High-velocity stars are rare objects that move significantly faster than typical disk stars. Their trajectories extend over tens of kiloparsecs, making them sensitive probes of the Galactic gravitational potential, including the dark matter distribution. Several research directions have emerged around this topic: some studies investigate the origin of such stars through ejection mechanisms from the Galactic center or disruption of dwarf satellites, while others use them to test halo models. This paper stands apart: it combines large samples from different surveys and applies the classical machinery of stellar kinematics to extreme objects rather than to ordinary disk stars.
The author assembled three programs of high-velocity stars: 591 stars from LAMOST DR7 and Gaia DR2 (Li et al. 2021), 87 stars from large surveys (Li et al. 2023), and 519 stars associated with the Large Magellanic Cloud (Liao et al. 2024). For each star, coordinates, proper motions, and radial velocities were taken from Gaia DR3, and then space velocities U, V, W were computed in Galactic coordinates. From these velocities, the velocity ellipsoid was constructed: the three principal dispersions σ1, σ2, σ3 and their orientation via direction cosines were determined. The Oort constants A and B were obtained in two ways: A from the dependence of radial velocities on Galactic longitude, and B from the ratio of dispersions σ2/σ1. Additionally, the AD-diagram method was used to find the apex coordinates of the Solar motion and the convergent points of each sample.
The three programs yield markedly different parameters. Program I (591 stars, 0.10–15.40 kpc) gives dispersions of 489.12, 371.04, and 196.16 km/s and a Solar speed of 430.64 km/s. Program II (87 stars, up to 108.64 kpc) shows the largest dispersions (556.46, 475.50, 330.03 km/s) and a Solar speed of 443.10 km/s. Program III (519 stars, 0.29–16.44 kpc) has the smallest dispersions (378.39, 284.16, 186.00 km/s) and a Solar speed of 303.98 km/s. The Oort constants also differ: for Program I, A = 15.427 ± 0.25, B = -21.095 ± 0.22; for Program II, A = 3.883 ± 0.51, B = -11.029 ± 0.30; for Program III, A = 16.510 ± 0.25, B = -21.227 ± 0.22 (all in km/s/kpc). The average values across all three programs are A = 11.94 ± 0.29, B = -17.78 ± 0.24, angular speed |A-B| = 25.07 ± 5.01 km/s/kpc, and rotation speed at the solar radius Vo = 243.72 ± 15.61 km/s. Program II, which includes the most distant stars, yields anomalously low constants, which the author attributes to systematic errors due to large distances.
This work differs from neighboring studies in that it applies the standard kinematic machinery (velocity ellipsoid, Oort constants) to an extreme population of high-velocity stars rather than to disk stars or typical halo stars. This allows testing how suitable such objects are as tracers of the Galactic potential. The results show that the sample strongly depends on the origin of the stars: Program II, consisting of stars ejected from the Galactic center, gives distorted constants due to large distances and possibly inhomogeneity. In contrast, Programs I and III, although they include stars with different acceleration mechanisms, yield consistent values of A and B, close to literature values. This means that high-velocity stars can be used to study local dynamics, but with caution: their origin must be taken into account, and objects with reliable parallaxes should be selected.
In the future, the sample could be expanded with new data from Gaia DR4 and future spectroscopic surveys such as WEAVE and 4MOST. This would reduce statistical errors and allow testing systematic effects related to distance. It would also be interesting to separate stars by origin (ejection from the center, disruption of dwarf galaxies, runaways from clusters) and study how each group affects the derived parameters. Orbital modeling in a realistic Galactic potential would help separate true kinematic signals from sample artifacts.
The results are useful for refining the local rotation curve of the Galaxy and calibrating mass distribution models, including the dark halo. They also provide benchmarks for interpreting data on hypervelocity stars and their use as tests of gravitational theories.
The next step is to repeat the analysis on a more homogeneous sample with accurate parallaxes from Gaia DR4 and to add chemical composition data to separate populations by origin. Then, orbits can be constructed and compared with potential models to estimate the contribution of dark matter.
In stellar kinematics, the shape and orientation of the velocity ellipsoid in different Galactic components, especially in the halo, remain open questions. High-velocity stars, being rare, provide poor statistics, but their extreme velocities are sensitive to the anisotropy of the potential. This work shows that Oort constants derived from such stars can be biased if the sample is heterogeneous in origin. This connects the problem of local kinematic parameters with the global structure of the halo and the history of accretion.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The study combines 1197 high-velocity stars from three programs, using LAMOST and Gaia DR3 data, and applies the classical machinery of stellar kinematics: it computes space velocities, dispersions, velocity ellipsoid parameters, and Oort constants. The mean values obtained are A = 11.94 ± 0.29 km/s/kpc, B = -17.78 ± 0.24 km/s/kpc, and a rotation speed at the solar radius of 243.72 ± 15.61 km/s.
Where its strength is
Applying standard kinematic methods to an extreme population of high-velocity stars, rather than to ordinary disk stars, allows us to test their suitability as tracers of the Galactic potential, and combining three heterogeneous samples provides a more complete picture.
Where it could go next
Worth trying
Why this may matter
Refining the local dynamics of the Galaxy using high-velocity stars could help calibrate dark matter distribution models and shed light on the origin of these objects.
Nearby works in our archive
The archive is sparse around this work: the closest review is only 0.59, against a usual 0.64. It looks like few have drilled nearby.
The Galactic rotation speed at the Sun's distance determined in this work (243.72 km/s) means that the Solar System completes a full orbit around the Galactic center in about 210 million years. The last time we were on this same side of the Galaxy, early dinosaurs were roaming the Earth.