Tracing the first stars and galaxies of the Milky Way

TL;DR

This study uses high-resolution simulations to trace the formation and evolution of the first stars and galaxies in Milky Way-like systems, revealing their environments, merger histories, and present-day remnants.

Contribution

It introduces models for the first star and galaxy formation sites, quantifies their numbers, and provides fitting functions for their accretion history in Milky Way-mass halos.

Findings

Approximately 23,000 Pop III star-forming sites per host without feedback.

Lyman-Werner feedback reduces star-forming sites to ~550.

Most early halos form in isolation and are not externally enriched.

Abstract

We use 30 high-resolution dark matter halos of the $Caterpillar$ simulation suite to probe the first stars and galaxies of Milky Way-mass systems. We quantify the environment of the high-$z$ progenitors of the Milky Way and connect them to the properties of the host and satellites today. We identify the formation sites of the first generation of Population III (Pop III) stars ($z$ ~ 25) and first galaxies ($z$ ~ 22) with several different models based on a minimum halo mass including a simple model for Lyman-Werner feedback. Through this method we find approximately 23,000 $\pm$ 5,000 Pop III potentially star-forming sites per Milky Way-mass host, though this number is drastically reduced to ~550 star-forming sites when Lyman-Werner feedback is included, as it has critical effects at these length scales. The majority of these halos identified form in isolation (96% at $z$ = 15) and are not subject to external enrichment by neighboring halos (median separation ~1 pkpc at $z$ = 15), though half merge with a system larger than themselves within 1.5 Gyrs. Approximately 55% of the entire population has merged into the host halo by $z$ = 0. Using particle tagging, we additionally trace the Pop III remnant population to $z$ = 0 and find an order of magnitude scatter in their number density at small (i.e. r $<$ 5 kpc) and large (i.e. r $>$ 50 kpc) galactocentric radii at $z$ = 0. Using our large number of realizations, we provide fitting functions for determining the number of progenitor minihalo and atomic cooling halo systems that present-day dwarf galaxies and the Magellanic cloud system might have accreted since their formation. We demonstrate that observed dwarf galaxies with stellar masses below 10$^{4.6}$ M$_{\odot}$ are unlikely to have merged with any other star-forming systems.