{"id":"44d573cb-bacd-4413-b171-ebff53c7998f","arxiv_id":"1908.05457","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The paper derives WKB approximations and critical-layer behavior for a Taylor-Goldstein equation modified by horizontal eddy viscosity.","lead":"A mathematician works out approximate wave solutions for a model of stratified shear flow with horizontal turbulent viscosity. The paper gives formulas for how the wave amplitude changes near two special heights, and could help researchers interpret internal gravity waves in the atmosphere.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Central reduction error: Eq. (8) with AT=0 gives Q1=-ik AH'/u2 and Q0=(N^2/u1-U'')/u2 - k^2, not Eqs. (10)-(11); thus the WKB potential (16) and solution (18) solve a different equation.","rationale":"I read the central claim as the asymptotic solution of the reduced equation. The WKB step itself is standard, so the load-bearing point is not the slow-variation assumption; it is whether Eq. (15) with potential (16) is actually the normal form of the model obtained from Eq. (8). The coefficient comparison shows it is not. A typo in Eq. (10) alone would be minor, but the Q1 error changes the Liouville transformation and the potential, so all subsequent formulas must be rederived. The reader's verdict already notes an algebraic flaw; my read makes the flaw more central. No numerical validation is present to compensate. As printed, the central claim is unsupported, so the current version should not be accepted; a corrected derivation could in principle lead to a revisable paper.","tokens_in":17897,"tokens_out":24005,"duration_ms":215857,"concrete_test":"Use a computer algebra system to expand Eq. (8) with AT=0, collect the coefficients of w'', w', and w, and divide by u2=U-c-ikAH. If the printed Q0=(N^2/u1-U'')/u2 and Q1=-ik AH/u2 emerge, the concern is moot; if Q1=-ik AH'/u2 and Q0=(N^2/u1-U'')/u2 - k^2 emerge, then Eq. (18) is not a solution of the stated model. An even simpler spot check is the constant-coefficient limit N=0, U=const, AH=const: Eq. (8) reduces to w''-k^2 w=0, while Eqs. (9)-(11) as printed do not.","verdict_should_be":"REJECT","load_bearing_attack":"In Section 2.2 the paper sets AT=0 in Eq. (8) and then states the modified Taylor-Goldstein equation (9) with Q0=1/u2(N^2/u1-U'') and Q1=-ik AH/u2. Direct rearrangement of Eq. (8) with AT=0 gives, after dividing by u2=U-c-ikAH, u2 w'' - ik AH' w' + (N^2/u1 - U'' - k^2 u2) w = 0, so the normalized coefficients are Q1 = -ik AH'/u2 and Q0 = (N^2/u1 - U'')/u2 - k^2. Both printed coefficients differ: Q1 misses the derivative on AH, and Q0 is missing the -k^2 term. Since Q1 enters the Liouville transformation (12), the potential Qepsilon in Eq. (16) and hence the WKB solution (18) are not the normal form of the stated physical equation. The printed Q1 also cannot generate the dot(AH)^2 term in Eq. (16), because Q1^2 involves AH^2 rather than dot(AH)^2. A minimal check is the constant-coefficient limit N=0, U=const, AH=const, where Eq. (8) reduces to w''-k^2 w=0 but Eqs. (9)-(11) as printed give w'' - [ik AH/(U-c-ikAH)] w'=0. The slow-variation regime is not the first obstacle: the equation being asymptotically solved is a different one.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies internal gravity waves in a stratified shear flow with horizontal eddy viscosity. The authors reduce the fourth-order viscous Taylor-Goldstein equation to a second-order modified Taylor-Goldstein equation when vertical viscosity is absent, then apply a Liouville transformation and a WKB (physical optics) expansion under a slow-variation assumption. They present a leading-order WKB solution, a turning-point analysis giving Airy functions, and a critical-level analysis using Frobenius series. The central claim is that Eq. (18) is an asymptotic solution of the modified Taylor-Goldstein equation, and that near turning points and critical levels the solutions are respectively Airy functions and Frobenius series.","tokens_in":18233,"tokens_out":25675,"duration_ms":210192,"significance":"If the results were correct, the paper would provide useful analytic approximations for vertically propagating internal waves in a viscous stratified shear flow, complementing numerical studies. The paper includes a standard WKB derivation, a worked illustrative example, and a convergence proof for the Frobenius series, which are positive features. However, the significance is severely undercut by fundamental errors in the derivation of the governing equation: the modified Taylor-Goldstein equation as printed is not the reduction of Eq. (8), and the potential used in the WKB analysis does not correspond to the normal form of the physical model.","major_comments":[{"comment":"Setting AT=0 in Eq. (8) gives, after division by u2, the second-order equation w'' - (ik AH'/u2) w' + [(N^2/u1 - U'')/u2 - k^2] w = 0. The printed coefficients Q1 = -ik AH/u2 and Q0 = (N^2/u1 - U'')/u2 are therefore incorrect: Q1 is missing the derivative on AH, and Q0 is missing the -k^2 term. A simple constant-coefficient check with N=0, U=const, AH=const confirms the discrepancy: the physical equation reduces to w'' - k^2 w = 0, whereas Eqs. (9)-(11) give a nonzero first-derivative term. Since Q1 enters the Liouville transformation (12) and hence the potential (16), this error propagates into the central WKB result (18).","section":"Section 2.2, Eqs. (9)-(11)"},{"comment":"Even if Q0 and Q1 are corrected as above, the potential Qε in Eq. (16) is not the result of the Liouville transformation (12) followed by the rescaling Z=εz. Direct computation from the corrected Q1 = -ik AH'/u2 and Q0 = (N^2/u1 - U'')/u2 - k^2 gives a Qε whose ε^2 terms contain additional factors of u2 in the denominators and different signs than the printed (16). Consequently, Eq. (15) with the printed Qε is not the normal form of the physical equation derived from Eq. (8), and the WKB solution (18) is an asymptotic solution of a different equation than the stated viscous stratified shear flow model.","section":"Section 2.2, Eq. (16)"},{"comment":"The critical-level analysis is inconsistent with Eq. (13). The potential Q2 in (13) includes the contributions -Q1^2/4 and -Q1'/2, which for the non-constant viscosity AH(z) produce terms of order ζ^{-2} in the expansion near the critical level. These terms are absent from the printed Q22 in Eq. (25), so the indicial exponents and the oscillatory Frobenius solution (29) are not those of the modified Taylor-Goldstein equation with horizontal viscosity. This makes the critical-level results invalid for the model the paper claims to solve.","section":"Section 4.2, Eqs. (24)-(25)"}],"minor_comments":[{"comment":"The expression for ε after Eq. (18) has denominator A^2+B^2, but the correct denominator is |M0|^2 = sqrt(A^2+B^2). This affects the numerical estimate ε ≤ 0.3 in the example.","section":"Section 3.1, definition of ε"},{"comment":"The formula 'q(z) = M0(z) + ε p(z) / (2|M0(z)|^2)' is ambiguous; it should be clarified whether the division by 2|M0|^2 applies to ε p(z) only or to the whole sum.","section":"Section 3.2, definition of q(z)"},{"comment":"The symbols u3 and u4 in Eq. (16) are introduced without explanation; they should be defined as the coefficients of the ε^2 correction in the potential.","section":"Section 2.2, notation"},{"comment":"The turning point condition N^2 = k^2 u1u2 uses only the leading-order part of Qε; the ε^2 terms are neglected without an explicit statement that this is a leading-order approximation.","section":"Section 4.1, turning point condition"},{"comment":"There is a typo in the historical discussion: 'Jeffrey' should be 'Jeffreys', and 'the trios WKB' should be 'the trio'. In the Abbreviations list, 'Kevin-Helmholtz' should be 'Kelvin-Helmholtz'.","section":"Introduction and Abbreviations"}],"recommendation":"reject","confidential_remarks":"The manuscript's central derivation contains a load-bearing algebraic error: the modified Taylor-Goldstein equation (9)-(11) is not the correct reduction of Eq. (8), and the potential (16) used for the WKB analysis is not the normal form of the physical model. The turning-point and critical-level analyses inherit the same inconsistency. While the WKB methodology itself is standard and the Frobenius convergence proof is a positive feature, the errors are so fundamental that the main results do not apply to the stated physical problem. A major revision would require re-deriving the equations and redoing the asymptotic analysis; in its current form the paper is not suitable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a legitimate asymptotic exercise, but as written the paper has a load-bearing algebra error. The stress-test note is right about the reduction from Eq. (8) to Eqs. (9)–(11). Setting AT=0 and dividing by u2 gives Q1 = -ik AH'/u2 and Q0 = (N^2/u1 - U'')/u2 - k^2, not the printed Q1 = -ik AH/u2 and Q0 = (N^2/u1 - U'')/u2. The constant-coefficient limit is a clean check: Eq. (8) reduces to w'' - k^2 w = 0, while Eqs. (9)–(11) give a different ODE with an AH-dependent first-derivative term. That is not a cosmetic typo; it is the equation the paper claims to solve.\n\nWhat is genuinely good: the derivation from the Navier-Stokes-like momentum equation to the fourth-order ODE is standard and looks correct. The WKB setup is competent, the turning-point reduction to Airy's equation is textbook, and the Frobenius convergence proof is a real contribution—most papers in this space just cite the series without showing it converges. The literature review is also thorough.\n\nWhere it gets softer. First, the printed Eqs. (10)–(11) are inconsistent with Eq. (16): the latter contains (dot AH)^2, which only arises from Q1 = -ik AH'/u2, not from the printed Q1. So either (10)–(11) are typos and (16) is right, or vice versa. The paper never acknowledges the discrepancy, and a reader using (9)–(11) will be solving the wrong equation. Second, the scaling in Eq. (16) looks off. With Z = epsilon z and dot = d/dZ, the epsilon^2/u2 factor produces terms that are epsilon^4 in the original variable; the correct Q2 should have the AH-derivative terms at order 1 but with 1/epsilon^2 in the scaled equation. The paper needs to sort out whether it is solving (13) or a rescaled version. Third, there is no numerical check of the WKB solution against a direct integration of the ODE—for an applied-math paper that is a significant omission. Finally, the critical-level conclusion says the solution \"vanishes,\" while the analysis itself shows indefinite oscillation with no limit; that is misleading.\n\nBottom line: the intended result is plausible and the machinery is sound, but the paper as printed cannot be used without fixing the reduction error and the scaling. It deserves a serious referee, not a desk reject, because the derivation from (8) is correct and the asymptotic framework is standard. I would send it to review with a strong request for major revision, and I would not cite it in its current form.","headline":"Useful WKB analysis of a viscous Taylor-Goldstein variant, but the printed reduction from the fourth-order equation contains an algebraic slip that changes the ODE being solved.","tokens_in":18744,"tokens_out":11109,"would_cite":false,"duration_ms":93205,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["76E05","34E20","76B70"],"pacs":[],"model":"deepseek-v4-flash","headline":"WKB solution tracks internal waves in viscous stratified shear flows","keywords":["stratified shear flow","eddy viscosity","Taylor-Goldstein equation","WKB method","physical optics approximation","turning point","critical level","internal gravity waves"],"falsifier":"Compute $\\varepsilon$ for a realistic atmospheric wind profile with strong shear, such as a jet with a sharp curvature, using the paper's formulas for $M_0$; if $\\varepsilon$ is not much smaller than 1, the asymptotic solution (18) is not valid and will deviate from a direct numerical integration of the modified Taylor-Goldstein equation (15) for that profile.","tokens_in":17682,"feed_emoji":"🌊","tokens_out":6973,"duration_ms":60003,"temperature":0.7,"pith_summary":"The paper derives an approximate solution for the vertical velocity of internal gravity waves in a stratified shear flow when horizontal eddy viscosity is present but vertical eddy viscosity is ignored. Starting from the Navier-Stokes equations under the Boussinesq approximation, it reduces the perturbation equation to a modified Taylor-Goldstein equation of second order, then applies WKB theory to obtain a physical optics solution valid when the background wind and buoyancy frequency vary slowly. It also characterizes the solution near a turning point, where it is governed by Airy functions, and near a critical level, where a Frobenius series with oscillatory behavior applies. If correct, the result gives a closed-form way to study wave propagation and instability without solving the full fourth-order viscous problem.","feed_headline":"WKB solution tracks internal waves in viscous shear flows","feed_subtitle":"Slowly varying eddy viscosity admits an asymptotic wave solution, with Airy and Frobenius behavior at singular levels.","key_machinery":"The load-bearing object is the complex-valued potential $Q_\\varepsilon(Z)$ in (16) and the WKB ansatz (17). The argument proceeds by separating $Q_\\varepsilon$ into real and imaginary parts, writing the WKB exponent $M = M_0 + \\varepsilon M_1 + \\dots$, and solving the lowest-order equations to obtain $m_0$ and $n_0$; the first-order correction supplies the amplitude factor $\\sqrt{|M_0(0)/M_0(z)|}$ and phase correction $\\tfrac12 i(\\alpha-\\alpha_0)$ that convert the geometric optics plane wave into the physical optics solution (18). Near singular points, the same equation is locally approximated by the Airy equation at a turning point and by a regular singular equation with indicial exponents at a critical level.","core_discovery":"The central claim is that the modified Taylor-Goldstein equation (15) with potential (16) admits the physical optics WKB solution (18), provided $\\varepsilon = \\max_{z\\ge 0}|M_0'(z)/M_0^2(z)| \\ll 1$. The solution expresses the vertical velocity perturbation as an amplitude-corrected complex exponential whose phase and amplitude are determined by the leading-order WKB exponent $M_0$ and its integral, with the ratio $m_0/n_0$ entering through a phase shift $\\alpha$. Near a turning point $z_0$ where $N^2 = k^2(U-c)(U-c-ikA_H)$, the governing equation reduces to the Airy equation (20), giving the decaying branch $\\operatorname{Ai}$; near a critical level $z_c$ where $U(z_c)=c_r$, the solution is a Frobenius series (29) with characteristic exponents $\\lambda = \\tfrac12 \\pm \\rho_c e^{i\\varphi_c}$, oscillating infinitely rapidly and converging for all finite distances from the critical level.","pith_inferences":["The analysis is restricted to horizontal eddy viscosity; because vertical eddy viscosity often dominates in the atmospheric boundary layer, extending the WKB treatment to $A_T \\neq 0$ is the natural next test of whether the asymptotic forms survive.","The infinite oscillations of the Frobenius solution near the critical level suggest that the WKB approximation connects to over-reflection and absorption phenomena, which could be tested by computing wave-action flux across the critical level.","The $\\varepsilon$ criterion could be used to classify realistic wind profiles, flagging regions where WKB fails and where a numerical matching method such as the RKWKB integrator is required.","The local Airy and Frobenius solutions could in principle be matched to build a global uniform approximation, giving connection formulas that the paper does not derive."],"forward_implications":["The physical optics solution (18) gives a closed-form expression for the vertical velocity perturbation that can be evaluated without solving the fourth-order viscous equation numerically.","The validity condition $\\varepsilon = \\max_{z\\ge 0}|M_0'/M_0^2| \\ll 1$ provides a quantitative criterion for when horizontal eddy viscosity can be treated by WKB methods.","Near a turning point the solution decays as an Airy function on one side, which describes evanescent behavior of internal gravity waves in a viscous shear flow.","Near a critical level the Frobenius solution oscillates infinitely rapidly and its amplitude vanishes as $z \\to z_c$, consistent with wave absorption at critical levels.","The matrix-exponential formulation with the WKB-constructed fundamental matrix supplies a propagator that can be used in boundary-value solvers for the viscous problem."],"supporting_citations":[{"why":"Provides the extended Taylor-Goldstein equation with eddy viscosity that the present model starts from.","marker":"[48]"},{"why":"Supplies the assumption that eddy coefficients are independent of x and motivates the horizontal-only viscosity case.","marker":"[49]"},{"why":"Shows how vertically varying eddy viscosity and diffusivity enter the modified TG equation, framing the horizontal-only reduction.","marker":"[50]"},{"why":"Classic critical-layer analysis whose lowest-order solution the paper generalizes with the Frobenius series.","marker":"[58]"},{"why":"Source for the WKB validity conditions and asymptotic-series requirements used in Section 3.1.","marker":"[55]"},{"why":"Reference for the critical-level analysis and the form of the Frobenius solution.","marker":"[71]"},{"why":"Gives the definition of the small parameter via $|M'/M^2|$ and the WKB validity criterion for vertically varying shear flows.","marker":"[93]"}],"fun_headline_variants":["Viscous shear waves: WKB asymptotics at singular levels","Physical optics for stratified shear with viscosity","Airy and Frobenius branches in viscous wave asymptotics","Modified Taylor-Goldstein: WKB wave solution","Turning and critical levels in viscous shear WKB"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the WKB slow-variation condition $\\varepsilon = \\max_{z\\ge 0}|M_0'(z)/M_0^2(z)| \\ll 1$ holds for the atmospheric profiles of interest, but the paper only demonstrates this for one illustrative profile with weak shear.","fun_headline_variants_meta":{"raw":{"variants":["Viscous shear waves: WKB asymptotics at singular levels","Physical optics for stratified shear with viscosity","Airy and Frobenius branches in viscous wave asymptotics","Modified Taylor-Goldstein: WKB wave solution","Turning and critical levels in viscous shear WKB"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000805,"raw_usage":{"total_tokens":3519,"prompt_tokens":912,"completion_tokens":2607,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":528,"completion_tokens_details":{"reasoning_tokens":2530}},"tokens_in":528,"tokens_out":2607,"duration_ms":19969,"temperature":1.0,"reasoning_tokens":2530,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:13:32.771155+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute $\\varepsilon$ for a realistic atmospheric wind profile with strong shear, such as a jet with a sharp curvature, using the paper's formulas for $M_0$; if $\\varepsilon$ is not much smaller than 1, the asymptotic solution (18) is not valid and will deviate from a direct numerical integration of the modified Taylor-Goldstein equation (15) for that profile.","supporting_citations":[{"cited_title":"Instability and hydraulics of turbulent stratified shear flows","cited_arxiv_id":null,"evidence_quote":"Provides the extended Taylor-Goldstein equation with eddy viscosity that the present model starts from."},{"cited_title":"The effect of small viscosity and diffusivity on the marginal stability of stably stratified shear flows","cited_arxiv_id":null,"evidence_quote":"Supplies the assumption that eddy coefficients are independent of x and motivates the horizontal-only viscosity case."},{"cited_title":"Destabilization of a stratified shear layer by ambient turbulence","cited_arxiv_id":null,"evidence_quote":"Shows how vertically varying eddy viscosity and diffusivity enter the modified TG equation, framing the horizontal-only reduction."},{"cited_title":"The critical layer for internal gravity waves in a shear flow.J","cited_arxiv_id":null,"evidence_quote":"Classic critical-layer analysis whose lowest-order solution the paper generalizes with the Frobenius series."},{"cited_title":"Advanced Mathematical Methods for Scientists and Engineers; Springer: Berlin, Germany, 2010","cited_arxiv_id":null,"evidence_quote":"Source for the WKB validity conditions and asymptotic-series requirements used in Section 3.1."},{"cited_title":"An Introduction to Atmospheric Gravity Waves, 2nd ed.; Elsevier: Amsterdam, the Netherlands, 2013","cited_arxiv_id":null,"evidence_quote":"Reference for the critical-level analysis and the form of the Frobenius solution."},{"cited_title":"An analytical model of mountain wave drag for wind profiles with shear and curvature","cited_arxiv_id":null,"evidence_quote":"Gives the definition of the small parameter via $|M'/M^2|$ and the WKB validity criterion for vertically varying shear flows."}],"review_version":1}