{"id":"c67b35ee-bb62-4a73-87fb-c61b9e69390b","arxiv_id":"2504.16457","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":3,"one_line_summary":"Black hole flybys in AGN disks may produce gravitational-wave bremsstrahlung at a fiducial rate of 3.2 Gpc^-3 yr^-1, with a plausible range of 0.08 to 1194 Gpc^-3 yr^-1.","lead":"This paper simulates stellar-mass black holes scattering inside the gas disks around supermassive black holes and estimates how often their close flybys emit gravitational waves that LIGO could detect. The headline rate, around 3 events per cubic gigaparsec per year, is an order-of-magnitude guess with a huge uncertainty range and several unverified assumptions.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed 3.2 Gpc^-3 yr^-1 rate is not established: the fBH,enc = 10^3 fbrem,BH scaling can exceed unity, and plugging the stated fbrem into Eq. 9 disagrees with Table 2.","rationale":"The paper's strongest claim is the event rate for detectable GW bremsstrahlung from sBH-sBH scattering in AGN disks. The reader's verdict is CONDITIONAL, and my independent read locates the same weak point: the conversion from simulated per-encounter probability fbrem,BH to the population-level fraction fBH,enc is an unvalidated factor-of-10^3 guess that is not derived from the simulations or from data. I agree with the reader that this is the most load-bearing assumption because the fiducial 3.2 Gpc^-3 yr^-1 and the full rate range in Table 2 scale linearly with fBH,enc; if fBH,enc is wrong by an order of magnitude, the headline claim changes by an order of magnitude. Moreover, the text is internally inconsistent: Eq. 9 is normalized with fBH,enc/0.5, while Section 4 derives fBH,enc ~ 10^3 fbrem,BH ~ 3.9 for the drag prescription, which would make the fiducial rate roughly seven times larger than reported if Eq. 9 were evaluated consistently. I do not treat this as evidence of any impropriety; the simulations and the MMR criterion appear to be sound, and the discrepancy can be fixed with a renormalized or explicitly bounded encounter scaling. I also note the manuscript's own limitation in Section 5.1, where the 'first-principle' waveforms are acknowledged to be a quadrupole spline approximation; that supports the reader's decision not to rely on the templates for the central rate claim. The reader's CONDITIONAL verdict remains appropriate: the quantitative rate needs correction, but the underlying parameter study and the analytical MMR criterion are valuable and can survive revision. No verdict change is needed.","tokens_in":24066,"tokens_out":5166,"duration_ms":50109,"concrete_test":"Recompute the Section 4 rate using the paper's own stated assumptions: set fBH,enc = 10^3 fbrem,BH with fbrem,drag = 0.0039 and fbrem,trap = 0.0056 in Eq. 9, and compare to the 'Expected Rate' column of Table 2. Also re-derive fBH,enc directly: 2e4 x 10% x 50% = 10^3 is the total number of interacting sBHs in one disk, not encounters per BH. If fBH,enc exceeds unity, or if the recomputed rate differs from Table 2 by more than 10%, the headline rate must be renormalized and the encounter scaling replaced by the simulation's actual per-BH encounter multiplicity distribution.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline rate rests on the Section 4 scaling fBH,enc ~ 10^3 fbrem,BH. This factor is asserted from NBH,NSC ~ 2e4, '10% in disk', and '50% interact', then interpreted as ~10^3 encounters per inner BH. The arithmetic actually gives 10^3 total interacting sBHs in the disk, not an encounter multiplicity per BH, and for the simulated fbrem,drag = 0.39% it yields fBH,enc ~ 3.9, a dimensionless 'fraction' greater than unity. Eq. 9 is normalized with fBH,enc/0.5; using fBH,enc ~ 3.9 in that equation gives a drag >1 Msun rate near 23 Gpc^-3 yr^-1, not the 3.2 Gpc^-3 yr^-1 reported in Table 2. The stated rate range 0.08-1194 Gpc^-3 yr^-1 therefore inherits an unvalidated and internally inconsistent normalization. Separately, the paper itself concedes in Section 5.1 that the 'first-principle' templates are a simplified quadrupole spline approximation with acknowledged error sources, so the rate, not the template, is the load-bearing quantitative claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies gravitational-wave (GW) bremsstrahlung from hyperbolic encounters of stellar-mass black holes (sBHs) embedded in AGN disks. The authors use N-body simulations with gas drag and migration-trap prescriptions, including 1PN and 2.5PN corrections, to characterize how close-encounter rates and GW emission depend on SMBH mass, migration rate, mutual inclination, and sBH masses. They identify first-order mean-motion resonances as a key suppression mechanism and derive an analytical criterion for resonance capture. They then run 80,000 population-synthesis simulations and convert the simulated bremsstrahlung fraction into a volumetric detection rate, quoting a fiducial rate of 3.2 Gpc^-3 yr^-1 and a range of 0.08-1194 Gpc^-3 yr^-1. Finally, they construct approximate GW templates from selected trajectories and evaluate their SNR against LIGO A+ sensitivity.","tokens_in":24216,"tokens_out":8582,"duration_ms":76926,"significance":"If the rate estimate were reliable, the paper would identify a new, potentially detectable GW burst channel from AGN disks and provide motivation for unmodeled burst searches. The parameter study is systematic and valuable: the finding that 1PN precession suppresses close encounters relative to 2.5PN-only runs is a useful correction to earlier treatments, and the mean-motion-resonance criterion (Eq. 7) is checked against simulations and works well. The paper is also transparent about several limitations, including simplified migration forces and the exclusion of disk turbulence. However, the headline rate is not established: the conversion from the simulated bremsstrahlung fraction to the per-BH encounter fraction used in the rate integral is internally inconsistent, the normalization of Eq. (9) is asserted without derivation, and the detection proxy used in the rate calculation is not a validated LIGO criterion. These issues affect the central quantitative claim of the paper.","major_comments":[{"comment":"The derivation of fBH,enc is internally inconsistent and not reproducible. The text states that NBH,NSC = 2e4, that 10% of these BHs are in the disk, and that 50% interact, concluding that 'each inner black hole will have ~10^3 encounters' and hence fBH,enc ~ 10^3 fbrem,BH. The arithmetic actually yields 10^3 total interacting BHs in the disk, not an encounter multiplicity per BH. Using the reported fbrem,drag(>1 Msun) = 0.0039 gives fBH,enc = 3.9, which is unphysical for a fraction. Inserting fBH,enc = 3.9 into Eq. (9) gives a fiducial rate of roughly 23 Gpc^-3 yr^-1, not the 3.2 Gpc^-3 yr^-1 reported in Table 2. The tabulated rate corresponds to fBH,enc ~ 0.5, indicating that the stated 10^3 scaling has not actually been applied. If the scaling is applied correctly via fBH,enc = 1 - exp(-N_enc fbrem,BH) with N_enc = 10^3, the fiducial rate would be about 5.9 Gpc^-3 yr^-1. The headline rate must be reconciled with a consistent, physically meaningful definition of fBH,enc.","section":"Section 4, Eq. (9) and following paragraph"},{"comment":"The normalization constant '3 Gpc^-3 yr^-1' in Eq. (9) is asserted without derivation. The text says the calculation follows Tagawa et al. (2020), but the cited work gives a broad merger-rate range (0.02-60 Gpc^-3 yr^-1), and no calculation is shown that turns that information into the point value 3. Because all subsequent rates scale linearly with this constant, the rate estimate is not self-contained. The authors should derive this normalization explicitly from the integral in Eq. (8), using an AGN number density and SMBH mass function, or identify it as a free parameter and propagate it as such.","section":"Section 4, Eq. (9)"},{"comment":"The rates in Table 2 are quoted as 'detectable by LIGO' based on a peak GW energy threshold of >1 Msun or >3 Msun, with the statement 'assuming an observational distance of ~1 Mpc' (Section 3.1). This is not a validated detection criterion: no LIGO sensitivity curve, SNR threshold, or search duty cycle is folded into the population rate. The SNR calculations in Section 5.1 are performed for only four example events and are not used to derive a selection function. Consequently, the abstract's 'rate for ground-based gravitational-wave detections' is stronger than what the calculation actually supports. Either fold an SNR-based selection function into the rate calculation or revise the language to 'events emitting more than X Msun in GWs' rather than 'detectable events.'","section":"Sections 3.1 and 4"}],"minor_comments":[{"comment":"The sentence 'each inner black hole will have ~10^3 encounters' is an unsupported interpretation; the preceding numbers give 10^3 total interacting BHs, not an encounter multiplicity per BH, and the encounter multiplicity is a separate quantity that should be modeled or measured.","section":"Section 4, paragraph after Eq. (9)"},{"comment":"The term 'first-principle gravitational wave templates' overstates the method, since the paper itself lists the quadrupole spline approximation as a source of error; 'approximate quadrupole templates' would be more accurate.","section":"Section 5.1"},{"comment":"The close-encounter criterion is described as 'within 100Rg,1' in Section 2.3 but as 'within 50 times the sum of their gravitational radii' in Section 3.1; please specify which criterion is used for the population synthesis and the rates in Table 2.","section":"Section 3.1"},{"comment":"The bibliography contains malformed or duplicated entries, including 'et al., B. P. A. 2016' and two entries for Abbott et al. 2016 with different DOIs; these need correction in production.","section":"References"},{"comment":"The sentence 'most of the gap time center around ~0.5 sec' should be reworded, e.g., 'most time gaps are centered around ~0.5 s.'","section":"Figure 18"}],"recommendation":"major_revision","confidential_remarks":"The headline rate is the paper's main advertised result, and the internal inconsistency in Section 4 undermines it. The dynamical study and MMR analysis are solid and could support a revised paper, but the rate calculation and its detection claims need to be redone before the manuscript can be considered for publication. The paper's scope is somewhat broad for a letter; a full-length article with a corrected rate derivation and a proper detection criterion would be more appropriate."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The genuinely useful part is the simulation campaign and the resonance-capture criterion. The paper shows, with a decent parameter sweep and 80k runs, that 1-PN precession suppresses close encounters relative to 2.5-PN-only treatments, and that lower-mass SMBHs produce more detectable flybys. The analytic criterion in Eq. 7 for first-order MMR capture, with its mass-dependent alternating encounter/no-encounter structure, checks out against their N-body results. That is a real contribution, consistent with existing literature and not circular. The qualitative story about migration traps and MMR gaps is plausible and worth having.\n\nThe soft spot is the rate estimate in Section 4, and it is load-bearing. The stress-test note is right: the factor of 10^3 is not an encounter multiplicity per black hole; it is ~10^3 total interacting sBHs in the disk, and multiplying the simulated bremsstrahlung fraction 0.39% by 10^3 gives a dimensionless fBH,enc ~ 3.9, which is unphysical. Using that in Eq. 9 does not reproduce Table 2's 3.2 Gpc^-3 yr^-1. The '3 Gpc^-3 yr^-1' normalization in Eq. 9 is asserted, not derived from an integral over the AGN mass function. So the headline rate range 0.08-1194 Gpc^-3 yr^-1 inherits an invalid normalization. The paper also conflates a source rate with a detection rate: the threshold is peak radiated energy, not SNR times a detection volume. That is fixable, but it is not a detail.\n\nThe GW templates are called 'first-principle' but are quadrupole spline approximations, and the paper itself concedes the error sources in Section 5.1. The numerical error estimate O(10^-30) is implausible and should be corrected or deleted. These templates are illustrative, not the main claim, so this is a minor-to-moderate issue.\n\nRecommendation: send it to a serious referee, actually to two. The core dynamics and the MMR criterion justify referee time. The rate section needs to be rewritten or downgraded to an order-of-magnitude source-rate with properly justified normalization and an SNR-based detection step. The authors have the ingredients to fix this. I would not cite the rate as is, but I would cite the resonance criterion.","headline":"The MMR criterion and the simulation campaign are solid and worth referee time, but the headline 3.2 Gpc^-3 yr^-1 rate is not established because the fBH,enc scaling is internally inconsistent.","tokens_in":24932,"tokens_out":2824,"would_cite":true,"duration_ms":27457,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Close hyperbolic encounters of stellar-mass black holes in AGN disks emit gravitational-wave bremsstrahlung at rates ground-based detectors may already be able to see.","keywords":["gravitational-wave bremsstrahlung","hyperbolic black hole encounters","AGN disks","stellar-mass black holes","mean motion resonance","gravitational-wave bursts","population synthesis rates","post-Newtonian dynamics"],"falsifier":"Search the existing public burst data from current ground-based detectors with the paper's template waveforms (Figures 14-17): if the fiducial rate of $3.2\\,\\mathrm{Gpc}^{-3}\\,\\mathrm{yr}^{-1}$ is correct, a handful of such one-off flyby bursts should already be present in the data, whereas a null result with an upper limit below roughly $0.1\\,\\mathrm{Gpc}^{-3}\\,\\mathrm{yr}^{-1}$ would rule out the optimistic half of the claimed range.","tokens_in":23738,"feed_emoji":"🌌","tokens_out":12680,"duration_ms":107222,"temperature":0.7,"pith_summary":"This paper argues that the same active galactic nucleus (AGN) disk environment that funnels stellar-mass black holes into mergers also produces close hyperbolic flybys emitting gravitational-wave bremsstrahlung, and that these flybys are detectable by ground-based interferometers. From scattering simulations with gas drag or migration-trap forces plus first- and 2.5-order post-Newtonian terms, the authors find that detectable bursts are most frequent around supermassive black holes of $10^5$-$10^6\\,M_\\odot$. They estimate a fiducial detection rate of $3.2\\,\\mathrm{Gpc}^{-3}\\,\\mathrm{yr}^{-1}$, with a wide allowed range from $0.08$ to $1194\\,\\mathrm{Gpc}^{-3}\\,\\mathrm{yr}^{-1}$ depending on migration forces, AGN disk properties, and detection thresholds. They also provide first-principle waveform templates whose frequencies fall in or near current ground-based detector bands. If the rate estimate holds, AGN-disk flybys constitute a new gravitational-wave source population distinct from binary mergers.","feed_headline":"Black hole flybys in AGN disks may be a new GW burst source","feed_subtitle":"Migrating black holes in AGN disks could make detectable flyby bursts at ~3 per cubic gigaparsec per year.","key_machinery":"The load-bearing machinery is a direct scattering experiment: two stellar-mass black holes orbit a supermassive black hole, started five mutual Hill radii apart, with the outer black hole migrating inward under either a gas-drag force or a migration-trap force while both feel 1PN and 2.5PN post-Newtonian corrections in an N-body integrator. The key analytic result that organizes the outcomes is the critical total mass for resonance capture, which decides whether the pair is trapped in a low-order mean-motion resonance (quiet) or passes close enough for gravitational-wave emission (loud). The energy radiated in each close passage is computed with the standard quadrupole-approximation formula for hyperbolic encounters, $\\Delta E_{\\rm GW} = (85\\pi/12\\sqrt{2})\\, G^{7/2}\\mu^2 m_{12}^{5/2}/(c^5 r_p^{7/2})$, and the waveforms are produced by spline-differentiating the simulated trajectories in the quadrupole approximation, which yields the burst frequencies and signal-to-noise ratios used in the detectability estimates.","core_discovery":"The central claim is that gravitational-wave bremsstrahlung from hyperbolic encounters between stellar-mass black holes migrating in AGN disks is a real, possibly detectable population. The paper shows that whether such an encounter emits strong gravitational waves is controlled by mean-motion-resonance capture: pairs whose total mass exceeds a critical value, $M_{\\rm tot}=24(f^{2/3}-1)^3 M_{\\rm SMBH}/\\big((f^{2/3}+1)^3 k_{\\rm hill}^3\\big)$ for a $p:q$ resonance (with $f=q/p$ and $k_{\\rm hill}$ the initial separation in mutual Hill radii), are locked into resonance and stay well separated, while pairs below the boundary scatter to periastra of a few gravitational radii and emit up to several solar masses of energy in one burst. Detectable scattering is more frequent around lower-mass SMBHs because the mutual Hill radius is smaller, making close approaches easier; at $10^5\\,M_\\odot$ SMBHs, resonance capture suppresses encounters again, and at higher SMBH masses the larger Hill radius spreads the black holes apart. Combining the simulated burst fraction with a population model of AGN disks and scaling by SMBH mass yields a fiducial rate of $3.2\\,\\mathrm{Gpc}^{-3}\\,\\mathrm{yr}^{-1}$ for encounters emitting more than one solar mass of gravitational-wave energy, with the range $0.08$-$1194\\,\\mathrm{Gpc}^{-3}\\,\\mathrm{yr}^{-1}$. The paper's rate estimate is therefore a statement that AGN-disk flybys could already be hiding in ground-based data.","pith_inferences":["If resonance capture is a real gatekeeper, the chirp-mass distribution of detected AGN-disk bursts should show deficits at total masses corresponding to low-order resonances; looking for a resonant 'comb' in burst masses would test the mechanism independently of the rate.","The same encounters produce repeated bursts with sub-second gaps (centered near $0.03$-$0.04\\,\\mathrm{s}$ for high-energy events), a distinctive time-frequency fingerprint that unmodeled burst searches could use to separate AGN-disk flybys from detector glitches.","Replacing the assumed $10^3$ encounters per black hole with a self-consistent multi-body simulation of the migrating population would either firm up or rescale the rate estimate; this is a testable modeling improvement rather than a prediction.","If the fiducial rate is near correct, AGN-disk flybys would add an impulsive, non-Gaussian foreground to the stochastic gravitational-wave background in the roughly $5$-$100\\,\\mathrm{Hz}$ band, separable from the merger background by its burst character."],"forward_implications":["Detectable bremsstrahlung bursts are most common around SMBHs of $10^5$-$10^6\\,M_\\odot$, so searches should weight low-mass AGN nuclei rather than the most massive ones.","Including 1PN precession suppresses close encounters relative to 2.5PN-only treatments, so earlier scattering studies that omitted 1PN likely overpredict encounter and emission rates.","Resonance capture splits the population by total sBH mass: pairs near 4:3, 5:4, and 6:5 resonances stay quiet, while pairs below the critical mass can produce strong bursts or mergers, producing alternating regions in the mass-ratio plane.","The predicted flyby rate ($\\sim 3\\,\\mathrm{Gpc}^{-3}\\,\\mathrm{yr}^{-1}$ fiducial) is comparable to or higher than the estimated AGN-disk merger rate, making flybys a competitive channel.","Some simulated waveforms peak near $100\\,\\mathrm{Hz}$, within current detector bands, while others peak near $5\\,\\mathrm{Hz}$, making them targets for future low-frequency ground-based detectors."],"supporting_citations":[{"why":"Supplies the gravitational-wave energy-loss formula for hyperbolic encounters that scores each close passage.","marker":"Peters (1964)"},{"why":"Provides the rate-integral framework, AGN disk and nuclear cluster parameters, the initial mass function, and the merger-rate comparison.","marker":"Tagawa et al. (2020)"},{"why":"Shows migrating sBHs in AGN disks can reach arbitrarily small separations and supplies the drag and trap migration prescriptions used here.","marker":"Li et al. (2022)"},{"why":"The representative prior scattering study using only 2.5-PN, which this paper argues overpredicts close-encounter rates.","marker":"Samsing et al. (2022)"},{"why":"Supplies the N-body integration package with the post-Newtonian and migration forces added.","marker":"Rein & Liu (2012)"},{"why":"Provides the 2.5-PN radiation-reaction terms included in the dynamics.","marker":"Blanchet (2006)"},{"why":"Reference quadrupole-approximation hyperbolic waveforms used to validate the new templates.","marker":"Capozziello & De Laurentis (2008)"},{"why":"One of the two existing searches for GW bremsstrahlung used to set the current observational context.","marker":"Morrás et al. (2022)"},{"why":"The other existing search, finding no significant events in O3b, used to argue current non-detections are consistent.","marker":"Bini et al. (2024)"}],"fun_headline_variants":["AGN disk black hole flybys could ring LIGO with bursts","Hyperbolic sBH encounters in AGN disks may emit GW bursts","New GW burst source: black hole scattering in AGN disks","Flyby bursts from AGN disks: a possible LIGO signal"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The rate estimate assumes that each migrating stellar-mass black hole experiences about $10^3$ close interactions with other black holes during the AGN disk lifetime; this factor is an order-of-magnitude guess not derived from the simulations or from data, and when multiplied by the simulated bremsstrahlung fraction it implies an encounter fraction larger than unity, so the normalization of the headline rate is not independently established.","fun_headline_variants_meta":{"raw":{"variants":["AGN disk black hole flybys could ring LIGO with bursts","Hyperbolic sBH encounters in AGN disks may emit GW bursts","New GW burst source: black hole scattering in AGN disks","Flyby bursts from AGN disks: a possible LIGO signal"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000263,"raw_usage":{"total_tokens":1723,"prompt_tokens":1193,"completion_tokens":530,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":809,"completion_tokens_details":{"reasoning_tokens":454}},"tokens_in":809,"tokens_out":530,"duration_ms":4807,"temperature":1.0,"reasoning_tokens":454,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T11:03:25.345502+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Search the existing public burst data from current ground-based detectors with the paper's template waveforms (Figures 14-17): if the fiducial rate of $3.2\\,\\mathrm{Gpc}^{-3}\\,\\mathrm{yr}^{-1}$ is correct, a handful of such one-off flyby bursts should already be present in the data, whereas a null result with an upper limit below roughly $0.1\\,\\mathrm{Gpc}^{-3}\\,\\mathrm{yr}^{-1}$ would rule out the optimistic half of the claimed range.","supporting_citations":[{"cited_title":"2008, Astropart","cited_arxiv_id":null,"evidence_quote":"Reference quadrupole-approximation hyperbolic waveforms used to validate the new templates."}],"review_version":1}