REVIEW 4 major objections 6 minor 17 references
Inter-detector time delay fails as a standalone test of LIGO-Virgo events and may signal a continuous correlated background.
Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →
Event-window cross-correlations in LIGO–Virgo pairs are statistically indistinguishable from off-source noise in 26/27 measurements, so the time-delay criterion fails as a standalone test and a continuous correlated component is proposed.
T0 review reviewed 2026-07-14 challenge →
load-bearing objection Solid empirical extension of Creswell: under band-pass-only processing, event-window C(τ) peaks sit inside the noise ensemble in 26/27 LIGO–Virgo pairs; the SGWB reading is an open interpretation, not a demonstrated result. the 4 major comments →
On correlated noise in the LIGO-Virgo network: a test of the stochastic gravitational-wave background hypothesis
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
Core claim
Across nine candidate events and all three LIGO-Virgo detector pairs, the peak of the normalised cross-correlation C(τ) in the event window is statistically indistinguishable from the peaks found in surrounding noise windows in 26 of 27 cases, including every measurement that involves Virgo. The sole exception is the H1×L1 pair of GW190412. The authors therefore argue that the inter-detector lag cannot serve as a robust, model-independent validation criterion and that the persistent correlations point to a continuous correlated component in the strain data.
What carries the argument
The empirical p-value of the peak of the normalised cross-correlation C(τ): for each detector pair a 0.2 s event window is compared with N=200 off-source windows of the same length after only a fourth-order Butterworth band-pass (35-350 Hz), without whitening or template subtraction; p_emp is the fraction of noise peaks that meet or exceed the event peak.
Load-bearing premise
That the statistical sameness of event and noise correlations, together with Virgo's independence from LIGO, is enough to establish a continuous physical background rather than residual instrumental or analysis effects whose amplitude still fits existing upper limits.
What would settle it
Extend the same band-pass-only cross-correlation test to all six detector pairs of the H1-L1-V1-KAGRA network; if the correlated structure is a physical stochastic background it must appear in every KAGRA pair at the light-travel delays fixed by the intercontinental baselines, and a null result in those pairs would refute the hypothesis.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper extends the Creswell et al. inter-detector cross-correlation test to nine O2/O3 LIGO–Virgo candidates, computing the normalized C(τ) in a 0.2 s event window against an empirical ensemble of N=200 off-source windows under a band-pass-only pipeline (no whitening, no Tukey window, no template). Across 27 pair-measurements (H1×L1, H1×V1, L1×V1), 26 event peaks are statistically indistinguishable from the noise ensemble (p_emp ≥ 0.12), including every pair involving Virgo; the sole exception is GW190412 in H1×L1 (p_emp = 0.005). The authors conclude that C(τ) is not a robust standalone event/noise separator and interpret the persistent off-source correlation structure, especially in LIGO–Virgo pairs, as evidence for a continuous correlated physical component for which a stochastic gravitational-wave background is a plausible candidate. A whitening check is reported for two events and removes the GW190412 excess.
Significance. Extending the Creswell-style test to Virgo and to a larger O2/O3 sample is a useful, model-independent contribution: Virgo’s independent hardware and environment genuinely tighten the shared-infrastructure loophole that limited the original H1–L1-only analyses. The empirical core (Table I; 26/27 non-detections under the stated pipeline) is clearly documented and falsifiable, and the authors’ own whitening test is a strength. If the stronger reading—that the data contain a continuous physical correlated component at a level relevant to short-window C(τ)—were quantitatively established, it would matter for both CBC validation practice and SGWB searches. As written, that stronger reading is not yet demonstrated; the paper’s lasting value is more likely the careful negative result on C(τ) as a standalone criterion and the Virgo control, provided the interpretation is brought into line with the evidence.
major comments (4)
- [Section II–III / abstract] Section II–III and the abstract: the null result (26/27 pairs with p_emp ≥ 0.12) is equally consistent with C(τ) on 0.2 s band-passed strain simply being an insensitive detection statistic for the CBC signals in the sample. Matched filtering exists precisely because raw cross-correlation lacks sensitivity at typical network SNRs. The paper never quantifies the expected |C*_ev| (or its lag) under the standard CBC hypothesis for the reported SNRs and the adopted window/filter. Without that benchmark, indistinguishability from the off-source ensemble does not by itself support a continuous physical component or undermine the events; it may only restate that this statistic is weak. This calculation is load-bearing for the interpretive claim.
- [Section IV E] Section IV E (and title/abstract/conclusions): the SGWB reading is advanced as “the most plausible” explanation without any comparison of the observed |C*| amplitudes (or their frequency content) to existing LVK upper limits on Ω_GW, nor to the expected short-window cross-correlation for a background at those limits. The paper itself leaves the amplitude question open. Until that check is done—or the claim is clearly demoted to an untested hypothesis—the leap from “event and off-source C(τ) look alike under band-pass-only processing” to “evidence for an SGWB” is not supported by the data presented.
- [Section IV C] Section IV C: Virgo independence rules out shared LIGO infrastructure, but not globally coherent terrestrial correlated noise (e.g. Schumann resonances and other EM disturbances), which is a standard foreground in SGWB cross-correlation analyses and can couple into widely separated interferometers without common hardware. The manuscript does not engage this literature or test whether the off-source C(τ) structure is consistent with known terrestrial coupling versus a GW background (overlap-reduction-function shape, frequency dependence, etc.). That omission weakens the claim that instrumental explanations are “ruled out.”
- [Table I / Section III] Table I and Section III: several event peaks sit at unphysical lags (e.g. GW170818 H1×L1 at τ_ev = −11.5 ms, outside the ±10 ms light-travel window), yet with unremarkable amplitudes. Under a continuous GW component one still expects the correlation structure to be shaped by light-travel geometry and the overlap reduction function. The paper should report the lag distribution of the noise ensemble and test whether it is consistent with a GW origin or with broadband/local noise; unphysical event lags currently cut against a straightforward SGWB interpretation of the same statistic used for the events.
minor comments (6)
- [Section II] Section II: free analysis choices (w = 0.2 s, N = 200, guard zone ±2 s, min separation 1 s, band [35, 350] Hz) should be motivated or subjected to a brief robustness check; at least state whether p_emp rankings are stable under modest changes of w and N.
- [Section IV B] Section IV B: the whitening test is valuable but reported only for GW170814 and GW190412. Extending it to the full sample (or stating why not) would strengthen the claim that whitening systematically suppresses the off-source structure.
- [Figures 1–4] Figures 1–4: the noise ensemble is hard to read as overplotted curves; consider a percentile envelope (e.g. 68/95%) plus the event curve so that “within the noise envelope” is visually quantitative.
- [Table I] Table I footnote and GW200129 discussion: the L1 glitch is appropriately flagged; consider marking the affected p_emp values as upper/lower limits or excluding them from the 26/27 headline count in the abstract.
- [Section IV E] Reference [16] (teleparallel/Bondi–Sachs residual radiation) is used to motivate a persistent gravitational component after a burst; a short clause should clarify that this is a specific theoretical framework, not a standard GR result, so readers can weight it accordingly.
- Typographical/date issues: manuscript date “July 14, 2026”; occasional missing spaces in compound words (e.g. “TheLIGO-Virgocollaboration”, “abeadfreetoslide”); standardize detector-pair notation (H1×L1 vs H1–L1).
Circularity Check
Empirical cross-correlation test is self-contained and model-independent; only a non-load-bearing self-citation supplies theoretical motivation for the optional SGWB interpretation.
specific steps
-
self citation load bearing
[Section IV E, paragraph beginning “This interpretation is not introduced here in an ad hoc manner”]
"In the context of the teleparallel equivalent of general relativity applied to the Bondi-Sachs space-time, it has been shown that the energy of gravitational radiation depends on the functions that generate the news functions, and it has been argued that, after a burst, gravitational radiation may remain present in the background structure of the space-time rather than disappearing completely. … This possibility provides a theoretical basis for interpreting the correlated off-source structure observed here as compatible with a stochastic gravitational-wave background [16]."
The only theoretical scaffolding offered for residual post-burst gravitational radiation (and hence for reading the observed off-source C(τ) structure as an SGWB) is a prior paper by overlapping authors. The citation is not required for the empirical ranking of event versus noise peaks, nor does it quantitatively constrain amplitudes; it merely supplies a non-ad-hoc narrative for an optional interpretation.
full rationale
The paper’s primary result is a direct statistical comparison: for each of 27 detector-pair measurements the peak of the normalized Pearson cross-correlation C(τ) inside a 0.2 s event window is ranked against an empirical ensemble of N=200 off-source windows of identical length drawn from the same 256 s stretch (band-pass only, no whitening or templates). The empirical p-value (Eq. 4) and the tabulated outcomes (Table I) follow immediately from that ranking; nothing is fitted and nothing is predicted from a prior model. The sole self-citation ([16], overlapping authors Ulhoa & Carneiro) appears only in Section IV E as optional theoretical motivation for why residual gravitational radiation might persist after a burst, thereby rendering an SGWB interpretation “not ad hoc.” That citation is not used to derive any number, to force uniqueness of the interpretation, or to exclude alternatives; the paper itself leaves amplitude consistency with existing SGWB upper limits open. Consequently the central empirical claim stands without the self-citation, producing only minor circularity of the non-load-bearing kind.
Axiom & Free-Parameter Ledger
free parameters (4)
- event/noise window width w =
0.2 s
- number of off-source windows N =
200
- band-pass frequency interval =
[35, 350] Hz
- guard zone and minimum window separation =
±2 s / 1 s
axioms (3)
- domain assumption A band-pass-only pipeline without PSD whitening or Tukey windowing constitutes an unbiased test of inter-detector independence.
- domain assumption Instrumental and analysis correlations shared by H1 and L1 are absent from H1–V1 and L1–V1 pairs because of Virgo’s independent hardware, software, power frequency and seismic environment.
- ad hoc to paper Persistence of comparable cross-correlation structure in off-source windows implies a continuous physical component rather than residual non-GW noise or analysis artefacts.
invented entities (1)
-
continuous correlated physical component in the strain data (identified with a stochastic gravitational-wave background)
no independent evidence
Cite this review
Pith. "Pith review of On correlated noise in the LIGO-Virgo network: a test of the stochastic gravitational-wave background hypothesis." pith.science (2026). https://pith.science/paper/LG7IOUOU
@misc{pith2026260710816,
author = {Pith},
title = {Pith review of: On correlated noise in the LIGO-Virgo network: a test of the stochastic gravitational-wave background hypothesis},
year = {2026},
howpublished = {\url{https://pith.science/paper/LG7IOUOU}},
note = {Machine review of arXiv:2607.10816}
}
abstract
We revisit the inter-detector cross-correlation criterion used in gravitational-wave validation by extending the analysis of Creswell et al.~\cite{Creswell2017} to nine LIGO--Virgo candidate events from the O2 and O3 observing runs. For each event, we analyze all three detector pairs (H1$\times$L1, H1$\times$V1, L1$\times$V1) and compare the normalized cross-correlation function $C(\tau)$ in the event window with an empirical ensemble of $N=200$ surrounding off-source windows, using a band-pass-only pipeline. In 26 of 27 pair-measurements, the event peak is statistically indistinguishable from the corresponding noise ensemble, including all pairs involving the independent Virgo detector. The only exception is GW190412 in the H1$\times$L1 pair. We conclude that the cross-correlation statistic does not provide a robust standalone separation between event and noise. We interpret the persistence of comparable correlation structure in the LIGO--Virgo pairs as evidence that the strain data may contain a continuous correlated physical component, for which a stochastic gravitational-wave background is a plausible candidate.
Figures
Reference graph
Works this paper leans on
-
[1]
On the time lags of the LIGO signals,
J. Creswell, S. von Hausegger, A. D. Jackson, H. Liu, and P. Naselsky, “On the time lags of the LIGO signals,” JCAP2017, 013 (2017), arXiv:1706.04191
Pith/arXiv arXiv 2017
-
[2]
Ob- servation of gravitational waves from a binary black hole merger,
B. P. Abbottet al.(LIGO Scientific, Virgo), “Ob- servation of gravitational waves from a binary black hole merger,” Phys. Rev. Lett.116, 061102 (2016), arXiv:1602.03837
Pith/arXiv arXiv 2016
-
[3]
Näherungsweise integration der feldgle- ichungen der gravitation,
A. Einstein, “Näherungsweise integration der feldgle- ichungen der gravitation,” Sitzungsber. Preuss. Akad. Wiss. , 688–696 (1916)
1916
-
[4]
Conference on the role of gravitation in physics,
R. P. Feynman, “Conference on the role of gravitation in physics,” inProceedings of the Chapel Hill Conference, edited by C. M. DeWitt and D. Rickles (1957) reprinted in:The Role of Gravitation in Physics, Edition Open Access (2011)
1957
-
[5]
Kennefick,Traveling at the Speed of Thought: Ein- stein and the Quest for Gravitational Waves(Princeton University Press, 2007)
D. Kennefick,Traveling at the Speed of Thought: Ein- stein and the Quest for Gravitational Waves(Princeton University Press, 2007)
2007
-
[6]
Radiation and boundary conditions in the theory of gravitation,
A. Trautman, “Radiation and boundary conditions in the theory of gravitation,” Bull. Acad. Polon. Sci., sér. sci. math., astr. et phys.6, 407–412 (1958)
1958
-
[7]
On the physical significance of the Rie- mann tensor,
F. A. E. Pirani, “On the physical significance of the Rie- mann tensor,” Acta Phys. Polon.15, 389–405 (1956)
1956
-
[8]
Detection and generation of gravitational waves,
J. Weber, “Detection and generation of gravitational waves,” Phys. Rev.117, 306–313 (1960)
1960
-
[9]
Evidence for discovery of gravitational radia- tion,
J. Weber, “Evidence for discovery of gravitational radia- tion,” Phys. Rev. Lett.22, 1320–1324 (1969)
1969
-
[10]
A brief history of gravitational waves,
J. L. Cervantes-Cota, S. Galindo-Uribarri, and G. F. Smoot, “A brief history of gravitational waves,” Universe 2, 22 (2016)
2016
-
[11]
A guide to LIGO-Virgo detector noise and extraction of transient gravitational-wave signals,
B. P. Abbottet al.(LIGO Scientific, Virgo), “A guide to LIGO-Virgo detector noise and extraction of transient gravitational-wave signals,” Class. Quantum Grav.37, 055002 (2020), arXiv:1908.11170
Pith/arXiv arXiv 2020
-
[12]
Open data from the third observing run of LIGO, Virgo, KAGRA and GEO,
R. Abbottet al.(LIGO Scientific, Virgo, KAGRA), “Open data from the third observing run of LIGO, Virgo, KAGRA and GEO,” Astrophys. J. Suppl.267, 29 (2023), arXiv:2302.03676
Pith/arXiv arXiv 2023
-
[13]
GWpy: a Python package for gravitational-wave astrophysics,
Duncan M. Macleodet al., “GWpy: a Python package for gravitational-wave astrophysics,” SoftwareX13, 100657 (2021)
2021
-
[14]
R. Abbottet al.(LIGO Scientific, Virgo, KAGRA), “GWTC-3: Compact binary coalescences observed by LIGO and Virgo during the second part of the third observing run,” Phys. Rev. X13, 041039 (2023), arXiv:2111.03606
Pith/arXiv arXiv 2023
-
[15]
R. Abbottet al.(LIGO Scientific, Virgo), “GWTC-2: Compact binary coalescences observed by LIGO and 10 Virgo during the first half of the third observing run,” Phys. Rev. X11, 021053 (2021), arXiv:2010.14527
Pith/arXiv arXiv 2021
-
[16]
Energy of gravitational radiation and the background energy of the space-time,
J. W. Maluf, F. L. Carneiro, S. C. Ulhoa, and J. F. da Rocha-Neto, “Energy of gravitational radiation and the background energy of the space-time,” Physics Let- ters B858, 139041 (2024)
2024
-
[17]
Overview of KAGRA: De- tector design and construction history,
T. Akutsuet al.(KAGRA), “Overview of KAGRA: De- tector design and construction history,” Prog. Theor. Exp. Phys.2021, 05A101 (2021), arXiv:2005.05574
arXiv 2021
This paper was first reviewed by grok-4.5 on July 14, 2026.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.