The RAVE Survey: Constraining the Local Galactic Escape Speed

TL;DR

New constraints on the local escape speed of the Galaxy are reported, finding that the escape velocity lies within the range 498 <v(esc) <608 km s (-1) (90 per cent confidence), with a median likelihood of 544 km s(-1).

Abstract

We report new constraints on the local escape speed of our Galaxy. Our analysis is based on a sample of high velocity stars from the RAVE survey and two previously published datasets. We use cosmological simulations of disk galaxy formation to motivate our assumptions on the shape of the velocity distribution, allowing for a significantly more precise measurement of the escape velocity compared to previous studies. We find that the escape velocity lies within the range $498\kms < \ve < 608 \kms$ (90 per cent confidence), with a median likelihood of $544\kms$. The fact that $\ve^2$ is significantly greater than $2\vc^2$ (where $\vc=220\kms$ is the local circular velocity) implies that there must be a significant amount of mass exterior to the Solar circle, i.e. this convincingly demonstrates the presence of a dark halo in the Galaxy. For a simple isothermal halo, one can calculate that the minimum radial extent is $\sim58$ kpc. We use our constraints on $\ve$ to determine the mass of the Milky Way halo for three halo profiles. For example, an adiabatically contracted NFW halo model results in a virial mass of $1.42^{+1.14}_{-0.54}\times10^{12}M_\odot$ and virial radius of $305^{+66}_{-45}$ kpc (90 per cent confidence). For this model the circular velocity at the virial radius is $142^{+31}_{-21}\kms$. Although our halo masses are model dependent, we find that they are in good agreement with each other.

Figures (9)

• Figure 1: This figure shows the velocity distribution of the stars (dashed) and dark matter particles (solid) from our four simulated galaxies. These distributions are for particles whose distance from the centre of the galaxy is between 3 and 14 kpc, and the stellar particles are those belonging to the 'spheroid' population.
• Figure 2: This figure shows the likelihood estimates for the exponent $k$, which denotes the shape of the velocity distribution. This is shown for both the stellar (thick) and dark matter (thin) high velocity 'solar neighbourhood' samples. The solid, dotted, short dashed and long dashed lines represent simulated galaxies KIA1, KIA2, KIA3 and KIA4, respectively.
• Figure 3: This figure shows the fractional contribution of each Galactic component as a function of $v_{\rm min}$, i.e. the minimum radial velocity of the sample. These results come from the toy model described in Section \ref{['sec:vmin']}.
• Figure 4: Results from the follow-up work described in Section \ref{['sec:validation']} for 12 of our high radial velocity RAVE stars. The horizontal axis shows the difference between the velocity as reported in the RAVE catalogue compared to the weighted mean of all velocities taken during the follow-up campaign. Note the good agreement between the two measurements. Typical errors are $\sim 2{\rm \,km\,s^{-1}}$ for the RAVE catalogue and $<1 {\rm \,km\,s^{-1}}$ for the follow-up velocities.
• Figure 5: The relation between radial velocity (corrected for Solar motion) and longitude for stars in the RAVE catalogue. Note that the signature of the disc is clearly visible. The horizontal lines correspond to $v_{\rm r}=-300,-250,0,+250,+300{\rm \,km\,s^{-1}}$. The crosses simply denote stars with $\left|v_{\rm r}\right|>250{\rm \,km\,s^{-1}}$.