The little things matter: relating the abundance of ultrafaint satellites to the hosts’ assembly history

Abstract

Ultrafaint dwarf galaxies ($M_\star \le 10^{5}\, {\rm M}_\odot$) are relics of an early phase of galaxy formation. They contain some of the oldest and most metal-poor stars in the Universe which likely formed before the epoch of hydrogen reionization. These galaxies are so faint that they can only be detected as satellites of the Milky Way. They are so small that they are not resolved in current cosmological hydrodynamic simulations. Here, we combine very high-resolution cosmological N-body simulations with a semi-analytic model of galaxy formation to study the demographics and spatial distribution of ultrafaint satellites in Milky Way-mass haloes. We show that the abundance of these galaxies is correlated with the assembly history of the host halo: at fixed mass, haloes assembled earlier contain, on average, more ultrafaint satellites today than haloes assembled later. We identify simulated galactic haloes that experience an ancient Gaia-Enceladus-Sausage-like and a recent LMC-like accretion event and find that the former occurs in 33 per cent of the sample and the latter in 9 per cent. Only 3 per cent experience both events and these are especially rich in ultrafaint satellites, most acquired during the ancient accretion event. Our models predict that the radial distribution of satellites is more centrally concentrated in early-forming haloes. Accounting for the depletion of satellites by tidal interactions with the central disc, we find a very good match to the observed radial distribution of satellites in the Milky Way over the entire radial range. This agreement is mainly due to the ability of our model to track ‘orphan’ galaxies after their subhaloes fall below the resolution limit of the simulation.

How unusual is the Milky Way’s assembly history?

“…As shown in the simulations of the APOSTLE project (Fattahi et al 2016;Sawala et al 2016), the EAGLE model reproduces well the observed stellar mass function for dwarf galaxies, both those that are satellites orbiting in the MW and those in the field around the Local Group. Several important properties of dwarf galaxies, such as their sizes and star formation histories, are also consistent with observations (Sales 2016;Sawala et al 2016;Campbell et al 2017;Digby et al 2019;Bose et al 2019Bose et al , 2020. The number of 'luminous' satellites as a function of halo mass for our various samples is shown in Fig.…”

“…We also examined the radial distribution of satellites and found no discernible differences between the different categories of MW-like galaxies. However, we note that we do not take 'orphan' satellites into consideration, which can make a substantial difference to the radial profiles (Bose et al 2020).…”

Milky Way Satellite Census. I. The Observational Selection Function for Milky Way Satellites in DES Y3 and Pan-STARRS DR1

“…In particular, reionization and stellar feedback drastically suppress dwarf galaxy formation in low-mass halos (e.g., Bullock et al 2000;Somerville 2002;Brown et al 2014), and tidal interactions with the Galactic disk are expected to disrupt a significant number of systems (e.g., Garrison-Kimmel et al 2017;Nadler et al 2018;Kelley et al 2019). Semi-empirical models that account for these effectsalong with realistic satellite detection criteria-find that the observed satellite population is consistent with cold, collisionless dark matter (e.g., Jethwa et al 2018;Kim et al 2018;Newton et al 2018;Bose et al 2019;Nadler et al 2019aNadler et al , 2019b. Likewise, hydrodynamic simulations that self-consistently model galaxy formation in a cosmological context produce luminosity functions and radial distributions of satellites that are broadly consistent with observations of the Milky Way system (e.g., Wetzel et al 2016;Garrison-Kimmel et al 2019;Samuel et al 2020).…”

The detailed structure and the onset of galaxy formation in low-mass gaseous dark matter haloes

“…virial temperature for gas to become self-gravitating and collapse is not a constant, contrary to common assumptions (e.g. Bullock et al 2000;Okamoto & Frenk 2009;Bose et al 2020;Graus et al 2019). Moreover, the critical temperature exhibits a maximum at z ∼ 2, when the Haardt & Madau (2001) photoheating rate peaks.…”

“…10 Finding many more nearby field dwarfs in such low-mass haloes would require either a revision of our current understanding of how (and when) reionization occurred or of how galaxy formation proceeded before reionization. Interestingly, Jethwa, Erkal & Belokurov (2018) and Graus et al (2019) have recently claimed, based on simulations, that such a population of galaxies might be required to explain the sheer number and radial distribution of ultrafaint galaxies observed around the Milky Way but Bose et al (2020) argue that these results are affected by numerical resolution and that, when these effects are taken into account, the data are in good agreement with the observed number of ultrafaint satellites. In any case, our model provides a theoretical framework to address these issues and interpret that data without the need for cosmological simulations and thus, independent of computational cost or resolution limitations.…”