We present evidence for a metal-poor, [Fe/H]$sim-1.4$ $sigma$=0.2 dex, stellar halo component detectable at radii from 10 kpc to 70 kpc, in our nearest giant spiral neighbor, the Andromeda galaxy. This metal-poor sample underlies the recently-discovered extended rotating component, and has no detected metallicity gradient. This discovery uses a large sample of 9861 radial velocities of Red Giant Branch (RGB) stars obtained with the Keck-II telescope and DEIMOS spectrograph, with 827 stars with robust radial velocity measurements isolated kinematically to lie in the halo component primarily by windowing out the extended rotating component which dominates the photometric profile of Andromeda out to $<$50 kpc (de-projected). The stars lie in 54 spectroscopic fields spread over an 8 square degree region, and are expected to fairly sample the halo to a radius of $sim$70 kpc. The halo sample shows no significant evidence for rotation. Fitting a simple model in which the velocity dispersion of the component decreases with radius, we find a central velocity dispersion of $152kms$ decreasing by $-0.90kms/kpc$. By fitting a cosmologically-motivated NFW halo model to the halo stars we constrain the virial mass of M31 to be greater than $9.0 times 10^{11} msun$ with 99% confidence. The properties of this halo component are very similar to that found in our Milky Way, revealing that these roughly equal mass galaxies may have led similar accretion and evolutionary paths in the early Universe.
Download