar mass dependence, the properties of the last major merger of early-type galaxies are independent of their halo mass. This provides theoretical support for the conjecture of Pasquali et al (2007) that the stellar mass of an early-type galaxy is the main parameter that governs its isophotal shape. We argue that the observed dichotomy of early-type galaxies has a natural explanation within hierarchical structure formation, and does not require AGN feedback. Rather, we argue that it owes to the fact that more massive systems (i) have more massive progenitors, (ii) assemble later, and (iii) have a larger fraction of early-type progenitors. Each of these three trends causes the cold gas mass fraction of the progenitors of more massive early-types to be lower, so that their last major merger was dryer. Finally, our model predicts that (i) less than 10 percent of all early-type galaxies form in major mergers that involve two early-type progenitors, (ii) more than 95 percent of all boxy early-type galaxies with $M_* < 2 \times 10^{10} h^{-1} \Msun$ are satellite galaxies, and (iii) about 70 percent of all low mass early-types do not form a supermassive black hole binary at their last major merger. The latter may help to explain why low mass early-types have central cusps, while their massive counterparts have cores.