Investigating the Upper Scorpius OB association with HERMES. I. The spectroscopic sample and 6D kinematics

TL;DR

This work presents a large spectroscopic survey of Upper Scorpius using 2dF/HERMES to identify >1000 PMS stars and to measure their radial velocities, enabling full 6D kinematics when combined with Gaia EDR3. By cross-matching with literature samples and applying robust youth criteria, the authors construct a 544-star, high-quality 6D sample distributed into seven subgroups, and analyze their velocity dispersions, virial states, expansion, rotation, and traceback to reconstruct their dynamical histories. The results show that Upper Scorpius is predominantly unbound and expanding, with strong anisotropy in velocity space; initial configurations were likely large and sparse, arguing against a cluster-chain formation scenario where subgroups trigger each other. The study favors star formation scenarios driven by large-scale, possibly external triggering (e.g., GMC interactions or UCL feedback) producing multiple subgroups with independent formation histories, rather than a single, centrally bound cluster undergoing gas expulsion. These findings provide a detailed kinematic framework for Upper Scorpius and demonstrate the power of combining spectroscopic youth indicators with precise astrometry to unravel the assembly history of nearby OB associations.

Abstract

We present results from a large spectroscopic survey of the nearest young association to the Sun, Upper Scorpius, conducted using 2dF/HERMES on the Anglo-Australian Telescope. We use spectroscopic youth criteria such as Li-equivalent widths to identify >1000 pre-main sequence (PMS) members across the region and measure radial velocities, combining these with Gaia EDR3 5-parameter astrometry to obtain 6D kinematic information. We separate confirmed PMS association members into distinct kinematic groups and measure expansion and rotation trends in each. We find strong evidence for asymmetric expansion in several groups and derive expansion timescales from the greatest rates of expansion in each group. We also trace the past motion of these groups using an epicycle approximation and estimate the time since their most compact configuration. These kinematic properties are compared to literature ages and the star formation history of Upper Scorpius is discussed. We find evidence that a scenario in which star formation in the subgroups of Upper Scorpius proceeded independently, either by self-instability or external feedback from Upper Centaurus-Lupus, is more likely than a recently proposed "cluster chain" scenario in which these subgroups have triggered each other.

Figures (20)

• Figure 1: Gaia DR3 BP-RP colour vs. absolute G magnitude for sources in the Upper Sco region with $13<Gmag<16.5$ and $\varpi<5$ (blue) or $\varpi<2.5$ (orange). Absolute magnitudes are calculated using distance estimates from bailerjones21 and average $A_G$ and $E$(BP-RP) values from Gaia DR2 per $0.25\deg^2$ square by 50 pc los-distance volume, similar to the method employed by zari19. Overplotted is the 20 Myr PMS isochrone (pink) from baraffe15.
• Figure 2: Sky positions of photometrically selected targets colour-coded by their priority (Fig. \ref{['CMD']}). Sources identified as YSOs by variability in Gaia DR3 marton23 are indicated with plus symbols. The 2$^\circ$-diameter circles indicate the individual fields of observation of the AAT within which targets were selected. The fields were distributed to effectively cover the region of Upper Scorpius with the greatest density of likely YSOs.
• Figure 3: Histogram of calibrated radial velocities (RVs) for 6727 sources in our sample of HERMES targets.
• Figure 4: Equivalent width of the lithium 6708 Å line plotted against the Gaia DR3 BP - RP colour. Sources that pass one or more of our YSO criteria are plotted in green, sources that pass one or more of our YSO criteria and also pass our astrometric quality criteria are plotted in red, sources that meet no YSO criteria are plotted in blue (likely non-YSOs). The EW(Li) threshold value for selecting YSOs (100 mÅ) is indicated with a dashed line.
• Figure 5: Sky positions of confirmed YSOs observed with the AAT. Vectors indicate the proper motion of each source relative to the median of all Upper Sco YSOs in the sample, colour-coded based on the position angle of the proper motion (see the colour wheel in the top left as a key). The magnitude scale (mas yr$^{-1}$) of proper motion vectors is indicated by the scale bar in the bottom right.