Some insight on the assembly bias. ZOMG!

Together with three colleagues, at the end of 2015 I started a project to understand the physical origin and implications of the assembly bias. Now, the first results are finally out, so it's time to introduce the Zooming On a Mob of Galaxies project (ZOMG).

The name assembly bias denotes the different clustering properties of Dark Matter (DM) haloes of the same mass, depending on their formation history. Despite being known for many years now, a physical motivation of this effect is still missing. Additionally, this effect is typically studied using DM-only simulations, but the intriguing possibility that the assembly bias can have an impact on the baryonic content of haloes has not been studied yet. No worries though, we are here to amend this!

Our suite of high-resolution zoom-in numerical simulations follows the evolution of seven haloes. Four of these have been simulated also with the inclusion of baryons. We use a new definition of assembly time of the halo, which produce a strong assembly bias signal and study the origin and consequences of asembly bias.

The first results appeared on arxiv last October, and focused on the evolution of the DM components. We showed there that the different collapse time of haloes with the same mass is a consequence of their location in the cosmic web. Namely, haloes residing in filament experience a strong tidal field that (on average) drives the particles away of them, starving their accretion and making them old. Conversely, haloes located in knots undergo a constant supply of materil due to the convergin velocity field of the surrounding regions. This result allows us to propose an improvement to the excursion set theory that improve the ability to predict the final halo mass from the initial density field.

Figure 1 (from Borzyszkowski et al. 2017): particle distribution around two haloes of the ZOMG suite. Colors represent peculiar velocity with respect to the halo. Accreting haloes (left) lives in knots where the material flows onto them, while stalled objects (right) live in prominent filaments where the dark matter is mostly receding from the halo.

A second paper was submitted to arxiv at the beginning of this year. There, we investigated the effect of the assembly history on the central galaxy of the haloes simulated including baryons. Long story short, the accretion of baryons is similar to the DM one, so that old haloes remain old, and young haloes remain young, even if we include baryons. But at the galaxy level things are different! The central galaxies are almost insensitive to the assembly history of the halo. In fact, we find disc and irregular galaxies in both young and old haloes. However, the gas starvation in the latter produce a thickening of the stellar disc of their central galaxies, as a consequence of interactions with satellites that are not counter-balanced by the additional gravity coming from the gas (not) in the disc. Similarly, the stellar population of the central galaxy of old haloes is older than the one found in the central object of young hosts.

Figure 2 (from Romano-Diaz et al. 2017): Stellar components of the simulated galaxies. Colors represent the local velocity dispersion. Galaxies hosted by stalled haloes (left) have older and thicker stellar discs than those residing in accreting objects.

A third paper is in the final stage of internal review and will be released soon. In this work we study the impact of the assembly time (and history) on the properties of the subhalo populations. I will post further details as soon as the paper appears on arxiv. Stay tuned!