A robust determination of Milky Way satellite properties using hierarchical mass modelling

We introduce a new methodology to robustly determine the mass profile, as well as the overall distribution, of Local Group satellite galaxies. Specifically, we employ a statistical multilevel modelling technique, Bayesian hierarchical modelling, to simultaneously constrain the properties of individual Local Group Milky Way satellite galaxies and the characteristics of the Milky Way satellite population. We show that this methodology reduces the uncertainty in individual dwarf galaxy mass measurements up to a factor of a few for the faintest galaxies. We find that the distribution of Milky Way satellites inferred by this analysis, with the exception of the apparent lack of high-mass haloes, is consistent with the Λ cold dark matter (ΛCDM) paradigm. In particular, we find that both the measured relationship between the maximum circular velocity and the radius at this velocity, as well as the inferred relationship between the mass within 300 pc and luminosity, match the values predicted by ΛCDM simulations for haloes with maximum circular velocities below 20 km s−1. Perhaps more striking is that this analysis seems to suggest a more cusped ‘average’ halo shape that is shared by these galaxies. While this study reconciles many of the observed properties of the Milky Way satellite distribution with that of ΛCDM simulations, we find that there is still a deficit of satellites with maximum circular velocities of 20–40 km s−1.