pith. sign in

arxiv: 2510.05982 · v2 · submitted 2025-10-07 · 🌌 astro-ph.SR

Dormant black hole candidates from Gaia DR3 summary diagnostics

Johanna M\"uller-Horn , Hans-Walter Rix , Kareem El-Badry , Ben Pennell , Matthew Green , Jiadong Li , Rhys Seeburger This is my paper

Pith reviewed 2026-05-18 08:46 UTC · model grok-4.3

classification 🌌 astro-ph.SR
keywords dormant black holesGaia DR3binary starsred giantscompact objectsastrometric detectionradial velocity scatterneutron stars
0
0 comments X

The pith

A forward-modelling method using Gaia DR3 summary statistics identifies 389 red giant systems with companions heavier than 3 solar masses as dormant black hole or neutron star candidates.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops an approach to detect dormant black holes and neutron stars in binaries by fitting models to Gaia summary data instead of waiting for full orbital solutions. It simulates the expected re-normalised unit weight error and radial velocity scatter for a given companion mass and period, then uses MCMC to infer the most likely values while accounting for other orbital details. When applied to three million giant stars the method selects 389 objects whose best-fit companions exceed three solar masses. Recovery tests show it recovers known compact-object binaries and finds more systems at periods shorter than one year or longer than roughly six years than the standard Gaia catalogue does. These objects become concrete targets for spectroscopy or later Gaia releases to confirm the presence of compact objects.

Core claim

By simulating Gaia observables, in particular the re-normalised unit weight error and radial velocity scatter, we infer posterior distributions for companion mass and orbital period via MCMC sampling, marginalising over nuisance orbital parameters. We applied the method to three million giants and identify 389 systems with best-fit companion masses ≳ 3 M_⊙. Recovery simulations suggest our selection method is substantially more sensitive than the DR3 non-single-star catalogue, particularly for binaries with periods below 1 year and above ∼ 6 years.

What carries the argument

Forward-modelling framework that simulates Gaia observables (ruwe and RV scatter) to infer posterior distributions for companion mass and orbital period via MCMC sampling while marginalising over nuisance orbital parameters.

If this is right

  • Red-giant primaries yield cleaner signals because their light centroid is more likely dominated by a single component and their RV scatter is more reliable.
  • The method recovers known compact-object binaries and is more sensitive than the published non-single-star catalogue for periods below one year and above about six years.
  • The 389 giant candidates with companions above three solar masses become immediate targets for spectroscopic follow-up and Gaia DR4 analysis.
  • Main-sequence stars produce more confounding signatures so their 279 candidates are presented only as an appendix.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If a substantial fraction of the 389 candidates are confirmed, the local census of dormant black-hole binaries would increase by an order of magnitude, tightening constraints on formation channels.
  • Applying the same summary-statistic pipeline to Gaia DR4 astrometry and radial velocities could enlarge the sample further without requiring full orbit solutions for every system.
  • The technique could be adapted to search for other hidden massive objects, such as white dwarfs or low-mass black holes, in larger stellar samples.

Load-bearing premise

The observed ruwe and RV scatter are produced by orbital motion around one compact companion whose light centroid is dominated by the red giant primary with no significant contamination from other stars.

What would settle it

Spectroscopic radial-velocity monitoring that reveals the companion mass to be below 3 solar masses or shows the system to be a triple with a luminous inner binary would falsify the compact-object interpretation for that candidate.

read the original abstract

We present a rigorous identification of candidates for dormant black holes (BHs) and neutron stars (NSs) in binaries using summary statistics from Gaia DR3, rather than full orbital solutions. Although Gaia astrometric orbits have already revealed a small sample of compact object binaries, many systems remain undetected due to stringent quality cuts imposed on the published orbits. Using a forward-modelling framework that simulates Gaia observables, in particular the re-normalised unit weight error (ruwe) and radial velocity (RV) scatter, we infer posterior distributions for companion mass and orbital period via MCMC sampling, marginalising over nuisance orbital parameters. We validate our approach by comparing the predicted masses and periods against full orbit solutions from DR3, and by successfully recovering known compact object binaries as promising candidates. The method is best suited for systems with red giant primaries, which have more reliable Gaia RV scatter and a light centroid more likely dominated by one component, compared to main-sequence stars, and they are less likely to be triples with short-period inner binaries, which produce confounding signatures. We applied the method to three million giants and identify 389 systems with best-fit companion masses $\gtrsim 3\,M_\odot$. Recovery simulations suggest our selection method is substantially more sensitive than the DR3 non-single-star catalogue, particularly for binaries with periods below 1 year and above $\sim 6$ years. These candidates represent promising targets for spectroscopic follow-up and Gaia DR4 analysis to confirm the presence of compact objects. Candidate main-sequence stars with massive companions face a larger set of confounding effects. Therefore, we present an analogous catalogue of 279 additional main-sequence candidates only as an appendix.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper introduces a forward-modeling MCMC framework that uses Gaia DR3 summary statistics (ruwe and RV scatter) to infer companion mass and orbital period for potential dormant black-hole and neutron-star binaries, focusing on red-giant primaries. Validation is performed against published full-orbit solutions and known compact-object systems; the method is then applied to three million giants, yielding 389 systems with best-fit companion masses ≳ 3 M_⊙. Recovery simulations indicate substantially higher sensitivity than the DR3 non-single-star catalogue, especially for periods below 1 yr and above ∼6 yr. An appendix supplies an analogous but less secure list of 279 main-sequence candidates.

Significance. If the contamination rate from unresolved triples and intrinsic variability can be shown to be low, the work would meaningfully enlarge the known population of dormant compact-object binaries and supply concrete targets for spectroscopic follow-up and Gaia DR4 orbit solutions. The explicit validation against independent full-orbit solutions and the period-dependent recovery simulations are clear strengths that support the claimed sensitivity gain.

major comments (2)
  1. [§3] §3 (likelihood and nuisance parameters): the forward model maps all excess ruwe and RV scatter to orbital motion around a single compact companion but includes no nuisance parameters or marginalization for short-period inner binaries in hierarchical triples or for red-giant jitter; this assumption is load-bearing for the reliability of the 389 high-mass candidates.
  2. [§5] §5 (recovery simulations): the simulations quantify sensitivity relative to the DR3 NSS catalogue but do not inject populations of hierarchical triples or variable giants to measure the false-positive rate in the M_comp ≳ 3 M_⊙ tail; without this test the claimed sensitivity advantage for P < 1 yr and P > 6 yr remains incompletely validated.
minor comments (2)
  1. [Abstract] The abstract states that the method is 'best suited' for red giants but does not quote the quantitative priors or systematic error budget used in the MCMC; adding one sentence with these details would improve reproducibility.
  2. [Results] Table 1 (or equivalent summary table of candidates) should include the median posterior period and its 16–84 percentile range for each of the 389 systems to allow direct comparison with future DR4 orbits.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive review and for highlighting the strengths of our validation approach. We respond point-by-point to the major comments below.

read point-by-point responses

  1. Referee: [§3] §3 (likelihood and nuisance parameters): the forward model maps all excess ruwe and RV scatter to orbital motion around a single compact companion but includes no nuisance parameters or marginalization for short-period inner binaries in hierarchical triples or for red-giant jitter; this assumption is load-bearing for the reliability of the 389 high-mass candidates.

    Authors: We selected red-giant primaries in part because they are less likely to be members of hierarchical triples with short-period inner binaries that would produce confounding signals, and because their RV scatter is more reliably measured than for main-sequence stars. The forward model therefore focuses on a single-companion interpretation for this population. While we do not introduce additional nuisance parameters for jitter or triples (to keep the MCMC tractable), the method was validated against published full-orbit solutions and known compact-object systems. In revision we will expand the discussion in §3 to more explicitly address the impact of this modeling choice on the reliability of the high-mass tail. revision: partial

  2. Referee: [§5] §5 (recovery simulations): the simulations quantify sensitivity relative to the DR3 NSS catalogue but do not inject populations of hierarchical triples or variable giants to measure the false-positive rate in the M_comp ≳ 3 M_⊙ tail; without this test the claimed sensitivity advantage for P < 1 yr and P > 6 yr remains incompletely validated.

    Authors: The recovery simulations were constructed to measure detection efficiency for injected true signals and thereby demonstrate the sensitivity gain relative to the DR3 NSS catalog, particularly at short and long periods. We acknowledge that a direct measurement of the false-positive rate would require separate injection campaigns of hierarchical triples and variable giants. This was outside the scope of the present work. In the revised manuscript we will add an explicit discussion of this limitation, including a qualitative assessment of contamination risks for the red-giant sample and a statement that the 389 candidates are targets for spectroscopic and DR4 follow-up rather than confirmed detections. revision: partial

Circularity Check

0 steps flagged

No circularity: forward-model MCMC inference against external Gaia data is self-contained

full rationale

The paper derives companion-mass posteriors by forward-simulating ruwe and RV scatter observables and sampling them with MCMC against measured Gaia summary statistics, marginalizing nuisance orbital parameters. Validation compares inferred masses/periods to independent DR3 full-orbit solutions and recovers known compact-object binaries. Application to three million giants yields 389 candidates with best-fit M_comp ≳ 3 M_⊙; sensitivity is assessed via separate recovery simulations. No equation or step reduces a claimed prediction to a fitted input by construction, no self-citation is load-bearing for the central inference, and no ansatz or uniqueness theorem is imported from prior author work. The chain is externally anchored and does not collapse to its inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on domain assumptions about the suitability of red-giant primaries and the absence of confounding triple configurations rather than on new physical entities or large numbers of free parameters.

axioms (2)
  • domain assumption Red-giant primaries have reliable Gaia RV scatter and a light centroid dominated by one component
    Explicitly stated as the reason the method works best for giants.
  • domain assumption The systems are not triples containing short-period inner binaries that produce confounding signatures
    Invoked to justify applicability to the giant sample.

pith-pipeline@v0.9.0 · 5858 in / 1497 out tokens · 46777 ms · 2026-05-18T08:46:49.843968+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. What You Don't Know Won't Hurt You: Self-Consistent Hierarchical Inference with Unknown Follow-up Selection Strategies

    astro-ph.IM 2026-05 unverdicted novelty 6.0

    Hierarchical Bayesian inference allows accurate recovery of intrinsic astrophysical source populations even when follow-up selection is unknown and correlated with parameters of interest.

  2. An astrometric search for planets in debris disk systems

    astro-ph.EP 2026-04 unverdicted novelty 4.0

    Gaia astrometric quality metrics and a machine-learning classifier trained on known exoplanet hosts identify candidate stars with debris disks likely to host undetected planets.

  3. SED and Galactic kinematic diagnostics for dormant BH/NS binary candidates

    astro-ph.SR 2026-01 unverdicted novelty 4.0

    SED fitting plus kinematic diagnostics identify 182 top dormant BH/NS binary candidates from Gaia DR3, with 19 likely black hole systems having companion masses >=3 solar masses.