{"id":"7a56c157-634e-4ce3-abbe-606e68202b9b","arxiv_id":"2508.12697","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Using the new TaylorF2Ecck waveform, the authors constrain the 20 Hz orbital eccentricity of GW170817 and GW190425 to be below 0.011 and 0.028 at 90% credibility.","lead":"This paper applies a new gravitational-wave template that includes orbital eccentricity and periastron shift to two neutron star merger signals. It finds no sign of a noticeably eccentric orbit for either event when the waves entered the detectors' sensitive band.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The null e0 bounds rest on a truncated waveform family with no independent validation in the constrained e0<0.03 range; Appendix F shows divergence from pyEFPE even at e0=0, so model error could bias the claimed upper limits.","rationale":"The reader's weakest assumption identifies essentially the same point: TaylorF2Ecck is a truncated leading-order model, validated mainly against its own injections and sibling approximants, so a model error could bias the e0 upper limits. My stress-test sharpens this into a specific, testable risk: the paper's Appendix F provides direct evidence that the model family diverges from an independent eccentric waveform model in the moderate-eccentricity regime, and that the mismatch does not vanish in the circular limit. Since the claimed bounds (e0<0.011 and e0<0.028) lie below the range where any independent validation is reported, the central null result is not yet secured against model systematic error. The proposed injection-recovery test using pyEFPE or the full O(e0^6) model directly probes whether TaylorF2Ecck can recover eccentricity if it is actually present at the levels the paper claims to exclude. If the test passes, the concern is resolved and the central claim stands; if it fails, the quoted upper limits are model artifacts rather than astrophysical constraints. The correct verdict remains CONDITIONAL, as the reader said: the analysis is plausible and well-documented, but this specific validation gap should be closed before the null eccentricity claim is treated as definitive. The paper's other issues (the mislabeled Bayes-factor table, the unbacked abstract Bayes-factor statement, and the extrapolated 3.5PN recommendation) are presentation or secondary matters and do not change this assessment.","tokens_in":37383,"tokens_out":5720,"duration_ms":63432,"concrete_test":"Run zero-noise injection-recovery tests in which pyEFPE generates GW170817-like signals with e0 at 20 Hz equal to 0.005, 0.010, 0.015, 0.020, and 0.025, and TaylorF2Ecck recovers them using the same priors and fmin=20 Hz as in Sec. III. If the injected e0 falls outside the recovered 90% credible interval for any of these values, or if the recovered upper limit is systematically below the injected value, the TaylorF2Ecck truncation is biasing the claimed null result. An acceptable alternative is to use the full O(e0^6) 3PN model of Ref. [72] as the injection family instead of pyEFPE, which avoids the response-function convention mismatch discussed in Appendix F.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that GW170817 and GW190425 have essentially zero initial eccentricity at 20 Hz (e0<0.011 and e0<0.028 at 90%). This inference is only as trustworthy as TaylorF2Ecck's ability to represent eccentric signals in exactly that range. TaylorF2Ecck is a restricted leading-order implementation: O(e0^2) in the Fourier phase, O(e0) in amplitude, and only four harmonics (Eqs. 1-3, Sec. II A). Its validation consists of comparisons to TaylorF2Ecc and TaylorF2Ecch, which share the same leading-order restrictions and the same 3PN eccentric framework, plus injection-recovery tests where TaylorF2Ecck both generates and recovers the signal. These checks cannot falsify missing higher-order eccentric contributions or harmonic-content errors. The paper itself reports in Appendix F that pyEFPE, an independent eccentric model, diverges from TaylorF2Ecck beyond 3% mismatch for e0 > 0.08 and, notably, does not approach zero mismatch even as e0->0. The paper attributes the e0=0 mismatch to differing response-function conventions, but the practical consequence is that the two models have not been shown to agree on observable waveforms at the eccentricities being constrained. If the true waveform contains O(e0^4) phasing or additional harmonic structure that TaylorF2Ecck omits, the 90% upper limits could be systematically low, and the conclusion that both events are circularized field binaries would be weakened. The concern is not that the result is wrong; it is that the decisive assumption, model fidelity in an unvalidated regime, is untested.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript introduces TaylorF2Ecck, a frequency-domain inspiral waveform approximant for non-spinning eccentric binaries, implemented in LALSuite as a restricted version of the model in Ref. [72] with O(e0^2) corrections in the 3PN Fourier phase, O(e0) quadrupolar amplitudes, and harmonics (1,0), (1,-2), (2,-2), (3,-2). After sanity checks consisting of match calculations and zero-noise injection-recovery studies against TaylorF2Ecc, TaylorF2Ecch, and TaylorF2, the authors perform Bayesian parameter estimation on GW170817 and GW190425 using Bilby and report 90% credible upper limits of e0 < 0.011 and e0 < 0.028 at 20 Hz, respectively, with median values consistent with zero under both uniform and log-uniform priors. They also compare posteriors with and without periastron advance and, using circular TaylorF2 at 3PN-4.5PN order, argue that eccentric inspiral templates should include 3.5PN phasing contributions for reliable mass-ratio estimates.","tokens_in":37688,"tokens_out":5550,"duration_ms":54513,"significance":"If the upper limits are robust, the paper strengthens the case that GW170817 and GW190425 are consistent with field-formed, circularized binaries and that periastron-advance effects are not resolvable for these events. The work provides an openly available LALSuite implementation, a data/code release, and reproduces and extends earlier constraints from Lenon et al. (2020). The main caveats are that the abstract's Bayes-factor claim is not directly presented, the 3.5PN recommendation is inferred from circular-phasing runs rather than an eccentric 3.5PN model, and the waveform model lacks independent validation inside the constrained e0 range.","major_comments":[{"comment":"The statement in the abstract that 'Bayes factors show no strong evidence favoring the eccentric waveform over the quasi-circular waveform' is not supported by the results as reported. Table II lists Delta(log10 BF) values around 368 for both TaylorF2Ecck and TaylorF2Ecc, which appear to be log-evidences relative to Gaussian noise rather than pairwise model-comparison Bayes factors; the text's 'Delta(log10 BF) < 1' refers to a comparison between two eccentric models, TaylorF2Ecck and TaylorF2Ecc, not to a direct comparison with quasi-circular TaylorF2. To justify the abstract, the authors should report a direct Bayes factor between TaylorF2Ecck and the quasi-circular TaylorF2 model and correct the Table II column label.","section":"Abstract; Sec. III B; Table II"},{"comment":"The conclusion that eccentric models should incorporate initial-eccentricity contributions at least up to 3.5PN order is inferred from parameter-estimation runs of the quasi-circular TaylorF2 with 3PN, 3.5PN, 4PN, and 4.5PN phasing, not from an eccentric 3.5PN waveform. The observed shift in the mass-ratio posterior when 3.5PN circular phasing is added to TaylorF2Ecc demonstrates a circular-phasing sensitivity, but it does not by itself establish the size or importance of the missing O(e0^2) 3.5PN eccentric terms in TaylorF2Ecck. The claim should be reworded as a recommendation based on a circular-phasing proxy, or the missing eccentric 3.5PN model should be constructed and tested.","section":"Sec. III C; Sec. IV"},{"comment":"The validation of TaylorF2Ecck in the eccentricity range relevant to the reported upper limits is incomplete. The injection-recovery tests use TaylorF2Ecck as both the injection generator and the recovery model, so they cannot falsify missing higher-order eccentric phasing or harmonic-content errors. Appendix F shows that pyEFPE, an independent eccentric model, has mismatch exceeding 3% for e0 > 0.08 and does not approach zero mismatch as e0 approaches 0, a difference attributed to response conventions. Because the 90% credible upper limits are 0.011 and 0.028, they lie in a region where agreement with an independent model is not demonstrated; a low-e0 cross-check, for example recovering pyEFPE injections with TaylorF2Ecck, or an explicit statement of this limitation is needed before the upper limits can be taken as robust against model systematics.","section":"Sec. II B; Appendix F"}],"minor_comments":[{"comment":"The Table II header contains typos: 'TaylorEcck (Ecck)' should be 'TaylorF2Ecck (Ecck)' and the second model should be 'TaylorF2Ecc (Ecc)'.","section":"Table II"},{"comment":"The paper uses the same equation numbering in different sections, for example Eq. (1) appears in both Sec. II A and Sec. III A, which makes cross-referencing confusing; renumber equations sequentially or use section-prefixed labels.","section":"General"},{"comment":"The astrophysical conclusion that both events are consistent with field formation should be stated together with the caveat that the eccentricity limits are derived from a leading-order-truncated eccentric waveform model, as discussed in the major comments.","section":"Sec. IV"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid implementation study with reproducible code and a plausible null result. My main reservations are that the abstract overstates the Bayes-factor comparison, the 3.5PN inference is indirect, and the cross-validation gap with pyEFPE is not addressed inside the constrained range. These are fixable in revision; I do not see a load-bearing error that would require rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: this paper's headline numbers (e0 < 0.011 for GW170817, e0 < 0.028 for GW190425 at 20 Hz, 90%) confirm what Lenon et al. already reported with a simpler model. The new ingredient is the inclusion of periastron advance and extra harmonics in the waveform, and the demonstration that those extras do not change the eccentricity conclusion. That is a legitimate null result, and the parameter-estimation work behind it is solid. The paper ships code and data, which deserves credit.\n\nWhat is actually new: a restricted LALSuite implementation of the authors' own Tiwari et al. 2019 approximant, applied to two public LVK events. The observation that 3PN-truncated eccentric models bias the mass-ratio posterior, and that this bias disappears with 3.5PN circular phasing, is the most interesting piece—though it is an extrapolation from quasi-circular tests, not a direct demonstration for the eccentric model.\n\nWhere the soft spots are: the model validation is essentially self-consistent. TaylorF2Ecck is compared against TaylorF2Ecc and TaylorF2Ecch, which share the same leading-order eccentric structure, and the injection-recovery tests use TaylorF2Ecck as both injector and recoverer. Appendix F shows that the independent pyEFPE model does not converge to zero mismatch even as e0 → 0, which the authors attribute to response-function conventions, but the practical upshot is that no independent waveform agrees with TaylorF2Ecck in the e0 < 0.03 regime where the constraints are set. If the true waveform has O(e0^4) phasing or additional harmonic content that the leading-order truncation omits, the quoted upper limits could be biased low. I do not think this is fatal—the limits are consistent with prior work and the astrophysical expectation—but it is a real caveat that the paper should acknowledge more prominently.\n\nAlso, the abstract's claim that \"Bayes factors show no strong evidence favoring the eccentric waveform over the quasi-circular waveform\" is not directly backed by a table or figure: Table II compares TaylorF2Ecck against TaylorF2Ecc, not against the circular TaylorF2. The 3.5PN recommendation is inferred from circular-PN-order tests. These are presentation gaps, not load-bearing flaws.\n\nWho this is for: GW data analysts working on eccentricity and inspiral-only events. It deserves a serious referee: the PE is careful, the result is concise, and the model caveats are correctable in revision. My recommendation is to send it to peer review with the request that the authors add an independent waveform comparison or, failing that, state explicitly that the bounds are conditional on O(e0^2) truncation.\n\nI would not cite it in my own work in the near term, but I would keep it on the radar.","headline":"A careful but unsurprising null result: TaylorF2Ecck, a restricted version of the authors' own 2019 model, confirms Lenon et al.'s eccentricity bounds for two BNS events, with a caveat about model validation in exactly the constrained regime.","tokens_in":38293,"tokens_out":1829,"would_cite":false,"duration_ms":21359,"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":"Two neutron-star mergers show no detectable orbital eccentricity at 20 Hz.","keywords":["gravitational waves","initial eccentricity","TaylorF2ecck","inspiral","post-Newtonian","GW170817","GW190425","periastron advance"],"falsifier":"Re-analyze GW170817 and GW190425 with an independent eccentric inspiral model that includes $O(e_0^4)$ phase terms and more harmonics; if the 90% upper limits on $e_0$ move above 0.011 and 0.028, or if the Bayes factor starts favoring the eccentric model, the zero-eccentricity conclusion is a modeling artifact rather than a property of the data.","tokens_in":37144,"feed_emoji":"🌌","tokens_out":4870,"duration_ms":48324,"temperature":0.7,"pith_summary":"This paper introduces a new analytic template family, TaylorF2ecck, that models gravitational waves from non-spinning binaries on slightly eccentric inspiraling orbits, then uses it to re-analyze the two inspiral-dominated events GW170817 and GW190425. It tries to establish that both events are consistent with zero initial eccentricity at the 20 Hz reference frequency, with 90% credible upper limits of 0.011 and 0.028 respectively, and that no eccentric waveform is favored over the standard circular one. If true, these events carry no signature of dynamical assembly, supporting field formation for the observed neutron-star binaries. The paper also argues that eccentric phasing terms at 3.5PN order are needed for unbiased mass-ratio posteriors, not just 3PN terms.","feed_headline":"Two neutron-star mergers show no sign of orbital eccentricity","feed_subtitle":"A new analytic waveform puts 90% upper limits of 0.011 and 0.028 on initial eccentricity at 20 Hz.","key_machinery":"TaylorF2ecck is a fully analytic frequency-domain waveform for inspiraling non-spinning binaries in eccentric orbits, built from the stationary-phase approximation and a 3PN-accurate eccentric Keplerian description. It sums four harmonics $(1,0)$, $(1,-2)$, $(2,-2)$, $(3,-2)$, keeps eccentric corrections to $O(e_0^2)$ in the Fourier phase and $O(e_0)$ in the amplitude, and includes periastron advance consistently. This machinery lets the paper isolate eccentricity and periastron-advance effects in parameter estimation.","core_discovery":"Using a frequency-domain approximant that combines 3PN-accurate orbital phase with leading-order $O(e_0^2)$ eccentric corrections, periastron advance, and quadrupolar amplitudes with $O(e_0)$ harmonic terms, the authors find initial eccentricity at 20 Hz is negligible for both events: $e_0 < 0.011$ for GW170817 and $e_0 < 0.028$ for GW190425 at 90% confidence. Bayes factors do not favor the eccentric template over the quasi-circular template, and at these eccentricities the inclusion of periastron advance does not change the posteriors. A comparison of 3PN and 3.5PN quasi-circular phasing shows the mass-ratio posterior shifts substantially with the PN order of circular phase contributions, indicating that 3.5PN eccentric corrections are needed for reliable parameter estimation.","pith_inferences":["Applied to the same data with the same settings, the model gives an eccentricity estimate for GW200105 that is consistent with an independent fully precessing eccentric analysis but without strong Bayes evidence; higher-signal-to-noise neutron-star-black-hole inspirals may be the cleanest place to test periastron advance, since its effect grows with total mass.","The PN-order dependence of the mass-ratio posterior implies that published mass-ratio constraints derived from 3PN eccentric templates may carry a systematic bias comparable to the statistical error; re-analysis with 3.5PN eccentric phasing would settle this.","Lowering the analysis start frequency below 20 Hz would add many eccentricity-sensitive cycles; relative-binning techniques could make such a re-analysis computationally feasible and likely tighten the limits.","The mismatch between the TaylorF2 family and a fully precessing eccentric model begins to exceed 3% only above $e_0 \\approx 0.08$, suggesting the leading-order $O(e_0^2)$ phasing is adequate for weak-eccentricity events but not for moderate-eccentricity candidates."],"forward_implications":["The 90% upper limits $e_0 < 0.011$ for GW170817 and $e_0 < 0.028$ for GW190425 at 20 Hz mean both events are consistent with field-formed, circularized binaries; no dynamical-formation eccentricity signature is required by the data.","Periastron advance does not leave a measurable imprint in these two events, so for low-eccentricity BNS analyses simpler eccentric or circular templates remain adequate.","The mass-ratio posterior shifts between 3PN and 3.5PN circular phasing, so eccentric inspiral templates need 3.5PN eccentric phase terms before claimed mass-ratio constraints are trustworthy.","For GW170817-like signals with $e_0$ below about 0.01, circular templates are sufficient in parameter estimation, while eccentricity must be included above that threshold to avoid biased masses."],"supporting_citations":[{"why":"Supplies the 3PN-accurate eccentric Keplerian dynamics and harmonic amplitudes that TaylorF2ecck implements.","marker":"[72]"},{"why":"Provides the earlier TaylorF2ecc model with leading-order eccentric phase corrections used as the main comparison baseline.","marker":"[70]"},{"why":"Previous eccentricity bounds for GW170817 and GW190425 that this work corroborates and refines.","marker":"[69]"},{"why":"Defines the prior choices and eccentricity thresholds for BNS inspiral parameter estimation used in the injection-recovery tests.","marker":"[26]"},{"why":"Gives the quasi-circular TaylorF2 Fourier phasing that the new eccentric model extends.","marker":"[14]"},{"why":"Supplies 4.5PN-accurate circular Fourier phase coefficients used to test the PN-order dependence of the mass-ratio posterior.","marker":"[76]"},{"why":"Provides an independent eccentric waveform model used for mismatch comparisons that reveal divergence at moderate eccentricity.","marker":"[51]"}],"fun_headline_variants":["No eccentricity in GW170817 and GW190425","Neutron-star mergers show no orbital eccentricity","Initial eccentricity negligible for neutron-star mergers","Analytic model rules out eccentric neutron-star orbits","Mergers born with effectively zero eccentricity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central claim depends on the waveform model representing the eccentric signal faithfully at the very small eccentricities it is constraining; if the leading-order eccentric phase truncation or the limited harmonic set distorts the templates even at $e_0$ below 0.01, the reported upper limits could be biased toward zero.","fun_headline_variants_meta":{"raw":{"variants":["No eccentricity in GW170817 and GW190425","Neutron-star mergers show no orbital eccentricity","Initial eccentricity negligible for neutron-star mergers","Analytic model rules out eccentric neutron-star orbits","Mergers born with effectively zero eccentricity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000726,"raw_usage":{"total_tokens":3305,"prompt_tokens":1050,"completion_tokens":2255,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":666,"completion_tokens_details":{"reasoning_tokens":2182}},"tokens_in":666,"tokens_out":2255,"duration_ms":17692,"temperature":1.0,"reasoning_tokens":2182,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T17:19:47.174746+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-analyze GW170817 and GW190425 with an independent eccentric inspiral model that includes $O(e_0^4)$ phase terms and more harmonics; if the 90% upper limits on $e_0$ move above 0.011 and 0.028, or if the Bayes factor starts favoring the eccentric model, the zero-eccentricity conclusion is a modeling artifact rather than a property of the data.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the earlier TaylorF2ecc model with leading-order eccentric phase corrections used as the main comparison baseline."},{"cited_title":"Gamba, M","cited_arxiv_id":null,"evidence_quote":"Supplies 4.5PN-accurate circular Fourier phase coefficients used to test the PN-order dependence of the mass-ratio posterior."}],"review_version":2}