Gaussian processes, median statistics, Milky Way rotation curves

We use the Iocco et al. (2015) compilation of 2,780 circular velocity measurements as a baseline sample to analyze the Milky Way rotation curve. We focus on techniques that adequately capture the small-scale structure (structure on the order of ∼0.5 kpc) present in the Milky Way rotation curve and that can be used to show the azimuthal anisotropy of the Milky Way rotation curve. We fit the data with simple, few parameter, rotation curve functions which are unable to adequately capture the significant small-scale spatial structure in these data and so provide poor fits. We introduce and use the Gaussian Processes (GP) method to derive Milky Way rotation curves from the dataset which describe the small-scale structure more effectively than simple fits. The GP method rotation curves present significant small-scale spatial structure superimposed on a broad rise to galactocentric radius $R\approx 7$ kpc and a decline at larger $R$ . We find that our rotation curve is consistent with previous data, and study correlations in the residual circular velocities which are consistent with the measurements of McClure-Griffiths and Dickey (2016).