The origin of accreted stellar halo populations in the Milky Way using APOGEE,<i>Gaia</i>, and the EAGLE simulations

Abstract

Recent work indicates that the nearby Galactic halo is dominated by the\ndebris from a major accretion event. We confirm that result from an analysis of\nAPOGEE-DR14 element abundances and $\\textit{Gaia}$-DR2 kinematics of halo\nstars. We show that $\\sim$2/3 of nearby halo stars have high orbital\neccentricities ($e \\gtrsim 0.8$), and abundance patterns typical of massive\nMilky Way dwarf galaxy satellites today, characterised by relatively low\n[Fe/H], [Mg/Fe], [Al/Fe], and [Ni/Fe]. The trend followed by high $e$ stars in\nthe [Mg/Fe]-[Fe/H] plane shows a change of slope at [Fe/H]$\\sim-1.3$, which is\nalso typical of stellar populations from relatively massive dwarf galaxies. Low\n$e$ stars exhibit no such change of slope within the observed [Fe/H] range and\nshow slightly higher abundances of Mg, Al and Ni. Unlike their low $e$\ncounterparts, high $e$ stars show slightly retrograde motion, make higher\nvertical excursions and reach larger apocentre radii. By comparing the position\nin [Mg/Fe]-[Fe/H] space of high $e$ stars with those of accreted galaxies from\nthe EAGLE suite of cosmological simulations we constrain the mass of the\naccreted satellite to be in the range $10^{8.5}\\lesssim M_*\\lesssim\n10^{9}\\mathrm{M_\\odot}$. We show that the median orbital eccentricities of\ndebris are largely unchanged since merger time, implying that this accretion\nevent likely happened at $z\\lesssim1.5$. The exact nature of the low $e$\npopulation is unclear, but we hypothesise that it is a combination of\n$\\textit{in situ}$ star formation, high $|z|$ disc stars, lower mass accretion\nevents, and contamination by the low $e$ tail of the high $e$ population.\nFinally, our results imply that the accretion history of the Milky Way was\nquite unusual.\n