Searching for young runaways across the sky

TL;DR

This paper compiles a catalog of 3354 young stars likely ejected from star-forming regions within 500 parsecs, using 2D traceback methods, and discusses the challenges of confirming their kinematics with Gaia DR3 data.

Contribution

It introduces a homogeneous 2D traceback method to identify young runaway stars and potential ejected pairs, highlighting the need for spectral follow-up and addressing Gaia DR3 limitations.

Findings

Catalog of 3354 candidate young runaways within 500 pc.

Identification of potential ejected star pairs with similar traceback times.

Gaia DR3 radial velocities are unreliable for low-mass, young stars due to spectral complexities.

Abstract

We present a catalog of 3354 candidate young stars within 500 pc that appear to have been ejected from their parent associations with relative speeds of >5 km/s. These candidates have been homogeneously selected through performing a 2d spherical traceback of previously identified pre-main sequence candidates to various star forming regions, ensuring that the traceback age as well as the estimated age of a star is consistent with the age of the population, and excluding contaminants from the nearby moving groups that follow the dominant velocity currents in the field. Among the identified candidates we identify a number of pairs that appear to have interacted in the process of the ejection, these pairs have similar traceback time, and their trajectory appears to be diametrically opposite from each other, or they have formed a wide binary in the process. As the selection of these candidates is performed solely in 2d, spectral follow-up is necessary for their eventual confirmation. Unfortunately, recently released Gaia DR3 radial velocities appear to be unsuitable for characterizing the kinematics of low mass stars with ages <100 Myr, as the accretion, activity, and a variety of other spectral features that make them distinct from the more evolved stars do not appear to have been accurately accounted for in the data, resulting in significant artificially inflated scatter in their RV distribution.