{"id":"b6241aac-681f-4d24-a5c3-54e702dc0b8e","arxiv_id":"2506.01002","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"In a global ocean simulation, pair dispersion is controlled by slow, balanced currents even when internal waves dominate the small-scale kinetic energy spectrum.","lead":"This paper uses a global ocean simulation to test whether ocean eddies or internal waves control how pairs of floating particles drift apart. It finds that internal waves add energy to small-scale currents but barely affect particle separation, which matters for interpreting new satellites like SWOT.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The causal claim that IGWs do not affect dispersion is inferred from spectral consistency alone; no wave-filtered Lagrangian experiment is run, leaving the wave-vortex partition and its errors untested.","rationale":"I read the paper as a careful, well-executed diagnostic study of LLC4320. Its strengths include two energetic regions, a clear seasonal contrast, multiple independent Lagrangian indicators, and unusually honest reporting of uncertainties, including the explicit admission that the smallest FSLE separations are close to the inertial-oscillation scale. The reader's CONDITIONAL verdict is appropriate. The most load-bearing weakness is not an internal inconsistency but an untested causal inference: the paper shows that the balanced component of the Eulerian spectrum has the shape expected for the observed dispersion regime, and then infers that waves do not affect dispersion. This is a consistency argument, not a direct test. A direct Lagrangian experiment with wave-filtered velocities would settle whether the partition actually matters for particle statistics, and it is feasible with the same dataset because the spatiotemporal transform and advection machinery already exist. If that test reproduces the full-field summer FSLE, the central claim is genuinely supported and the partition is validated causally; if not, the claimed insensitivity to IGWs would need to be abandoned or substantially qualified. Given the paper's appropriate hedging ('consistent with the hypothesis', 'no evidence of an impact'), the conditional verdict stands rather than a rejection. My concern is the same one the reader identified, so I agree with the weakest-assumption analysis and would not change the verdict.","tokens_in":18775,"tokens_out":3483,"duration_ms":39869,"concrete_test":"Advect a fresh set of N ≈ 3600 triplets in each region and season using the low-frequency velocity field obtained by zeroing all (k, ω) components with ω² > f²(1 + L_R²k²) (i.e., the inverse transform of the green-dot spectrum in Fig. 10), then recompute the FSLE and relative diffusivity. If the flat summer plateau of Fig. 7 is reproduced with the balanced-only flow, the conclusion holds; if the plateau becomes scale-dependent or shifts, IGWs (or misclassified high-frequency motions) do affect dispersion. As a robustness check, repeat with L_R varied by ±50% and with a frequency-only filter at f and M2 to test the sensitivity of the partition to its defining parameters.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central inference rests on the wave-vortex partition of Sec. 5b: using the 10th-baroclinic-mode dispersion relation ω² = f²(L_R²k² + 1) with L_R = 65 km (winter) and 20 km (summer), the authors separate IGWs from balanced motions and show in Sec. 5c that the balanced summer spectrum scales as k^-3, consistent with the observed nonlocal FSLE plateau. The paper then concludes, in the abstract and conclusions, that high-frequency IGWs do not impact relative dispersion. The load-bearing assumption is that this partition cleanly separates waves from balanced motions. This is not independently validated: the criterion is a linear, single-mode dispersion curve applied in regions with strong mean flows, Doppler shifting, and a continuum of vertical modes; ageostrophic submesoscale motions with frequencies near f or M2 could be misclassified, and L_R itself is regionally and seasonally uncertain. If the k^-3 balanced spectrum is partly an artifact of the partition, the agreement with the Lagrangian plateau is coincidental and the causal claim collapses. Moreover, the authors state in the Conclusions that the inertial-oscillation scale V/f ≈ 4.59 km is close to the smallest FSLE separation (≈ 4.17 km), so the diagnostic may not even resolve wave effects if present; this is a self-declared limitation. No experiment advects particles in a wave-filtered velocity field, which would directly test the causal claim rather than infer it from Eulerian spectra.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"Using hourly surface velocities from the global MITgcm LLC4320 simulation, the authors advect synthetic particle triplets (initial separation R0≈3.48 km) for 30 days in February and August 2012 in the Kuroshio Extension, with a companion analysis in the Gulf Stream. They compute relative dispersion, relative diffusivity, separation kurtosis, and the FSLE, and compare the inferred dispersion regimes with the slopes of Eulerian wavenumber kinetic energy spectra. Winter results are broadly consistent with local dispersion and a β≈2 spectrum; summer results show nonlocal dispersion (flat FSLE, δ² diffusivity) despite a spectral slope β≈2.3. Using frequency-wavenumber spectra and a 10th-baroclinic-mode dispersion relation, the authors partition the flow into low-frequency balanced and high-frequency IGW components and find a summer balanced spectrum close to k^-3, which they argue resolves the discrepancy and implies that IGWs do not affect relative dispersion in these regimes.","tokens_in":19043,"tokens_out":7356,"duration_ms":77232,"significance":"The empirical contribution is solid: the analysis uses a state-of-the-art global simulation with tidal forcing, applies multiple independent Lagrangian diagnostics with bootstrapped uncertainties, and reproduces the main seasonal contrast in two energetic western-boundary-current regions. If the causal interpretation is accepted, the paper provides a useful framework for predicting surface dispersion from balanced, geostrophic velocities and for interpreting SWOT and ODYSEA data. The main limitation is that the central causal claim is inferential: no particle experiment with filtered wave fields is performed, and the wave-vortex partition is not independently validated. Given the acknowledged scale limitation near V/f, the paper's conclusions are somewhat stronger than the evidence directly supports.","major_comments":[{"comment":"The wave-vortex partition is the load-bearing step for the summer interpretation, but it is not validated for the regions and seasons studied. The partition is obtained by integrating E(k,ω) on either side of the dispersion curve ω² = f²(L_R² k² + 1) for the 10th baroclinic mode, with L_R = 65 km (winter) and 20 km (summer). In the presence of strong mean flows, Doppler shifting, and a continuum of vertical modes, a single linear dispersion curve can misclassify ageostrophic submesoscale motions as waves or, conversely, label wave energy as balanced. If the k^-3 summer spectrum of the low-frequency component is partly an artifact of this choice, the agreement with the flat FSLE would be coincidental. Please provide sensitivity tests over a plausible range of L_R and vertical mode number, or an independent validation of the partition, before the causal claim is accepted.","section":"Sec. 5b, Fig. 10"},{"comment":"The paper's title and abstract assert that high-frequency IGWs do not impact relative dispersion, but the evidence is spectral consistency, not a controlled Lagrangian experiment. A direct test is feasible: advect the same particle set in a velocity field from which the high-frequency component (ω² > f²(L_R² k² + 1)) has been removed, and compare the resulting FSLE and relative diffusivity with Figs. 5 and 7. Without such a test, the possibility remains that IGWs contribute to dispersion in a way that is masked by the dominant balanced strain, or that the two components interact nonlinearly. I consider this experiment, or an equivalently strong causal identification strategy, necessary to support the stated conclusion.","section":"Sec. 5c and Sec. 7"},{"comment":"The authors explicitly note that the inertial-oscillation scale V/f ≈ 4.59 km is close to the smallest FSLE separation (δ ≈ 4.17 km) and that resolving smaller scales would require higher-resolution simulations. This is an acknowledged blind spot: any wave influence on relative dispersion at scales near or below the inertial scale is not sampled. Consequently, the conclusion 'No evidence of an impact of internal waves on pair dispersion was found' is only valid for separations larger than roughly 4 km and should be qualified accordingly in the abstract and conclusions.","section":"Sec. 7, inertial-oscillation paragraph"}],"minor_comments":[{"comment":"The winter spectral slope is reported as β≈2 with a fit-range-dependent range 5/3≲β≲2.4; later the winter FSLE exponent γ=0.29 is converted to β≈2.4 and called compatible with the upper bound. The paragraph would be more transparent if it stated that the winter agreement relies on the upper end of the spectral-slope uncertainty.","section":"Sec. 3, Fig. 3"},{"comment":"The phrase 'integrated using a fourth-order Runge-Kutta method and TRACMASS in space' mixes time integration and trajectory scheme; please clarify whether TRACMASS is used for spatial interpolation and cite the appropriate interpolation method.","section":"Sec. 2"},{"comment":"Please define the Lagrangian kinetic energy spectrum E(ω) and its normalization; the axis label E(t) [m²/s] appears inconsistent with a frequency-domain quantity.","section":"Sec. 5a, Fig. 8"},{"comment":"There is a typo, 'a wealth a smaller eddies', which should read 'a wealth of smaller eddies'.","section":"Sec. 3"},{"comment":"The filtered spectra are shown without uncertainty shading, whereas the total spectra and Lagrangian diagnostics have bootstrap intervals; since the k^-3 slope of the low-frequency component is central, please add uncertainty estimates for the filtered components.","section":"Figs. 10 and 14"}],"recommendation":"major_revision","confidential_remarks":"I agree with the conditional assessment in the reader's report: the empirical analysis is careful, but the causal claim needs either a direct filtered-advection experiment or a clearly restricted conclusion. Should the authors add such an experiment, I would support acceptance; without it, the paper is a valuable diagnostic study whose title and abstract should be softened. No concerns about novelty or attribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nRead this if you care about whether internal waves affect surface dispersion, or about what SWOT velocities can and cannot tell us. The paper is a solid, careful analysis of LLC4320 in two western boundary currents, and the main result is the seasonal contrast: in winter, Lagrangian FSLE, diffusivity, and kurtosis all line up with the full kinetic energy spectrum (local dispersion, beta ~2-2.4); in summer, the full spectrum would predict local dispersion (beta ~2.3) but the FSLE is flat (nonlocal). That mismatch is real and it is the interesting hook.\n\nWhat is new is not the individual diagnostics—FSLE and spectral bridging are standard—but the reconciliation. Using frequency-wavenumber spectra and a 10th-baroclinic-mode dispersion relation, they show the summer total spectrum is dominated by internal waves at scales below about 50 km, and the balanced, rotational part of the flow scales as k^-3, which is exactly the steep spectrum expected for the observed nonlocal dispersion. The same pattern appears in both Kuroshio and Gulf Stream, which helps. The statistics are handled honestly: bootstrapped uncertainties, independent computation of enstrophy for the ballistic check, and they flag where the FSLE fit is ambiguous.\n\nThe soft spot is the causal claim. The paper concludes internal waves do not impact dispersion, but no experiment ever advects particles in a wave-filtered velocity field. The argument is spectral consistency: the balanced spectrum matches the Lagrangian regime, therefore waves are irrelevant. That is plausible but it leans on the wave-vortex partition, which uses a linear, single-mode dispersion curve with L_R that shifts from 65 to 20 km seasonally, in a region with strong mean flow and Doppler shifting. Misclassify a bit of ageostrophic submesoscale energy as wave energy and the k^-3 balanced spectrum could be partly an artifact. The authors, to their credit, also note that the smallest FSLE separation (about 4.2 km) is close to the inertial-oscillation scale V/f (4.6 km), so the diagnostic may not even resolve wave effects if they exist. These are real caveats, but they are stated in the paper, not hidden. The paper itself uses hedged language in the abstract and conclusions, which is appropriate.\n\nThe bottom line is that this is a good paper with a conditional central inference. It deserves a serious referee and, with the causal claim softened or tested, it is publishable. I would bring it to a reading group to discuss the wave-vortex decomposition, and I would cite it when thinking about submesoscale dispersion. Recommend: send to review, ask for a direct test of the wave-filtered advection or a clearer statement that the result is consistency-based, not causal.","headline":"A careful, honest analysis of a global simulation that cleanly shows a seasonal contrast in dispersion regimes, with a plausible but not fully tested conclusion that internal waves do not affect pair dispersion.","tokens_in":19599,"tokens_out":2628,"would_cite":true,"duration_ms":25035,"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":"Internal waves add ocean energy but do not control how drifters spread.","keywords":["Lagrangian pair dispersion","internal gravity waves","balanced motions","kinetic energy spectrum","finite-size Lyapunov exponent","submesoscale turbulence","Kuroshio Extension","Gulf Stream"],"falsifier":"Run the same Lagrangian experiments with the high-frequency wave component removed from the velocity field by zeroing all energy above the dispersion-relation curve, and compare the finite-size Lyapunov exponent to the full-field result; the paper's claim predicts identical curves down to about 4 km. Alternatively, launch pairs at initial separations well below the inertial-oscillation scale, for example 1 km, in the full field: if wave motions then change the finite-size Lyapunov exponent or produce a local dispersion scaling absent in the filtered field, the claim that internal waves do not affect dispersion fails at those scales.","tokens_in":18523,"feed_emoji":"🌊","tokens_out":7804,"duration_ms":71058,"temperature":0.7,"pith_summary":"This paper asks whether the way two floating particles drift apart can be read off the ocean's kinetic-energy spectrum, as quasi-geostrophic turbulence theory predicts. Using a global simulation that resolves both meso- and submesoscale eddies and internal gravity waves, the authors find that the rule holds in winter but appears to fail in summer: the summer spectrum is shallow enough to predict local dispersion, yet particles separate in a nonlocal regime dominated by large eddies. The discrepancy dissolves when the spectrum is split into slow, nearly balanced motions and fast internal waves. The wave part dominates kinetic energy at scales below about 50 km, while the balanced part shows the steep $k^{-3}$ spectrum that nonlocal dispersion requires. The paper concludes that high-frequency internal waves barely affect relative dispersion at the resolved scales, and that the balanced, rotational flow alone controls how pairs spread.","feed_headline":"Internal waves add ocean energy but not drifter spreading","feed_subtitle":"A simulation shows summer's wave-dominated spectrum hides a steeper balanced flow that actually governs pair dispersion.","key_machinery":"The load-bearing tool is the frequency-wavenumber kinetic-energy spectrum $E(k,\\omega)$ partitioned by the linear internal-gravity-wave dispersion relation $\\omega^2=f^2(L_R^2 k^2+1)$, using the 10th baroclinic mode, the highest resolved baroclinic mode in the LLC4320 simulation, with deformation radius $L_R\\simeq65$ km in winter and $L_R\\simeq20$ km in summer. This curve divides motions into slow, nearly balanced meso- and submesoscale motions and fast internal-wave motions; integrating each side separately yields the balanced-only and wave-only wavenumber spectra. The companion diagnostics are the finite-size Lyapunov exponent $\\lambda(\\delta)$, whose power-law decay signals local dispersion and whose plateau signals nonlocal dispersion, and the Helmholtz decomposition into rotational and divergent kinetic energy. Together these tools let the authors attribute the summer small-scale spectral energy to waves while showing that the balanced spectrum is steep ($k^{-3}$), reconciling the Lagrangian indicators with spectral theory.","core_discovery":"In both the Kuroshio Extension and the Gulf Stream, winter dispersion is local, with the finite-size Lyapunov exponent decaying as a power of separation and tracking a kinetic-energy spectrum with exponent $\\beta\\simeq2$ to $2.4$, whereas summer dispersion is nonlocal, with an extended plateau in the Lyapunov exponent indicating strain from the largest scales, even though the measured total spectrum has $\\beta\\simeq2.3$, which would predict local dispersion. By forming frequency-wavenumber spectra and partitioning energy along the 10th-baroclinic-mode dispersion relation $\\omega^2=f^2(L_R^2 k^2+1)$, the authors show that in summer internal waves dominate the total spectrum at scales below roughly 50 km, while the nearly balanced, mainly rotational component follows $E(k)\\sim k^{-3}$. The steep balanced spectrum is exactly what predicts nonlocal dispersion, so the apparent inconsistency disappears. The paper's central claim is that high-frequency internal gravity waves do not measurably impact relative dispersion in this simulation, and that surface pair spreading is controlled by the balanced, larger-scale flow component.","pith_inferences":["A direct numerical test of the paper's hypothesis would be to advect synthetic particles in the full velocity field and in the frequency-filtered balanced field alone; the paper's claim predicts nearly identical finite-size Lyapunov exponent curves down to the smallest resolved separations.","The same frequency-wavenumber partition could be applied to SWOT-derived surface velocities in other regions and seasons to map where filtered spectra predict dispersion better than raw spectra, a test the paper does not perform.","If the claim generalizes, biologically and chemically relevant tracer spreading at scales above roughly 10 km could be estimated from balanced altimetry alone, with wave-driven dispersion confined to scales smaller than the inertial-oscillation scale.","The paper's null result could be sharpened by seeding pairs at separations below 1 km in a yet-higher-resolution simulation, where the inertial-oscillation and tidal scales would be resolved and wave effects would have room to appear."],"forward_implications":["In summer conditions like these, the total kinetic-energy spectrum is not a reliable predictor of relative dispersion; the frequency-filtered balanced spectrum is.","Geostrophic velocities from wide-swath altimeters such as SWOT may describe the flow component that actually controls pair dispersion, provided high-frequency wave energy is filtered out rather than aliased into the field.","In winter, where internal waves are weak, satellite-derived geostrophic currents should support direct Lagrangian predictions of surface dispersion.","The insensitivity to waves is limited to the resolved separation range; at initial separations at or below the inertial-oscillation scale, about 4.6 km, waves may still contribute, and the present experiment cannot rule that out.","Future missions that measure the low-frequency surface current component could fill the gap where waves dominate the energetic spectrum in summer."],"supporting_citations":[{"why":"Provides the global LLC4320 simulation whose surface velocity fields are used to advect the synthetic particles.","marker":"Forget et al. 2015"},{"why":"Supplies the frequency-wavenumber spectrum methodology and the 10th-baroclinic-mode dispersion-relation partition between balanced motions and internal gravity waves.","marker":"Torres et al. 2018"},{"why":"Establishes the dimensional bridging relations between spectral slope and local versus nonlocal dispersion regimes that the paper tests.","marker":"Foussard et al. 2017"},{"why":"Reviews the Lagrangian statistics and spectral-link predictions used to interpret relative dispersion, diffusivity, and kurtosis.","marker":"LaCasce 2008"},{"why":"Presents the opposing result that high-frequency motions substantially increase small-scale diffusivity, which the paper's findings contest.","marker":"Sinha et al. 2019"},{"why":"Shows that internal-wave effects on pair dispersion depend on the specific meso- and submesoscale flow, framing the season-dependent response studied here.","marker":"Wang et al. 2018"},{"why":"Quantifies where geostrophy holds in a tide- and eddy-resolving model, supporting the paper's interpretation of SWOT-derived velocities.","marker":"Yu et al. 2021"},{"why":"Compares SWOT-derived and real drifter dispersion and finds balanced motions dominate above about 10 km, aligning with the winter behaviour reported here.","marker":"Tranchant et al. 2025"},{"why":"Argues that fixed-lengthscale indicators should feel inertial oscillations, a claim the paper's summer finite-size Lyapunov exponent does not reproduce.","marker":"Beron-Vera and LaCasce 2016"}],"fun_headline_variants":["Internal waves add ocean energy but not drifter spreading","Ocean simulation: internal waves don't stir drifters","Waves don't disperse drifters, balanced flow does","Internal waves off the hook for ocean surface spread","Dispersion at sea ruled by balanced motions, not waves"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion rests on the assumption that a single wave-frequency curve cleanly sorts every motion into slow balanced eddies versus fast internal waves; if some fine-scale eddy motions are misclassified as waves, the steep balanced spectrum that explains summer dispersion disappears.","fun_headline_variants_meta":{"raw":{"variants":["Internal waves add ocean energy but not drifter spreading","Ocean simulation: internal waves don't stir drifters","Waves don't disperse drifters, balanced flow does","Internal waves off the hook for ocean surface spread","Dispersion at sea ruled by balanced motions, not waves"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000283,"raw_usage":{"total_tokens":1709,"prompt_tokens":1021,"completion_tokens":688,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":637,"completion_tokens_details":{"reasoning_tokens":610}},"tokens_in":637,"tokens_out":688,"duration_ms":6568,"temperature":1.0,"reasoning_tokens":610,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T11:53:36.444569+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same Lagrangian experiments with the high-frequency wave component removed from the velocity field by zeroing all energy above the dispersion-relation curve, and compare the finite-size Lyapunov exponent to the full-field result; the paper's claim predicts identical curves down to about 4 km. Alternatively, launch pairs at initial separations well below the inertial-oscillation scale, for example 1 km, in the full field: if wave motions then change the finite-size Lyapunov exponent or produce a local dispersion scaling absent in the filtered field, the claim that internal waves do not affect dispersion fails at those scales.","supporting_citations":[{"cited_title":"Berti, X","cited_arxiv_id":null,"evidence_quote":"Establishes the dimensional bridging relations between spectral slope and local versus nonlocal dispersion regimes that the paper tests."},{"cited_title":"H., 2008: Statistics from Lagrangian observations","cited_arxiv_id":null,"evidence_quote":"Reviews the Lagrangian statistics and spectral-link predictions used to interpret relative dispersion, diffusivity, and kurtosis."},{"cited_title":"Balwada, N","cited_arxiv_id":null,"evidence_quote":"Presents the opposing result that high-frequency motions substantially increase small-scale diffusivity, which the paper's findings contest."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows that internal-wave effects on pair dispersion depend on the specific meso- and submesoscale flow, framing the season-dependent response studied here."},{"cited_title":"Legresy, A","cited_arxiv_id":null,"evidence_quote":"Compares SWOT-derived and real drifter dispersion and finds balanced motions dominate above about 10 km, aligning with the winter behaviour reported here."},{"cited_title":"J., and J","cited_arxiv_id":null,"evidence_quote":"Argues that fixed-lengthscale indicators should feel inertial oscillations, a claim the paper's summer finite-size Lyapunov exponent does not reproduce."}],"review_version":1}