🔭 astrophysics

Runaway OB stars within 1 kpc of the Sun

This study analyzes a volume-complete sample of OB stars within 1 kpc of the Sun to determine runaway star probabilities, finding that approximately 17.5% of O-type and 6.9% of B-type stars are runaways based on a fixed velocity threshold, a result consistent with previous research but influenced by sample selection and methodology.

Original authors: Juan Martinez Garcia, Nicholas Wright, Alexis Quintana

Published 2026-07-17

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Juan Martinez Garcia, Nicholas Wright, Alexis Quintana

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the Milky Way not as a static, silent ocean of stars, but as a bustling, chaotic city where massive, brilliant buildings are constantly being constructed and occasionally demolished. In this cosmic metropolis, the most massive and luminous stars—known as O-type and B-type stars—are like the city's skyscrapers: they burn bright, live fast, and are born in crowded neighborhoods called star clusters. Usually, these giants stay close to their birthplaces, but sometimes, something dramatic happens. A star might get kicked out of its neighborhood, perhaps because a neighbor exploded in a supernova or because of a violent gravitational tussle with other stars. These ejected travelers, zooming through space at high speeds, are called "runaway stars."

Astronomers care about these cosmic drifters because they are like crime scene witnesses. By tracking where they are going and how fast they are moving, scientists can figure out exactly how they were kicked out and what their original neighborhood looked like before they left. It's a bit like finding a car speeding down a highway and trying to guess which garage it was stolen from and what kind of crash caused it to fly off the lot. However, figuring out who is a runaway and who is just a normal commuter is tricky. You need to know exactly how fast the "traffic" (the general rotation of the galaxy) is moving to spot the outliers. If you get the background speed wrong, you might think a normal star is a runaway, or miss a real one entirely.

This is exactly the puzzle Juan Martínez García and their team at Keele University tackled in their new study. They decided to take a census of the most massive stars within a 1,000-light-year radius of our Sun—a neighborhood that is supposed to be fully mapped out, meaning they didn't just look at the brightest or closest stars, but tried to count every single one of them. They gathered data on 40 O-type stars and nearly 25,000 B-type stars. To separate the runners from the walkers, the team first built a super-accurate model of how the galaxy normally spins and moves. They subtracted this "normal" motion from the actual movements of the stars to see who was left with a "peculiar" speed—a speed that didn't fit the pattern.

They used two different ways to spot the runners. The first was a strict speed limit: if a star was moving faster than 23 kilometers per second relative to the local traffic, they flagged it as a runaway. The second method was more statistical, looking for stars that were moving so much faster than the average that they were clearly statistical outliers. The results were revealing. Using the speed limit method, they found that about 17.5% of the O-type stars are runaways, while roughly 7% of the B-type stars are. This suggests that the biggest, most massive stars are much more likely to be ejected from their homes than their slightly smaller cousins.

The team also checked their work by looking at a smaller group of stars where they had 3D speed data (including how fast they were moving toward or away from us), and the results held up, suggesting their 2D method was a solid approximation. They compared their findings to previous studies and found that while other researchers had reported wildly different numbers (ranging from 5% to over 50%), those differences likely came from looking at incomplete or biased lists of stars. By using a complete sample, this study suggests that the true runaway fraction for massive stars is likely right around the 7% to 17.5% mark.

Interestingly, the study hints that the mechanism for these ejections might differ by star size. Theoretical models suggest that the "supernova explosion" method (where a companion star blows up) is unlikely to produce the high number of runaway O-type stars the team found. Instead, the data leans toward the "dynamical ejection" scenario, where gravitational slingshots in crowded star clusters fling stars out. While the team didn't pinpoint the exact birth cluster for every runaway, their work provides a much clearer map of the "escapees" in our cosmic backyard, setting the stage for future research to trace them back to their origins.

Technical Summary: Runaway OB stars within 1 kpc of the Sun

Problem Statement
Runaway stars are high-velocity objects ejected from their birth environments, providing critical insights into the kinematic history of stellar clusters and the mechanisms driving stellar ejection. While the term traditionally applies to OB stars due to their observability, estimates of the "runaway fraction" (the proportion of OB stars that are runaways) have historically shown wide dispersion, ranging from ~1.5% to over 50%. This inconsistency is largely attributed to variations in the underlying stellar samples (often magnitude-limited or spectroscopically biased) and the diverse methods used to define and select runaway stars. Previous studies have struggled to provide a unified, volume-complete census of runaway stars to resolve these discrepancies.

Methodology
This study utilizes a recently published, volume-complete sample of OB stars within 1 kpc of the Sun, derived from the catalogue by Quintana et al. (2025). The final sample comprises 40 O-type stars and 24,488 B-type stars. The analysis proceeds through the following steps:

Galactic Rotation Modeling: The authors fit a Galactic rotation model to the observed proper motions of the sample using a Markov Chain Monte Carlo (MCMC) approach. The model treats the galactocentric velocity as a function of Galactic radius and longitude, parameterizing the azimuthal velocity ( $V_{\phi}$ ) and radial velocity ( $V_R$ ) and their first derivatives with respect to radius. The model accounts for the Sun's motion relative to the Galactic Center.
Residual Velocity Calculation: By subtracting the model-predicted velocities from the observed velocities, the authors derive the peculiar velocities of the stars in the Galactic plane ( $\ell$ ) and vertical ( $b$ ) directions.
Statistical Characterization: The distribution of these residual velocities is fitted to a Maxwell-Boltzmann distribution to characterize the kinematic dispersion of the non-runaway population.
Runaway Selection Criteria: Two distinct methods are employed to identify runaway stars:
- Method 1 (Fixed Threshold): A fixed 2D peculiar velocity threshold of $23 \text{ km s}^{-1}$ is applied. This value is chosen based on the intersection of the observed probability density and the Maxwell-Boltzmann fit, representing the velocity where a star is more likely than not to be a runaway.
- Method 2 (Normalized Dispersion): Stars are identified as runaways if their uncertainty-normalized peculiar velocities exceed a significance threshold ( $E > 1$ ), effectively selecting stars with velocities significantly higher than the sample dispersion relative to their measurement errors.
Monte Carlo Simulation: To account for measurement uncertainties and model parameter variations, a Monte Carlo experiment (500 simulations) is run. The probability of a star being a runaway is calculated based on the frequency with which it exceeds the selection criteria across simulations.
3D Calibration: The 2D results are calibrated using a subset of 3,884 stars with available radial velocities (from APOGEE and LAMOST) to compute 3D peculiar velocities and verify the 2D threshold against a 3D Maxwell-Boltzmann fit.

Key Results

Runaway Fractions: Using the fixed velocity threshold (Method 1), the study finds a runaway fraction of 17.5% ( $^{+0.1}_{-2.5}$ ) for O-type stars and 7.0% ( $\pm 0.1$ ) for B-type stars. Method 2 yields lower fractions: 10.0% for O-type and 2.5% for B-type stars.
Velocity Distribution: The residual velocities of the sample are well-described by a Maxwell-Boltzmann distribution with a parameter $b \approx 11 \text{ km s}^{-1}$ (2D) and $b \approx 14 \text{ km s}^{-1}$ (3D). The 3D analysis confirms that the 2D threshold of $23 \text{ km s}^{-1}$ is a reasonable approximation for a full 3D treatment, corresponding to a 3D threshold of approximately $28 \text{ km s}^{-1}$ .
Individual Candidates: The study confirms the runaway status of several well-known candidates (e.g., $\mu$ Col, AE Aur, $\zeta$ Pup, 68 Cyg, $\lambda$ Cep) based on proper motions. It also identifies stars previously considered runaways based solely on radial velocity (e.g., $\xi$ Per, X Per) as having insufficient 2D proper motion to be classified as runaways under these criteria, highlighting the limitations of 2D-only analysis for specific objects.
Spatial Distribution: Runaway stars, particularly B-types, show a more diffuse spatial distribution compared to the general OB population. O-type runaways remain closer to the Galactic plane, likely due to their younger ages. The vertical distribution of runaways shows a reduced weight for the "thin" disk component compared to the general OB population, indicating runaways are more likely to be found at higher Galactic latitudes.

Significance and Claims
The authors claim that the primary significance of this work lies in the use of a volume-complete sample of OB stars within 1 kpc. Unlike previous studies that relied on magnitude-limited or spectroscopically biased samples (which may overrepresent runaways because they are easier to observe in non-embedded environments), this volume-complete approach mitigates selection biases.

The results demonstrate that O-type stars have a significantly higher runaway fraction than B-type stars, a trend hinted at in previous literature but now quantified with a complete local sample. The authors suggest that the high runaway fraction of O-type stars supports the Dynamical Ejection Scenario (DES) as the dominant mechanism for their ejection, whereas the lower fraction for B-type stars may imply a larger contribution from the Binary Supernova Scenario (BSS). The paper concludes that while the derived fractions are consistent with some previous studies, the differences observed in the literature are largely attributable to the underlying samples and selection methodologies rather than intrinsic astrophysical variations. The study provides a robust baseline for future work tracing these stars back to their birth clusters.

More like this