The Structure and 3D Kinematics of Vela OB2

TL;DR

This study uses spectroscopic data combined with Gaia measurements to analyze the 3D kinematics of stars in Vela OB2, revealing multiple kinematic groups, anisotropic expansion, and complex formation history over approximately 10 million years.

Contribution

The paper provides the first detailed 3D kinematic analysis of Vela OB2 using combined spectroscopic and Gaia data, uncovering its substructure and expansion dynamics.

Findings

Multiple kinematic groups identified within Vela OB2

Strong anisotropic expansion observed in the association

NGC 2547 is an interloper, not part of Vela OB2

Abstract

The kinematics of stars in OB associations can provide insights into their formation, dynamical evolution, and eventual fate. The low-mass stellar content of OB associations are sufficiently numerous as to provide a detailed sampling of their kinematic properties, however spectroscopy is required to confirm the youth of individual stars and to get 3D kinematics. In this paper we present and analyse results from a large spectroscopic survey of Vela OB2 conducted using 2dF/HERMES on the AAT. This spectroscopy is used to confirm the youth of candidate young stars and determine radial velocities, which are combined with proper motions and parallaxes from Gaia to measure 3-dimensional positions and velocities. We identify multiple separate kinematic groups in the region, for which we measure velocity dispersions and infer their virial states. We measure expansion rates for all these groups and find strong evidence for anisotropic expansion in the Vela OB2 association of at least 11$\sigma$ significance in all three dimensions, as well as some evidence for expansion in the $\gamma$ Vel and P Puppis clusters. We trace back the motions of these groups into the past and find that the open cluster NGC 2547 is an interloper in the Vela OB2 region and actually formed $>$100 pc away from the association. We conclude that Vela OB2 must have formed with considerable spatial and kinematic substructure over a timescale of $\sim$10 Myr, with clear temporal substructure within the association, but no clear evidence for an age gradient.