REVIEW 4 major objections 4 minor 10 references
The collision between the Milky Way and Andromeda and the fate of their Supermassive Black Holes
T0 review · 4 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read The paper claims that in direct N-body simulations of the Milky Way–Andromeda collision, the two central supermassive black holes reach the same final separation of about 50 parsecs and stall on a nearly circular orbit, independent of the…
desk verdict Preliminary N-body study with an honest limitations section, but the headline 50 pc SMBH stalling radius is not yet supported; treat it as a progress report, not a result. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The argument is carried by direct N-body integration with particle-particle forces and a dynamical-friction term added to mimic the diffuse intergalactic medium. The galaxies are built from a disk, a bulge, and an extended halo, with the black holes inserted as massive particles at the centers; the black holes' masses are one-thousandth of each galaxy's total mass. This setup lets the black holes lose orbital energy dynamically as the galaxies merge, and the repeated result of that sinking is the paper's stalled 50-parsec orbit.
What would settle it
A direct, higher-resolution simulation of the same collision with the observed black-hole masses ($\sim 4\times10^6$ and $\sim 10^8$ solar masses) and enough particles to resolve separations below 50 pc would settle the claim: if the two black holes do not stall near 50 pc under those conditions, the reported convergence is a numerical artifact of the chosen masses or resolution.
Extended reading notes
Core claim
On the paper's own terms, the central discovery is a numerical convergence: in all three simulated initial-velocity cases, the separation between the two supermassive black holes ends at roughly 50 parsecs, where they appear to stall on a nearly circular orbit. The authors state that this final separation is reached independently of the magnitude and orientation of the initial relative velocity, and they interpret the large-scale motion as regular. They caution that their particle number ($6.5\times10^4$) is too low to resolve the subsequent evolution and that the final separation depends significantly on resolution.
Load-bearing premise
The load-bearing premise is that the two black holes can be represented by particles of one-thousandth of their host galaxy's mass (about $10^9$ solar masses), even though the real Milky Way and Andromeda black holes are far lighter; if that artificial mass ratio is wrong, the sinking speed and the stalling radius change.
Editorial extensions
If this is right
- If the claim holds, the Milky Way's and Andromeda's central black holes will not fly apart after the merger; they will end up bound to each other at roughly 50 parsecs.
- The uncertainty in Andromeda's transverse velocity shifts the predicted merger time by billions of years, but does not change the final black-hole separation in these models.
- A 50-parsec stalled binary is still far from gravitational-wave emission; further hardening would require stars, gas, or a third massive body to carry away angular momentum.
- The halo cutoff radius is a controlling parameter for how fast the host galaxies merge, which means measurements of the outer Milky Way and Andromeda halos directly affect the timing of the black holes' encounter.
Reading between the lines
- A natural extension is to rerun the same collision with the observed black-hole masses ($\sim 4\times10^6$ and $\sim 10^8$ solar masses) rather than the $10^9$-solar-mass particles used here; weaker dynamical friction would plausibly stall the pair at a larger separation or on a longer timescale.
- If the same 'same final separation regardless of initial velocity' pattern appears in other galaxy-merger simulations, it would suggest that dynamical friction acts as a strong attractor that erases the memory of incoming orbits, making the post-merger black-hole separation a predictable function of the host mass ratio.
- The result gives a concrete scale to target observationally: searches for dual active galactic nuclei after major mergers could check whether separations cluster near tens of parsecs when the total mass ratio is comparable to Milky Way/Andromeda.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This proceedings paper presents direct N-body simulations of the Milky Way-Andromeda encounter using the HiGPUs code, with each galaxy modeled as a disk, bulge, and Hernquist halo and with central supermassive black holes (SMBHs) of 1.0e9 and 1.6e9 solar masses. Three runs vary Andromeda's transverse velocity (30, 50, and 70 km/s) while fixing the radial velocity at 120 km/s. The paper reports that the galaxy merger time is highly sensitive to the tangential velocity and to the halo cutoff radius, and that the two SMBHs stall at a separation of about 50 pc regardless of the initial relative velocity. The authors explicitly state that the final BH separation depends significantly on particle number and that their resolution is too low to follow the BHs to smaller scales.
Significance. If the 50 pc stalling result were robust, it would be a notable input to the final-parsec problem and to predictions for gravitational-wave detection from a Milky Way-Andromeda SMBH binary. The qualitative sensitivity of the merger time to tangential velocity and halo extent is plausible and consistent with earlier work. The paper is honest about its limitations, which is a strength, but the central SMBH claim is presented as a finding despite an acknowledged resolution dependence. The work is best read as a preliminary progress report rather than a converged physical prediction.
major comments (4)
- [Section 3, Figure 3] The claim that the two SMBHs 'reach the same final separation of about 50 pc, independently of the magnitude and orientation of the initial relative velocity' is not supported by the evidence. The three simulations vary only the transverse speed (30, 50, 70 km/s); the radial velocity is fixed at 120 km/s and the spin orientations are unchanged. Moreover, the very next sentence admits that 'the final value of their separation depends significantly on the number of particle in the simulation.' Without any convergence study, the 50 pc value cannot be distinguished from a resolution-dependent numerical artifact, so the headline claim overreaches the data.
- [Section 2, SMBH initial conditions] The adopted SMBH masses, 1.0e9 and 1.6e9 solar masses, are roughly 250 times and 16 times the observationally inferred masses of the Milky Way and M31 black holes. Dynamical friction and two-body relaxation depend directly on the SMBH-to-star mass ratio, so this choice affects both the sinking rate and the stalling radius. The authors note this is a limitation but provide no test of how the BH separation changes with the assumed mass fraction, leaving the central result conditional on an acknowledged unrealistic setup.
- [Section 2, halo cutoff radius] The text states that 'our simulations show a significant correlation between the cut-off radius of the two halos and the time of the interaction' and that a cutoff of 70 disk scale lengths was therefore chosen. This is potentially circular if the cutoff was adjusted to produce a desired merger time. The reference to Shull (2014) suggests an independent motivation, but the wording 'For this reason' implies a post hoc choice. Please clarify whether the cutoff was fixed a priori or tuned, and report how merger time varies with cutoff radius if the latter.
- [Section 3, numerical methods] No softening length, block-time-step parameters, or particle mass resolution are reported, although these strongly influence the two-body dynamics and the stalling radius. A convergence test with at least one higher-N run is essential to determine whether the 50 pc separation is physical or a resolution-induced attractor. Without it, the BH trajectory result cannot be evaluated as a predictive statement.
minor comments (4)
- [Section 1] The text contains a typo: 'Hernquists profile' should be 'Hernquist profile'.
- [Section 3] The integration scheme is described as '6th order Hermites' and the friction term as 'Chandraskhar's formula'; both should be corrected to 'Hermite' and 'Chandrasekhar', respectively.
- [References] The reference to 'Raychaudury & Lynden-Bell' is missing the ampersand and the name is misspelled; the correct spelling is 'Raychaudhury & Lynden-Bell'.
- [Figure captions] The figure captions use 'unit of 100 kpc' for both galaxy separation and BH separation, but the text claims a BH separation of 50 pc. If the BH separation is indeed shown on a 100 kpc scale, the 50 pc value would be invisible; please clarify the units and consider a separate inlay for the BH trajectories.
Circularity Check
No significant circularity; the merger times and black hole separations are obtained from stated N-body initial conditions, not from fitting or from self-citation of the target result.
full rationale
The paper's central results -- the merger time as a function of M31's tangential velocity and the final roughly 50 pc separation of the two SMBHs -- are produced by direct N-body integration of stated initial conditions, so they are not equivalent by construction to the inputs. The choice of halo cutoff radius is described as informed by simulations showing a correlation with merger time, and the cutoff is then fixed at 70 disk scale lengths citing Shull (2014); this is a parameter selection, not a fitted input renamed as a prediction. The SMBH masses are admittedly larger than observed ones because of the particle resolution, and the authors explicitly state that the final separation depends significantly on particle number; these are acknowledged modeling limitations and numerical caveats, not circular reasoning. The only self-citation is to the HiGPUs code, which is a methodological tool rather than a load-bearing theorem, and no uniqueness argument is imported from the authors' prior work. Thus, no step in the paper reduces to its own input or to an unverified self-citation chain.
Assumptions & free parameters
free parameters (4)
- SMBH mass fraction =
0.001 of galaxy total mass (1.0e9 Msun MW, 1.6e9 Msun M31)
- Halo cutoff radius =
70 disk scale lengths (~150 kpc for MW)
- IGM density rho0 =
~10 times critical density
- M31 tangential velocity Vt0 =
30, 50, 70 km/s
assumptions (4)
- domain assumption The Milky Way and Andromeda form a bound system with a first close approach in 3-5 Gyr.
- domain assumption Galaxies are modeled as exponential disks plus Hernquist bulges and halos with specified mass ratios.
- domain assumption Chandrasekhar's dynamical friction formula describes the dissipative effect of the IGM on the galaxy motion.
- domain assumption Direct particle-particle N-body integration with the stated particle number adequately captures the collision dynamics.
Cite this review
Pith. "Pith review of The collision between the Milky Way and Andromeda and the fate of their Supermassive Black Holes." pith.science (2026). https://pith.science/paper/N35E33TU
@misc{pith2026190807278,
author = {Pith},
title = {Pith review of: The collision between the Milky Way and Andromeda and the fate of their Supermassive Black Holes},
year = {2026},
howpublished = {\url{https://pith.science/paper/N35E33TU}},
note = {Machine review of arXiv:1908.07278}
}
read the original abstract
Our Galaxy and the nearby Andromeda Galaxy (M31) form a bound system, even though the relative velocity vector of M31 is currently not well constrained. Their orbital motion is highly dependent on the initial conditions, but all the reliable scenarios imply a first close approach in the next 3-5 Gyrs. In our study, we simulate this interaction via direct N-body integration, using the HiGPUs code. Our aim is to investigate the dependence of the time of the merger on the physical and dynamical properties of the system. Finally, we study the dynamical evolution of the two Supermassive Black Holes placed in the two galactic centers, with the future aim to achieve a proper resolution to follow their motion until they form a tight binary system.
Figures
Reference graph
Works this paper leans on
-
[1]
Capuzzo-Dolcetta, R. et al. 2013, Journ. of Comp. Phys., v. 236, p. 580-593
work page 2013
-
[2]
1987, Galactic Dynamics, Princeton, NJ, Princeton University Press
Binney J., Tremaine S. 1987, Galactic Dynamics, Princeton, NJ, Princeton University Press
work page 1987
-
[3]
Cox, T. J. & Loeb, A. 2008, MNRAS, 386, 461–474
work page 2008
-
[4]
Van der Marel, R. P. et al. 2012, ApJ, 753, 9
work page 2012
-
[5]
Salomon, J. B. et al. 2016, MNRAS, 456, 4432
work page 2016
-
[6]
Van der Marel, R. P. et al. 2019, ApJ, 872, 24
work page 2019
-
[7]
1995, Astronomical Data Analysis Software and Systems IV, 77, 398-401
Teuben, P. 1995, Astronomical Data Analysis Software and Systems IV, 77, 398-401
work page 1995
-
[8]
Shull, J. M. 2014, ApJ, 784, 142
work page 2014
Show all 10 references
-
[9]
& Lynden-Bell, D
Raychaudury, S. & Lynden-Bell, D. 1989, Mon. Not. R. astr. Soc., 240,195-218
1989
-
[10]
Widrow, L. M. & Dubinski, J. 2005, ApJ, 631, 838-855
2005
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.