{"id":"7e409308-a8bd-4157-b81d-067a4f8efbf9","arxiv_id":"2509.09165","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Merger gravitational wave signals contain a 'direct wave' from the plunging companions, screened by the remnant's potential, with frequency near the superradiant value for high-spin remnants and SNR above 10 in GW150914-like events.","lead":"Researchers identify a new component in black hole merger gravitational waves, the direct wave, emitted as the black holes crash together, and show it may be detectable with current detectors. The result suggests ringdown analyses should include this signal, and it offers a new way to probe the spin and frame dragging of the remnant.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"SXS:BBH:0305 direct-wave identification is not yet cleanly separated from QNM-filter leakage or time-domain ringing; the Table I SNR inherits this uncertainty.","rationale":"The paper's EMRI analysis is strong: the saddle-point derivation and agreement with Eq. (5)/(7) provide independent support for the existence of direct waves in black-hole perturbation theory, and the masking study (Fig. 6) localizes the emission near the light ring. But the most consequential claim—that direct waves are a robust observable in current GW events—depends on the comparable-mass demonstration. That demonstration uses one SXS waveform, one assumed time alignment, and a filter residual with no quantitative validation. The reader already marked the report CONDITIONAL, and my concern is essentially the same weakest assumption, sharpened: the filter's all-pass magnitude means the failure mode is not suppression but incomplete mode removal and time-domain ringing, which can produce exactly the kind of residual shown. This does not invalidate the EMRI results or the theoretical framework, and it is an addressable verification issue rather than a demonstrated error, so I do not move the verdict away from CONDITIONAL. The proposed QNM-only mock directly tests whether the Fig. 5 residual is a physical direct wave or a filter artifact. Eq. (9) is also stated without proof and is not obviously implied by the two factors in Eq. (2), but it is secondary to the SNR claim.","tokens_in":13294,"tokens_out":9991,"duration_ms":128604,"concrete_test":"Construct a QNM-only mock: fit the post-peak SXS:BBH:0305 strain with the same mode set (N_p=7, N_r=2, plus (3,2,0)) in a window where QNM fits are reliable, extend it backward through merger with a taper, and apply the identical rational filter. If the filtered mock's residual has instantaneous frequency within ~0.1 mΩ_H of the black curve in Fig. 5, or an Eq. (13) SNR exceeding a few percent of Table I, then the claimed direct wave is consistent with filter leakage rather than prompt emission. As a complementary robustness check, repeat the extraction with N_p=10, N_r=4 and with an explicit quadratic-QNM filter at 2ω_220; the direct-wave residual and SNR should be unchanged if the interpretation is correct.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing step is the extraction of the claimed direct wave from the comparable-mass NR waveform SXS:BBH:0305 (Fig. 5), because the headline detectability statement (Table I, network SNR can exceed ~10) is computed from that residual via Eqs. (12)-(13). The rational QNM filter (SM §3, Eqs. 26-28) is all-pass in magnitude on the real axis, so the formal failure mode is not 'amplitude suppression'; the real risk is incompleteness and time-domain ringing. F_tot has zeros only at the chosen QNM frequencies (N_p=7, N_r=2, plus one (3,2,0) mixing mode). Any unmodeled QNM—higher or retrograde overtones, other spherical-spheroidal mixings, or quadratic QNMs from comparable-mass nonlinearities—passes through and appears as a residual, and its instantaneous frequency can sweep near mΩ_H by chance. Conversely, each zero introduces a phase feature that can ring in the time domain. The only NR-vs-theory check is qualitative: the SXS strain peak is assumed to correspond to the prograde light-ring crossing (Fig. 5 caption), and no mismatch statistic or sensitivity to N_p, N_r is reported. Therefore the residual shown as 'direct wave' is not yet cleanly separated from filter/systematics, and the SNR numbers inherit that uncertainty.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper argues that post-merger gravitational waves from binary black hole coalescence contain a distinct 'direct wave' component—prompt radiation emitted as the companion plunges inside the light ring and ergosphere—that is not a quasinormal mode. Using Teukolsky perturbation theory and a saddle-point approximation, the authors derive a time-dependent complex frequency ω_G(t) for this component and argue that the screening factor D̂ vanishes at ω = mΩ_H − i nκ, so that for high spins the direct wave is modulated but not suppressed, oscillating near the superradiant frequency. They identify this component in EMRI waveforms from Sasaki–Nakamura integrations and in the comparable-mass SXS:BBH:0305 waveform after applying rational QNM filters, and they estimate its detectability in a GW150914-like event, finding SNR values that can exceed 10 with current detectors. The paper concludes that direct waves should be included in ringdown spectroscopy and provide a probe of the near-horizon/ergosphere region.","tokens_in":13776,"tokens_out":6693,"duration_ms":76553,"significance":"If the identification is correct, this is a substantive contribution: it extends the notion of horizon/redshift modes, provides a concrete non-QNM component of merger-ringdown signals, and makes a falsifiable detectability claim for LVK O4/O5-era observations. The saddle-point derivation (Eqs. 4–7) is transparent and the EMRI simulation provides a useful self-consistency test of the analytic frequency evolution. The paper also makes its simulation data available, which is a strength. However, the central claim for comparable-mass binaries rests on one NR waveform and on a specific QNM-filter choice, and the SNR estimates inherit these systematics; the current draft does not yet supply the quantitative validation needed to establish the direct wave as a robust observable.","major_comments":[{"comment":"The identification of the direct wave in SXS:BBH:0305 is obtained by applying QNM rational filters with N_p=7, N_r=2 plus one spherical-spheroidal mixing mode. This filter has zeros only at the selected QNM frequencies; any unmodeled QNM—higher or retrograde overtones, additional ℓ−m mixing modes, or quadratic QNMs—passes through and appears in the residual. Each zero can also introduce time-domain ringing. No sensitivity of the residual to N_p, N_r, or to the inclusion of additional modes is reported. Since the subsequent SNR estimates (Table I) are computed from this filtered residual, the direct-wave identification is not yet cleanly separated from filter systematics.","section":"SM §3, Eqs. (26)–(28) and Fig. 5"},{"comment":"The comparable-mass validation relies on a single SXS waveform and on the assumption that the SXS strain peak corresponds to the particle crossing the prograde light ring. The agreement with the EMRI prediction is described only qualitatively ('overall good'), with no mismatch statistic, no alignment uncertainty, and no test of the sensitivity to the peak-to-light-ring mapping. This assumption is load-bearing because it sets the time origin and normalization used in the SNR estimate. A quantitative comparison (e.g., residual-norm mismatch, alignment variation, multiple SXS cases) is needed before the claimed detectability can be accepted.","section":"Fig. 5 and Table I"},{"comment":"The statement that D̂_ℓmω = 0 at ω = mΩ_H − i nκ for all integer n is asserted without derivation ('Here we further extend this result... as illustrated in Fig. 2'). This screening property is central to the argument that horizon modes are eliminated while the direct wave survives through the time-dependent ω_G. A derivation or a precise reference should be provided; otherwise the theoretical foundation for the greybody modulation is incomplete.","section":"Eq. (9)"},{"comment":"The EMRI 'prediction' and the EMRI simulation both come from the same linearized Teukolsky/Sasaki–Nakamura formalism, so their agreement validates the saddle-point evaluation but is not an independent confirmation that the direct wave exists in comparable-mass mergers. The paper should state this limitation explicitly and place the weight of the physical claim on the SXS comparison, which currently has the uncertainties noted above.","section":"Eqs. (2)–(5) and Fig. 3"},{"comment":"The quoted network SNRs for O4 do not follow from Table I. The text states that the O4 three-detector network reaches SNR 11.0 at −5M_t and 29.4 at −10M_t, but the Table entries give O4 L+V network SNRs of sqrt(7.1²+4.5²)=8.4 at −5M_t and sqrt(19.0²+11.9²)=22.4 at −10M_t. Please reconcile the quoted numbers. In addition, the SNR estimates have no error bars; they should include noise-only variance and systematic uncertainties from the filter choice, time alignment, and the use of a single NR waveform.","section":"Table I and §Detectability"}],"minor_comments":[{"comment":"The phrase 'filter out all possible QNMs' is stronger than what is demonstrated by the finite N_p, N_r filter. Please soften it to reflect the finite set of modes actually removed.","section":"Introduction / Fig. 5 caption"},{"comment":"The time-domain SNR formula is introduced as Eq. (12) with the inner product in Eq. (13), but the text refers to 'Eq. (12)' when meaning the pair. Please clarify the numbering and state explicitly how multiple detectors are combined to form the network SNR.","section":"Eq. (12)–(13)"},{"comment":"The color-scale and marker legend are hard to read in the contour plot; please enlarge and ensure that the red dots and triangles are distinguishable in print.","section":"Fig. 2"},{"comment":"The data availability statement cites reference [74] with no DOI or persistent identifier; please provide a stable link or DOI so the simulation data can be retrieved.","section":"Data availability"}],"recommendation":"major_revision","confidential_remarks":"This is a thought-provoking paper with a clear potential impact on ringdown spectroscopy. The theoretical derivation for EMRIs is coherent, and the public data release is commendable. However, the comparable-mass claim and the detectability numbers depend on a single NR waveform, a particular QNM-filter setup, and an assumed time alignment; none of these are quantitatively validated at present. The inconsistency between the quoted O4 network SNR and Table I is a concrete issue that must be fixed. I believe these problems are addressable within the scope of a revision, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First, what to know: this is the most serious treatment of the direct wave/horizon-mode story since Mino-Brink and Zimmerman-Chen. The saddle-point derivation in Eqs. (4)-(7) is clean, and the EMRI simulations track the analytic instantaneous frequency well across spins (Figs. 3, 4). The masking/localization calculation is a genuinely nice diagnostic. But the headline detectability claim does not stand on the evidence yet.\n\nThe genuinely new pieces are the time-dependent saddle-point description, the generalized screening zeros in Eq. (9), the first NR comparison, and the SNR estimates. The EMRI agreement is real but partly circular—both sides come from the same linearized Teukolsky/Sasaki-Nakamura machinery, so it mostly checks the saddle-point approximation. The SXS comparison is the more independent benchmark, and it is underpowered, exactly as the stress-test says: one waveform, assumed peak-to-light-ring alignment, and a QNM rational filter whose completeness is not probed. Unmodeled overtones or quadratic QNMs would leak through and produce a residual with time-dependent frequency; the filter zeros can also ring. No mismatch statistic, no variation of N_p,N_r, no error bars on Table I. The abstract's 'dominated' claim is stronger than the fractional residual shown in Fig. 5. Data availability is cited, but there is no code or persistent identifier in the text.\n\nThat said, these are fixable weaknesses, not fatal ones. The core picture—a prompt, time-dependent component sourced near the light ring and partially screened by the potential barrier—is coherent and consistent with the earlier horizon-mode literature. The paper is honest about the SXS discrepancy and about the limits of linear theory. The generalized screening statement, if correct, is a useful extension.\n\nThis is for the ringdown/spectroscopy audience: anyone fitting overtones or interpreting merger waveforms should know this component exists. I would send it to peer review. A referee should ask for a derivation of Eq. (9), a filter-sensitivity and alignment study across several SXS waveforms, and error bars on the SNR. With those, it becomes a solid reference; without them, the detectability claim should stay tentative.","headline":"A credible analytic extension of horizon-mode theory with a clean EMRI self-check, but the SXS residual and SNR estimate are not yet clean enough to support the O4 detectability claim.","tokens_in":14147,"tokens_out":5714,"would_cite":true,"duration_ms":66403,"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":"Merger gravitational waves are dominated by a prompt 'direct wave' from the plunging companion; for rapidly spinning remnants it oscillates near the superradiant frequency and is detectable in current events.","keywords":["direct wave","black hole ringdown","quasinormal modes","Kerr black holes","merger gravitational waves","superradiant frequency","frame dragging","black hole spectroscopy"],"falsifier":"Construct a synthetic gravitational-wave signal composed only of known quasinormal modes with fixed amplitudes, apply the same rational QNM filters, and check whether the filtered residual is exactly zero. If a spurious residual survives, the filter does not actually isolate direct waves. Alternatively, for a high-spin comparable-mass waveform, decompose the filtered residual and check whether its instantaneous complex frequency follows the predicted ω_G(t) rather than a sum of two QNM frequencies; the latter would indicate nonlinear mode-mixing rather than a direct wave.","tokens_in":13218,"feed_emoji":"🌊","tokens_out":8692,"duration_ms":88055,"temperature":0.7,"pith_summary":"Black hole ringdowns are usually described as a sum of quasinormal-mode (QNM) oscillations. This paper argues that the gravitational wave signal around the merger peak is instead dominated by a different component, the 'direct wave': radiation emitted promptly as the companion crosses the light ring and spirals into the ergosphere. Using black-hole perturbation theory and numerical-relativity waveforms, the authors show that after removing all QNMs with rational filters, a residual remains whose instantaneous frequency and decay rate match a saddle-point formula derived from the perturbation wave equation. For high-spin remnants (dimensionless spin ≥ 0.7), this direct wave locks onto a quasi-stable frequency near the superradiant frequency, reflecting frame dragging. The authors estimate that in a GW150914-like event the direct wave alone would have a signal-to-noise ratio above 10 with the current ground-based detector network, so it is a realistic observable that must be included in black hole spectroscopy.","feed_headline":"Direct wave in merger ringdowns is detectable at SNR above 10","feed_subtitle":"The plunge emission oscillates near the superradiant frequency and cannot be ignored in black hole spectroscopy.","key_machinery":"The central object is the direct wave and its saddle-point construction. From the frequency-domain wave equation for perturbations of a spinning black hole, the waveform at infinity is written as an integral over the particle trajectory. A two-variable steepest-descent evaluation in time and frequency yields the instantaneous complex frequency ω_G(t) of the direct wave and a contribution proportional to the greybody factor D̂ℓmω evaluated at ω_G(t). This factor vanishes at the horizon-mode frequencies mΩ_H - i nκ, explaining why the naive horizon mode is screened while the transient plunge radiation survives. The second piece of machinery is the QNM rational filter, which removes a chosen se","core_discovery":"The paper's central claim is that merger-stage gravitational waves from binary black holes are dominated by a prompt, non-QNM component they call the direct wave. In black-hole perturbation theory, a plunging compact object emits radiation whose saddle-point contribution is localized along the retarded-time–emission-time map u = t - x(t); the complex instantaneous frequency ω_G(t) = m Ω̂ - i ĝ is set by the local orbital frequency and plunging velocity. The traditional 'horizon mode' proposed in earlier work would oscillate at ω = mΩ_H - iκ, but the paper shows the transmission factor (the greybody factor) vanishes at all these horizon-mode frequencies, ω = mΩ_H - i nκ, so that mode is scree","pith_inferences":["If the direct wave is a generic feature, early-ringdown tests of general relativity that assume a pure QNM signal may be systematically biased; re-analyzing existing event residuals after QNM filtering could reveal prompt-emission excess power near the superradiant frequency.","The screening mechanism means the direct wave's amplitude and phase carry information about the near-horizon transmission (greybody) factor; deviations from the predicted D̂ℓmω would point to modified near-horizon structure.","The close match between an extreme-mass-ratio plunge calculation and a comparable-mass numerical waveform suggests a simple geodesic-plunge template could serve as a merger model for high-spin, unequal-mass binaries, potentially reducing systematics in parameter estimation."],"forward_implications":["QNM-based ringdown fits that start at or shortly after the strain peak will inadvertently absorb the direct wave into overtones, biasing the inferred remnant mass and spin; analyses must either start later or include a direct-wave template.","For high-spin remnants, the measured quasi-stable frequency near the superradiant frequency gives a direct, near-universal probe of frame dragging in the ergosphere.","The paper's screening result implies the previously proposed horizon mode is not observable in ordinary Kerr perturbations; any detected mode exactly at a horizon-mode frequency would indicate physics beyond the standard picture.","Direct waves contribute to the post-peak signal at SNR above 10 with current ground-based detectors and much more with next-generation instruments, making them a new target for gravitational-wave searches and tests of black hole dynamics."],"fun_headline_variants":["Direct wave in merger ringdowns now visible at SNR above 10","Plunge emission rings near superradiant frequency, detectable now","Merger ringdowns dominated by direct wave, not quasinormal modes","Direct wave from binary coalescence yields SNR over 10","New signal in ringdown: direct wave with SNR exceeding 10"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The existence and detectability of the direct wave rests on the assumption that the rational QNM filters (seven prograde and two retrograde overtones plus a mixing mode) remove all quasinormal content without altering the direct wave, and that the strain peak in the comparable-mass waveform corresponds to the prograde light-ring crossing; if either is wrong, the residual could be an artifact or mislocated.","fun_headline_variants_meta":{"raw":{"variants":["Direct wave in merger ringdowns now visible at SNR above 10","Plunge emission rings near superradiant frequency, detectable now","Merger ringdowns dominated by direct wave, not quasinormal modes","Direct wave from binary coalescence yields SNR over 10","New signal in ringdown: direct wave with SNR exceeding 10"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000489,"raw_usage":{"total_tokens":2232,"prompt_tokens":718,"completion_tokens":1514,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":462,"completion_tokens_details":{"reasoning_tokens":1423}},"tokens_in":462,"tokens_out":1514,"duration_ms":14591,"temperature":1.0,"reasoning_tokens":1423,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T19:33:50.465614+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Construct a synthetic gravitational-wave signal composed only of known quasinormal modes with fixed amplitudes, apply the same rational QNM filters, and check whether the filtered residual is exactly zero. If a spurious residual survives, the filter does not actually isolate direct waves. Alternatively, for a high-spin comparable-mass waveform, decompose the filtered residual and check whether its instantaneous complex frequency follows the predicted ω_G(t) rather than a sum of two QNM frequencies; the latter would indicate nonlinear mode-mixing rather than a direct wave.","supporting_citations":[],"review_version":1}