Chemical separation of disc components using RAVE

TL;DR

This paper uses RAVE survey data to chemically separate and analyze the kinematic properties of thin and thick disc stars in the solar neighborhood, revealing distinct velocity trends and dispersions.

Contribution

It introduces a probabilistic chemical separation method for disc components that can be applied to datasets with limited [$ ext{alpha}$/Fe] accuracy, enhancing kinematic analysis.

Findings

Negative $V_{\phi}$-metallicity trend in $ ext{alpha}$-low disc

Positive $V_{\phi}$-metallicity gradient in $ ext{alpha}$-high disc

Differences in velocity dispersions between the two disc components

Abstract

We present evidence from the RAdial Velocity Experiment (RAVE) survey of chemically separated, kinematically distinct disc components in the solar neighbourhood. We apply probabilistic chemical selection criteria to separate our sample into $\alpha$-low (`thin disc') and $\alpha$-high (`thick disc') sequences. Using newly derived distances, which will be utilized in the upcoming RAVE DR5, we explore the kinematic trends as a function of metallicity for each of the disc components. For our $\alpha$-low disc, we find a negative trend in the mean rotational velocity ($V_{\mathrm{\phi}}$) as a function of iron abundance ([Fe/H]). We measure a positive gradient $\partial V_{\mathrm{\phi}}$/$\partial$[Fe/H] for the $\alpha$-high disc, consistent with results from high-resolution surveys. We also find differences between the $\alpha$-low and $\alpha$-high discs in all three components of velocity dispersion. We discuss the implications of an $\alpha$-low, metal-rich population originating from the inner Galaxy, where the orbits of these stars have been significantly altered by radial mixing mechanisms in order to bring them into the solar neighbourhood. The probabilistic separation we propose can be extended to other data sets for which the accuracy in [$\alpha$/Fe] is not sufficient to disentangle the chemical disc components a priori. For such datasets which will also have significant overlap with Gaia DR1, we can therefore make full use of the improved parallax and proper motion data as it becomes available to investigate kinematic trends in these chemical disc components.