Quantifying the Milky Way, LMC and their interaction using all-sky kinematics of outer halo stars
Pith reviewed 2026-05-15 14:17 UTC · model grok-4.3
The pith
The LMC induces a 39 km/s reflex motion in the Milky Way outer halo, yielding enclosed masses of 3.36 x 10^11 solar masses for the Milky Way and 8.76 x 10^10 for the LMC.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
Using all-sky kinematics of outer halo stars, the authors apply simulation-based inference with a neural posterior estimator on 32,000 rigid MW-LMC models to find a reflex velocity of 38.6 km/s, an enclosed Milky Way mass within 50 kpc of 3.36 x 10^11 solar masses, an enclosed LMC mass within 50 kpc of 8.76 x 10^10 solar masses, and a pre-infall anisotropy of beta_0 equal to 0.68. These values indicate that the total LMC mass is at least 20 percent of the Milky Way mass.
What carries the argument
Neural posterior estimator trained on the means and dispersions of radial and tangential velocities from simulated halo stars in rigid MW-LMC potentials.
Load-bearing premise
The rigid MW-LMC simulations accurately reproduce the observed velocity means and dispersions without significant bias from unmodeled effects such as time-dependent potentials or substructure.
What would settle it
An independent measurement of the Milky Way center's motion relative to stars at 100 kpc that differs substantially from 39 km/s would falsify the reported reflex velocity magnitude.
read the original abstract
The recent pericentric passage of the Large Magellanic Cloud (LMC) has dislodged the Milky Way's (MW) centre of mass, inducing dynamical disequilibrium, the reflex motion, in the kinematics of outer stellar halo stars. Using data out to $160 \, \rm kpc$ from the combined H3+SEGUE+MagE outer halo survey, we constrain the mass of the MW and LMC, as well as the resulting reflex motion and the stellar halo velocity anisotropy. Using a suite of 32,000 rigid MW--LMC simulations, each with a MW stellar halo evolved to the present day in the combined MW--LMC potential, we perform Simulation Based Inference by training a neural posterior estimator on the means and dispersions of the radial and tangential velocities of stars from the combined H3+SEGUE+MagE outer halo sample. Relative to halo stars at $100 \, \rm kpc$, we find the magnitude of the reflex velocity to be $v_{\rm travel} = 38.6^{+8.3}_{-7.8}\,\rm km \, s^{-1}$. Simultaneously, we determine the enclosed MW mass, $M_{\rm MW}(< 50 \, \rm kpc) = 3.36 \pm 0.15 \times 10^{11}\, \rm M_{\odot}$ and the enclosed LMC mass, $M_{\rm LMC}(< 50 \, \rm kpc) = 8.76^{+1.94}_{-1.77} \times 10^{10}\, \rm M_{\odot}$. Our results suggest that the total LMC mass must be at least $\sim20\%$ that of the MW. The velocity anisotropy prior to the LMC's infall is constrained to be $\beta_0 = 0.68 \pm 0.02$. Finally, we demonstrate that neglecting the LMC in models biases the estimated MW mass to prefer more massive values.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper claims to use all-sky kinematics of outer halo stars from H3+SEGUE+MagE to constrain the MW and LMC enclosed masses within 50 kpc, the reflex velocity v_travel, and the halo velocity anisotropy β0 via simulation-based inference on 32,000 rigid MW-LMC simulations. Key results are v_travel = 38.6^{+8.3}_{-7.8} km s^{-1}, M_MW(<50 kpc) = 3.36 ± 0.15 × 10^{11} M_⊙, M_LMC(<50 kpc) = 8.76^{+1.94}_{-1.77} × 10^{10} M_⊙, and β0 = 0.68 ± 0.02, with the additional finding that ignoring the LMC biases MW mass estimates upward.
Significance. These findings, if robust, provide key measurements of the dynamical influence of the LMC on the MW halo and mass estimates that can be compared to other methods. The SBI approach with a large simulation grid is a positive aspect for deriving posteriors from complex kinematic data.
major comments (2)
- [Methods (simulation suite)] The rigid MW-LMC potential approximation in the 32,000 simulations is load-bearing for the neural posterior estimator results. As noted in the stress-test, unmodeled effects like dynamical friction and halo response could bias the recovered v_travel, M_MW, and M_LMC; the manuscript should include tests showing that the summary statistics (velocity means and dispersions) match those from live simulations or quantify the potential bias.
- [Results] The demonstration that neglecting the LMC biases MW mass estimates is central to the paper's motivation, but the magnitude of this bias and its dependence on the neural estimator should be quantified more explicitly, perhaps with a dedicated comparison in the results section.
minor comments (2)
- [Abstract] The abstract mentions the combined H3+SEGUE+MagE sample but the full text should specify the exact selection criteria and radial range (50-160 kpc) for clarity.
- Check for consistency in the reported uncertainties (e.g., symmetric vs asymmetric) across all parameters.
Simulated Author's Rebuttal
We thank the referee for their constructive comments and positive assessment of our manuscript. We address each major comment below and describe the revisions we will implement.
read point-by-point responses
-
Referee: [Methods (simulation suite)] The rigid MW-LMC potential approximation in the 32,000 simulations is load-bearing for the neural posterior estimator results. As noted in the stress-test, unmodeled effects like dynamical friction and halo response could bias the recovered v_travel, M_MW, and M_LMC; the manuscript should include tests showing that the summary statistics (velocity means and dispersions) match those from live simulations or quantify the potential bias.
Authors: We acknowledge that the rigid potential is a central modeling choice. The existing stress-test section already compares velocity means and dispersions from our rigid simulations against a limited set of live N-body runs and finds agreement within uncertainties. To address the referee's request more fully, we will expand this analysis in the revised manuscript by adding a small suite of live simulations, explicitly quantifying any systematic differences in the summary statistics, and reporting the resulting impact on the inferred parameters. These additions will appear in an extended Methods section. revision: yes
-
Referee: [Results] The demonstration that neglecting the LMC biases MW mass estimates is central to the paper's motivation, but the magnitude of this bias and its dependence on the neural estimator should be quantified more explicitly, perhaps with a dedicated comparison in the results section.
Authors: We agree that a more prominent and quantitative presentation of this result would strengthen the paper. The current manuscript demonstrates the bias via a direct comparison of posteriors with and without the LMC (Figure 8 and surrounding text). In the revision we will add a dedicated subsection in Results that isolates and quantifies the magnitude of the shift in M_MW and examines its sensitivity to the neural posterior estimator by testing alternative network architectures and summary-statistic choices. These changes will be included in the revised manuscript. revision: yes
Circularity Check
No significant circularity; results are outputs of simulation-based inference on independent models
full rationale
The derivation chain uses 32,000 rigid MW-LMC simulations to train a neural posterior estimator on velocity means and dispersions from the H3+SEGUE+MagE sample. The reported values (v_travel, M_MW(<50 kpc), M_LMC(<50 kpc)) are recovered parameters from this inference applied to observed data. No equation or step reduces these outputs to the inputs by construction, nor renames a fit as a prediction. The simulations are generated independently of the target observations, and the method does not invoke self-definitional relations, load-bearing self-citations, or smuggled ansatzes in the provided text. The central claims remain externally falsifiable against the data and alternative models.
Axiom & Free-Parameter Ledger
free parameters (3)
- MW enclosed mass within 50 kpc
- LMC enclosed mass within 50 kpc
- velocity anisotropy beta0
axioms (2)
- domain assumption Outer-halo star kinematics are dominated by the reflex motion induced by the LMC's pericentric passage
- domain assumption Velocity means and dispersions are sufficient statistics for recovering the underlying parameters
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.