{"id":"014eef9e-4569-4884-abdc-3b1c6b0632ab","arxiv_id":"1908.10635","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Anisotropic lightning sources do not always worsen Schumann-resonance correlated magnetic noise for stochastic gravitational wave searches, and the Virgo-KAGRA pair is least affected.","lead":"This paper studies how the global magnetic noise caused by lightning, known as Schumann resonances, can contaminate searches for the random gravitational wave background. It finds that the Virgo and KAGRA detector pair is the most resistant to this noise, even when lightning sources are strongly clustered.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The VK robustness claim rests on a frequency-independent, flash-rate-proportional anisotropy spectrum (Eqs. 20-22) that is not tested against measured anisotropic Schumann data; this is the load-bearing assumption.","rationale":"The reader's weakest_assumption identifies the same load-bearing concern: the frequency-independent, flash-rate-proportional anisotropy model of Eqs. (20)-(22). I agree with that identification. The VK robustness is a quantitative cancellation effect, and the paper itself notes that the properties of correlated magnetic noise depend crucially on geometry and geography. The model has independent support for the LIGO-Hanford-Livingston pair (Appendix B reproduces the measured Omega_mag), which is real evidence, but that validation does not exercise the VK cancellation on which the central claim depends. A direct measurement of the VK magnetic cross-spectrum is the cleanest test of whether the assumed anisotropy model is adequate. Since the concern is a missing validation and an unsurveyed sensitivity rather than an identified internal inconsistency, the reader's CONDITIONAL verdict should stand unchanged; the paper would be strengthened by adding either the proposed empirical check or an explicit bound on how much frequency-dependent anisotropy would be required to move VK across the SNR=5 line.","tokens_in":22550,"tokens_out":7604,"duration_ms":94164,"concrete_test":"Extract the Virgo-KAGRA magnetic-field cross-spectrum M12(f) in the 5-50 Hz band from the magnetometer coherence data of Ref. [28], compute <SB> for VK using this measured M12 with the same transfer functions and SNR=5 threshold as Sec. IV, and compare with the model prediction from Eqs. (14)-(19) with W from Eq. (22) at epsilon=0.8. If the measured-M12 <SB> lies above the SNR=5 line, or differs from the model by more than the factor needed to cross that line, the frequency-independent flash-rate anisotropy assumption is load-bearing.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that Virgo-KAGRA stays at or below the SNR=5 detectable amplitude for epsilon=0.8 for tensor, vector, and scalar modes—requires that the anisotropic part of the magnetic-field spectrum be exactly W(Omega)=(1-epsilon)+epsilon w(Omega), with w proportional to the mean annual flash-rate map and no frequency dependence. This is stated in Sec. IV A: 'Assuming that the anisotropic component is independent of frequencies.' The VK result is not a broad-band suppression; it emerges from phase cancellation in the integral of M12(f) against the oscillatory optimal filter (Eq. 12), as the paper emphasizes in Sec. V. A frequency-dependent anisotropy pattern would change the relative l-mode weights gamma_l^B(f) and hence the shape of M12(f), and there is no guarantee the cancellation survives. Moreover, the flash-rate map counts all flashes equally and ignores stroke current-moment spectra, so even the frequency-independent identification of w with flash rate is an unvalidated approximation. No uncertainty from the lightning dataset or from alternative source models is propagated into Figs. 3-4, and the paper's own conclusion (Sec. VI) admits that the results rely on the model's simplifications. This does not invalidate the paper, but it makes the quantitative VK ranking conditional.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper extends the authors' earlier analytic model of correlated magnetic noise from Schumann resonances to include anisotropic lightning source distributions. The anisotropy is parametrized by ε in W(Ω) = (1−ε) + εw(Ω), where w is derived from the mean annual flash-rate map. The model is applied to all pairs of five second-generation detectors (LIGO Hanford, Livingston, India, Virgo, KAGRA) and the impact of the correlated magnetic noise on stochastic GW detection is evaluated for unpolarized tensor, circularly polarized tensor, unpolarized vector, and unpolarized scalar modes, with flat and f^{2/3} spectra. The principal finding is that the Virgo–KAGRA (VK) pair remains the most robust against correlated magnetic noise even for strong anisotropy with ε = 0.8, staying at or below the SNR = 5 one-year detection threshold for all GW polarization types. The robustness is attributed not to a broadband suppression but to phase cancellation in the integral of the magnetic noise spectrum M12(f) against the oscillatory optimal filter.","tokens_in":22809,"tokens_out":3131,"duration_ms":45142,"significance":"If the result holds, it provides a concrete, falsifiable prediction for stochastic GW searches: the VK pair is the least contaminated by Schumann-resonance correlated noise among second-generation detector pairs, for tensor, vector, and scalar modes alike. The paper's strengths are its transparent derivation of Eq. (19) including the weighted coherence function, the use of a physically motivated source map, the systematic scan of ε rather than a fit tuned to the VK result, and the extension to non-tensor and circularly polarized GWs. The previous model from Ref. [24] has been checked against magnetometer measurements (Refs. [22, 23, 28]), and Appendix B further validates the updated transfer-function version against LIGO-Virgo data. The main caveat is that the anisotropic component of the magnetic spectrum is assumed frequency-independent and proportional to the climatological flash-rate map, an assumption that is load-bearing for the VK cancellation result and is not tested against measured anisotropic Schumann spectra.","major_comments":[{"comment":"The VK robustness result relies on the assumption that the anisotropic part of the magnetic spectrum is frequency-independent, W(Ω) = (1−ε) + εw(Ω). The phase cancellation that suppresses ⟨S_B⟩ for VK is computed from the frequency integral of M12(f) against the oscillatory optimal filter, and the l-mode weights γ_l^B(f) enter M12 through Eq. (19). A frequency-dependent anisotropy pattern would change the relative l-mode weights and hence the shape of M12(f); there is no demonstrated reason that the cancellation survives such a change. The authors should test this by repeating the calculation with a simple frequency-dependent extension of W (e.g., different source terms at the first and higher Schumann resonances) or by comparing against measured anisotropic Schumann cross-spectra between Virgo and KAGRA sites. Without such a check, the quantitative claim that VK is robust up to ε = 0.8 is conditional on an unvalidated modeling choice.","section":"Sec. IV A, Eqs. (20) and (21)"},{"comment":"The normalized function w(Ω) is identified with the observed mean annual flash-rate map, which counts all lightning flashes equally. Real lightning sources have different stroke current-moment spectra depending on region and storm type, so even a frequency-independent anisotropy would not strictly equal the flash-rate map: the magnetic-field amplitude per flash is not region-independent. The paper should either justify this identification with a reference to measured source spectra or demonstrate that plausible variations in w(Ω) do not change the VK ranking. As it stands, no uncertainty from the lightning dataset or from alternative source models is propagated into the curves in Figs. 3 and 4.","section":"Sec. IV A, Eq. (22)"},{"comment":"The conclusion that VK is 'potentially' the most insensitive pair is appropriately cautious, but the abstract and Section IV B 2 make the stronger assertion that VK is robust for all types of stochastic GWs without prominently flagging the model dependence. Since the quantitative ranking and the cancellation mechanism both rely on the frequency-independent anisotropy assumption, the central claim should be explicitly qualified as a model-dependent result in the abstract and in the main text, with the unvalidated assumption stated at the point where the VK claim is made.","section":"Sec. VI"}],"minor_comments":[{"comment":"The phrase 'the correlated magnetic' appears to be missing the word 'noise'; it should read 'the correlated magnetic noise.'","section":"Abstract"},{"comment":"The word 'lighting' is used where 'lightning' is meant; this typo appears in the abstract and in several places in the introduction.","section":"Abstract and Sec. I"},{"comment":"The shape function contains a typographical error: '|E𝓁(f)|2|' has an extra vertical bar and should read '|E𝓁(f)|2'.","section":"Eq. (16)"},{"comment":"The sentence 'which closely matchs with those estimated by [23]' contains a typo: 'matchs' should be 'matches.'","section":"Sec. II B"},{"comment":"The caption is crowded and the line labels for the four pairs are not clearly mapped to the four panels; adding explicit panel letters (a)–(d) would improve readability.","section":"Fig. 2 caption"},{"comment":"Reference [48] is an LIGO aLOG entry with a URL; this is not a stable archival reference. The authors should provide a permanent citation or archival DOI if one exists, or at least note that the entry is an internal LIGO document.","section":"Ref. [48]"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a coherent extension of the authors' own published model, and the central computation is internally consistent. The main concern is not circularity or internal inconsistency but the unvalidated frequency-independence of the anisotropic source model, which is load-bearing for the headline VK robustness claim. The paper's own limitations paragraph in Sec. VI acknowledges this, but the abstract does not. A revision that adds a sensitivity test to frequency-dependent anisotropy or, failing that, clearly demotes the VK claim to a model-conditioned prediction, would be appropriate. The self-citation pattern is not problematic here: Ref. [24] is the basis of the calculation and has been validated against magnetometer measurements."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a legitimate extension of a previously validated analytic model, and the central qualitative result—that Virgo–KAGRA is the pair least contaminated by correlated Schumann noise even under strongly anisotropic lightning—survives contact with the paper. The stress-test note points at the right weak spot, but it lands as a caveat rather than a fatal flaw.\n\nWhat is genuinely new: the paper takes the authors' 2017 isotropic model (which already matched magnetometer measurements of Schumann correlations) and adds an anisotropic lightning source distribution built from the observed OTD/LIS flash-rate map, parametrized by a single anisotropy strength epsilon. It then computes the correlated magnetic noise contribution to the stochastic-background cross-correlation statistic for all pairs among the five second-generation detectors, for unpolarized tensor, circularly polarized tensor, vector, and scalar GW modes. The VK result is not constructed; it emerges from the integral of an oscillating M12 against the optimal filter, with the detector pair's insensitivity below 25 Hz playing a big role. Appendix B is a nice check: with the updated LIGO transfer function, the model reproduces the frequency-dependent Omega_mag measured during O2, including zero-crossings and Schumann peaks.\n\nThe soft spots are real but bounded. The anisotropic spectrum assumes frequency independence and takes the flash-rate map as the angular weight, with no dependence on stroke current-moment spectra; that is an unvalidated approximation, and it feeds directly into the shape of M12(f) that produces the phase cancellation. The authors say as much in the conclusion. Because the VK suppression partly comes from the detector pair's weak sensitivity at the Schumann frequencies rather than from the anisotropy model, the qualitative robustness claim probably holds, but the quantitative ranking and the factors-of-a-few in epsilon should be read as model-dependent. The coupling parameters ri(f) for KAGRA and India remain uncertain; the paper is transparent about this and tests an updated LIGO transfer function.\n\nCitation pattern: self-citation to ref [24] is appropriate here, since the whole point is extending that validated model, and they cite the external magnetometer measurements. No red flags.\n\nRecommendation: send it to a serious referee. The question is whether the anisotropy model is physical enough; a referee should push on whether measured Schumann anisotropic spectra exist to test the ansatz, and whether the epsilon scan is wide enough. For the stochastic-GW and detector-characterization community, this is worth reading.","headline":"Extends a validated Schumann-noise model to anisotropic lightning and finds the Virgo-KAGRA pair surprisingly robust; the main caveat is the untested frequency-independence of the anisotropy ansatz.","tokens_in":23346,"tokens_out":2231,"would_cite":true,"duration_ms":24678,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Lightning noise won't blind Virgo-KAGRA to stochastic GWs","keywords":["stochastic gravitational waves","Schumann resonances","correlated magnetic noise","lightning source anisotropy","cross-correlation analysis","overlap reduction function","Virgo","KAGRA"],"falsifier":"A year-long measurement of the magnetic-field cross-spectrum between the Virgo and KAGRA sites, compared mode by mode with the model's prediction at $\\epsilon=0.8$, would settle the claim: if the anisotropic part of the Schumann spectrum has a different frequency dependence than the flat weight assumed in Eqs. (20)-(22), or is not shaped like the climatological flash-rate map, the predicted ranking and the VK margin would change. A simpler check is whether the observed ratio $\\langle S_B\\rangle_{\\epsilon}/\\langle S_B\\rangle_{\\epsilon=0}$ for any pair follows the computed curve.","tokens_in":22323,"feed_emoji":"⚡","tokens_out":5676,"duration_ms":52398,"temperature":0.7,"pith_summary":"This paper claims that the correlated magnetic noise produced by Earth's Schumann resonances does not prevent detection of stochastic gravitational waves, even when the lightning sources that drive those resonances are strongly concentrated on tropical continents. Extending an earlier isotropic analytical model, the authors assign a weight $W(\\hat{\\Omega})=(1-\\epsilon)+\\epsilon w(\\hat{\\Omega})$ to each sky direction using the observed mean annual lightning flash-rate map, with $\\epsilon=0.8$ representing strong anisotropy. They find that the noise impact does not always grow with anisotropy: for some detector pairs and polarization modes the integral of the oscillating magnetic-noise spectrum against the optimal filter partially cancels. The robust result is that Virgo and KAGRA, among all pairs of second-generation detectors, have correlated magnetic noise at or below the amplitude of a stochastic background detectable at signal-to-noise ratio 5 in one year, for tensor, vector, and scalar modes.","feed_headline":"Lightning noise won't blind Virgo-KAGRA to stochastic GWs","feed_subtitle":"Even strongly anisotropic lightning sources keep correlated magnetic noise below detectable levels for this detector pair.","key_machinery":"The machinery is the analytical correlated-noise spectrum $M_{12}(f)=\\frac{1}{8\\pi}P_B(f)\\sum_\\ell \\frac{|E_\\ell(f)|^2}{|E_\\ell(f'_\\ell)|^2} \\gamma^B_\\ell(\\hat{r}_1,\\hat{r}_2)$, with the coherence function $\\gamma^B_\\ell$ carrying the geometry of the detector pair and the sky weight $W(\\hat{\\Omega})$; the sum runs over Schumann transverse-magnetic modes with Lorentzian line shapes $|E_\\ell(f)|^2$. The extension replaces the isotropic sky average by a weighted average over the observed lightning map, so the coherence function becomes $\\gamma^B_\\ell(\\hat{r}_1,\\hat{r}_2) \\propto \\int d^2\\hat{\\Omega}\\, W(\\hat{\\Omega}) P^1_\\ell(\\hat{\\Omega}\\cdot\\hat{r}_1)P^1_\\ell(\\hat{\\Omega}\\cdot\\hat{r}_2)[\\hat{e}_1(\\hat{\\Omega})\\cdot\\hat{X}_1][\\hat{e}_2(\\hat{\\Omega})\\cdot\\hat{X}_2]$. The cross-correlation statistic $\\langle S_B\\rangle$ is then computed by integrating $M_{12}$ against the optimal filter $\\tilde{Q}(f)\\propto \\sum_A \\Omega^A_{\\rm gw}(f)\\gamma^A_{12}(f)/[f^3 P_1(f)P_2(f)]$, and it is the phase relationship between the oscillating $M_{12}$ and $\\tilde{Q}$ that determines whether anisotropy increases or decreases the impact.","core_discovery":"The central discovery is that anisotropic lightning loading changes the correlated magnetic noise $M_{12}(f)$ in a way that is mostly coherent and small in the spectrum, yet can swing the detection impact by factors of two to three or more depending on the detector pair and the GW polarization. For the LIGO Hanford-Livingston pair the absolute noise stays large under all assumed anisotropies, while for Virgo-KAGRA a large phase cancellation in $\\langle S_B\\rangle$ keeps the equivalent $\\Omega_{\\rm gw}h^2$ well below the one-year, SNR=5 threshold even at $\\epsilon=0.8$. The authors emphasise that this robustness is a combined effect of the pair's large separation, its detector noise spectra, and its sensitivity peaking above roughly 40 Hz, away from the low-frequency Schumann band. They also show analytically that the exact nulling condition of the isotropic model survives only for specially symmetric source distributions, so in general no choice of coupling vectors can cancel the noise.","pith_inferences":["The magnetometer cross-correlation between Virgo and KAGRA that has already been reported in the literature could be re-analysed to test the predicted $\\epsilon$ dependence directly; the paper does not perform that comparison.","Because lightning activity varies seasonally and diurnally while the flash-rate map is an annual average, the model probably smooths over transient increases in anisotropy; a time-resolved estimate might find intervals where the Virgo-KAGRA margin shrinks.","The same phase-cancellation mechanism suggests a design rule for future detector sites: choose pairs with large separation and high-frequency sensitivity to stay safe against Schumann noise, a hint the paper states once but does not develop.","If future transfer-function measurements show that KAGRA couples to magnetic fields more strongly than the assumed $\\kappa_i=2$, $b_i=2.67$ values, the Virgo-KAGRA margin could close, so the conclusion is conditional on that coupling."],"forward_implications":["If the model is right, the Virgo-KAGRA pair should be the least contaminated by correlated magnetic noise for unpolarized tensor, circularly polarized tensor, vector, and scalar stochastic backgrounds, and should be preferred in searches for all of these.","The LIGO Hanford-Livingston pair remains dominated by correlated magnetic noise under every assumed anisotropy, so improvements in magnetic coupling or noise subtraction are needed before that pair can contribute to a stochastic detection.","Anisotropy can suppress rather than amplify correlated noise in some configurations, because the product of the oscillating magnetic-noise spectrum and the optimal filter can cancel by phase; monotonic scaling with anisotropy is not a safe assumption.","A detector pair whose sensitivity begins above roughly 40 Hz largely avoids the low-frequency Schumann band, which is why Virgo-KAGRA stays robust even when the lightning source distribution is strongly anisotropic.","For astrophysical backgrounds with $\\Omega_{\\rm gw}\\propto f^{2/3}$, the same qualitative ranking holds and the Virgo-KAGRA pair still falls below the one-year SNR=5 threshold for correlated magnetic noise."],"supporting_citations":[{"why":"Supplies the analytical model of the correlated magnetic noise spectrum $M_{12}$ that this paper extends to anisotropic lightning sources.","marker":"[24]"},{"why":"Provides the cross-correlation statistic and optimal filter formalism used to translate magnetic noise into an effective gravitational-wave amplitude.","marker":"[18]"},{"why":"Gives the axisymmetric transverse-magnetic modes of the Earth-ionosphere cavity used to describe the Schumann resonances.","marker":"[42]"},{"why":"Reports magnetometer measurements of correlated magnetic noise between detectors, the empirical behaviour the model is built to reproduce.","marker":"[22]"},{"why":"Provides the coupling and transfer-function parameters that the paper adopts in its fiducial setup.","marker":"[23]"},{"why":"Supply the mean annual lightning flash-rate data used to define the anisotropic weight function $w(\\hat{\\Omega})$.","marker":"[45, 46]"},{"why":"Reports the observed Schumann-resonance correlation between Virgo and KAGRA, the empirical motivation for studying this detector pair.","marker":"[28]"},{"why":"Gives the analytical overlap reduction functions for vector and scalar GW modes used to compute the signal-to-noise ratio for non-tensor polarizations.","marker":"[32]"},{"why":"Provides the updated LIGO transfer-function calibration used in Appendix B to check whether reduced coupling changes the conclusions.","marker":"[48]"}],"fun_headline_variants":["Lightning can't mask stochastic GWs for Virgo-KAGRA","Virgo-KAGRA pair resists correlated lightning noise","Stochastic GW detection robust to anisotropic lightning","Correlated lightning noise fails to challenge Virgo-KAGRA","Anisotropic lightning won't blind Virgo-KAGRA"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The anisotropic part of the magnetic-field spectrum is assumed to be frequency-independent and directly proportional to the mean annual lightning flash-rate map; if the true Schumann source distribution has a different frequency dependence, the computed noise amplitudes and the Virgo-KAGRA ranking could change.","fun_headline_variants_meta":{"raw":{"variants":["Lightning can't mask stochastic GWs for Virgo-KAGRA","Virgo-KAGRA pair resists correlated lightning noise","Stochastic GW detection robust to anisotropic lightning","Correlated lightning noise fails to challenge Virgo-KAGRA","Anisotropic lightning won't blind Virgo-KAGRA"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000311,"raw_usage":{"total_tokens":1816,"prompt_tokens":1031,"completion_tokens":785,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":647,"completion_tokens_details":{"reasoning_tokens":702}},"tokens_in":647,"tokens_out":785,"duration_ms":7394,"temperature":1.0,"reasoning_tokens":702,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:37:25.024032+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A year-long measurement of the magnetic-field cross-spectrum between the Virgo and KAGRA sites, compared mode by mode with the model's prediction at $\\epsilon=0.8$, would settle the claim: if the anisotropic part of the Schumann spectrum has a different frequency dependence than the flat weight assumed in Eqs. (20)-(22), or is not shaped like the climatological flash-rate map, the predicted ranking and the VK margin would change. A simpler check is whether the observed ratio $\\langle S_B\\rangle_{\\epsilon}/\\langle S_B\\rangle_{\\epsilon=0}$ for any pair follows the computed curve.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the analytical model of the correlated magnetic noise spectrum $M_{12}$ that this paper extends to anisotropic lightning sources."},{"cited_title":"Christensen, Phys","cited_arxiv_id":null,"evidence_quote":"Reports magnetometer measurements of correlated magnetic noise between detectors, the empirical behaviour the model is built to reproduce."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the updated LIGO transfer-function calibration used in Appendix B to check whether reduced coupling changes the conclusions."}],"review_version":1}