{"id":"c32739be-d615-4b98-a122-7f2736852ab8","arxiv_id":"2506.17002","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A new local elliptic formulation for two-layer constant-vorticity Euler flows yields global bifurcation curves that terminate exactly when conformal equivalence or non-stagnation on the interface breaks down.","lead":"This paper rigorously constructs large-amplitude periodic solutions for steady two-layer Euler flows with piecewise constant vorticity in a channel, using a novel elliptic-system reformulation and global bifurcation theory. It also provides numerical evidence that solution branches end either in interface corners (with stagnation) or in layers of arbitrarily thin width.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The global theorem rests on computer-algebra computations in Proposition 3.1 that are not reproducible; the claimed Lopatinskii lower bounds are asserted without code or intermediate expressions, and an algebraic slip there would invalidate the Fredholm step.","rationale":"The paper is well-structured and the geometric reformulation is plausible, but the central existence theorem depends critically on the ellipticity and Fredholm-index analysis of Section 3. The reader's stated weakest assumption concerned the equivalence of the reformulation to the physical problem and the limiting behavior in (1.3). In my reading, the most fragile point is instead the unverified computer-algebra core of Proposition 3.1: the claimed determinant factorization, root count, and Lopatinskii lower bounds are asserted without enough detail to audit. This does not necessarily mean the theorem is wrong; the authors are experienced and the argument is coherent. However, because the Fredholm property is the entry point to the global bifurcation theorem, a single algebraic error in those computer-algebra computations would invalidate the main result. The reader's rationale also flagged the lack of reproducible computational verification, so my concern overlaps with the reader's overall judgment, though not with the specific weakest-assumption formulation. I therefore do not recommend moving away from the conditional acceptance, but I would make the release of a verification script or the inclusion of the key symbolic computations a condition of full acceptance.","tokens_in":51092,"tokens_out":40335,"duration_ms":442660,"concrete_test":"Independently re-derive the linearized system in Section 3.1 with a symbolic algebra system (e.g., SymPy or Mathematica) at a generic solution point of (2.4): verify that det(A-hat) factors exactly as in (3.1a), that the complex roots of det(A-hat) in nu have the claimed upper/lower half-plane distribution, and that the contour integrals defining Lambda_0 and Lambda_1 yield the stated 5x5 minors with the claimed lower bounds for e^0_dD and e^1_dD. In particular, evaluate detr(mu^1_alpha*) at a non-shear solution and at the shear solution (2.9) with q = q*, and confirm the expression given in Proposition 3.1. If the expressions agree, the Fredholm step is supported; if they differ, the global bifurcation theorem is not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing step is Proposition 3.1, where the linearized operator D_zeta F^circ(zeta,q) is proved to be L-elliptic. The determinant factorization (3.1a), the distribution of the complex roots of det(A-hat), and especially the lower bounds for the Lopatinskii constants e^0_dD and e^1_dD are obtained with the aid of computer algebra, but no code, no intermediate matrices Lambda_0, Lambda_1, and no explicit formulas for the relevant 5x5 minors are supplied. In particular, the assertions detr(mu^0_alpha*) = 1/(32 k^2 |1+k chi_{1x}(x,0)| |xi|) and detr(mu^1_alpha*) = sqrt((u0+q)^2+v0^2)/(16 (1+s_x)^2 k^3) are not checkable from the paper alone. If either computation contains an algebraic error, D_zeta F may fail to be a Fredholm operator of index zero, and Theorem 2.4 cannot be used. This is not a claim of fraud; it is a verifiability gap in the most load-bearing calculation of the proof, and it is not covered by the later analytic sections. An independent symbolic recomputation would either close the gap or reveal a concrete obstruction.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves the first global existence result for large-amplitude steady two-layer constant-vorticity Euler flows in a periodic channel. The authors reformulate the free-boundary problem as a local system of nine first-order PDEs plus two scalar constraints, using conformal maps for each layer composed with a harmonic 'horizontal distortion' s that aligns the two interface parametrizations (Section 2). They verify Agmon-Douglis-Nirenberg L-ellipticity of the linearized operator at all solutions in the admissible set U (Section 3, Proposition 3.1), obtain a Crandall-Rabinowitz branch from shear flows (Section 4), rule out closed loops by a nodal argument (Section 5), and prove that uniform bounds excluding stagnation or conformal breakdown imply compactness (Section 6). Theorem 1.1 concludes that the global curve exists and the quantity (1.3) tends to zero along it, so the only possible terminal behaviors are interface stagnation or loss of the conformal strip equivalence in C^1. Section 7 provides numerical continuation evidence (corner formation or near-wall limits) but does not claim to compute the limiting waves.","tokens_in":51384,"tokens_out":23699,"duration_ms":229004,"significance":"This is a substantial and well-structured contribution. The horizontal-distortion reformulation is a genuine technical innovation: it turns a two-layer free-boundary problem with possible overhangs and stagnation points into a local elliptic system, opening the door to Schauder estimates, Fredholm index theory, and Buffoni-Toland global bifurcation. The proof architecture is complete - local bifurcation, nodal analysis, uniform regularity, and maximality - and the numerics are presented cautiously, with the authors explicitly stating that limiting configurations are not asserted to exist. The numerical section also contains a quantitative, checkable prediction: the conjugate-flow formula (7.8c) is verified to six decimal places in Figure 8, lending credibility to the exploration. If the Lopatinskii computations in Section 3 are correct, the main theorem stands as a strong existence result. The concern that (1.3) does not construct a limiting solution is not, on reading, a flaw: Theorem 1.1 is carefully worded as a statement about limits along the curve, and the paper is explicit that the limiting flow is not recovered.","major_comments":[{"comment":"The Lopatinskii condition is the load-bearing step of the Fredholm analysis, but the two displayed identities - det(mu^0_alpha*) = 1/(32 k^2 |1 + k chi_{1x}(x,0)| |xi|) and det(mu^1_alpha*) = sqrt((u0+q)^2+v0^2)/(16 (1+s_x)^2 k^3) - together with the homotopy lower bounds for e^0_dD(t) and e^1_dD(t) in Section 3.2, are asserted to follow from 'computer algebra' with no code, no scripts, no explicit matrices Lambda_0, Lambda_1, and no intermediate residue computations. Since L-ellipticity of D_zeta F^circ at all solutions is what makes D_zeta F Fredholm (hypothesis (ii) of Theorem 2.4), an algebraic slip in these determinants would invalidate the use of the global bifurcation theorem. Unlike the interior-symbol determinant (3.1a), which is displayed explicitly and is in principle checkable, this part of the proof is not checkable from the manuscript. Please supply either a complete hand-written derivation (entries of Lambda_0 and Lambda_1, the relevant 5x5 minors, and the residues at the three poles (3.2)) or the computer-algebra worksheets/scripts with enough documentation to reproduce every displayed bound.","section":"Section 3.1, Proposition 3.1 (and Section 3.2 homotopy)"},{"comment":"The 'closed' half of the nodal argument contains an unproved rigidity assertion: after applying the strong maximum principle to the scalar elliptic equation (5.3), the paper states that if v0 is constant then 'some straightforward calculation' shows the solution is the trivial shear zeta_* of (2.9). This assertion is load-bearing because it is exactly what rules out alternative (b) of Theorem 2.4 in Corollary 5.4; if a non-shear solution with v0 identically constant existed, the loop alternative would remain open. The calculation needs to show not only that the physical flow in the lower layer is a horizontal shear, but also that the strip parameters h0, h1 and the functions chi_i, eta_i, s equal the specific values in (2.9), which requires use of the scalar constraints (2.4t) and (2.4u). Please include the details.","section":"Section 5, Lemma 5.2"}],"minor_comments":[{"comment":"Theorem 1.1 states that H, k, omega0, omega1 are real with H in (0,1), but the construction requires k > 0: the period 2*pi/k, the positivity of h0 = kH and h1 = k(1-H) in (2.9), and the dispersion relation (4.1) all presuppose it. Please add k > 0 (or at least k != 0) to the hypotheses.","section":"Theorem 1.1 / Section 1.1"},{"comment":"In the paragraph following (1.1), the text says 'the vorticities omega_1 != omega_2 are distinct', but only omega_0 and omega_1 have been defined; presumably omega_0 != omega_1 is meant.","section":"Section 1.1"},{"comment":"The heading reads 'the nodal peroperty (5.1)'; 'peroperty' should read 'property'.","section":"Corollary 5.4"},{"comment":"The matrix displays for the homotopy segment 1 < t <= 2 and for the terminal symbol tilde A(2) are typeset unclearly - for example, the fourth diagonal block appears as '2 xi^2 nu - 2 nu^2 xi' and one row reads '0 0 0 0 xi^2 + nu^2'. These should be re-typeset so the intended matrices can be read.","section":"Section 3.2"},{"comment":"In (7.6) the mesh points are written as 't_I = (I-1) pi / N, i = 1, ..., N-1', with inconsistent index letters; the index should be made consistent.","section":"Section 7.1, equation (7.6)"},{"comment":"In the sentence on 'angular velocities [33, 44, 36]', reference [33] (Fraenkel's book on maximum principles) looks mismatched to the claim; please verify the citation.","section":"Section 1.3"}],"recommendation":"major_revision","confidential_remarks":"The manuscript fits the journal's scope and the main theorem is significant if the Section 3 computations are correct. My recommendation of major_revision hinges almost entirely on making the Lopatinskii verification reproducible; I would ask the editor to require the computer-algebra code or a full analytic derivation before acceptance. The paper's other claims are well supported, prior work [54] is cited appropriately, and the numerical section is honestly framed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis is a serious global bifurcation result, probably the first to handle two-layer constant-vorticity Euler free-boundary problems with both overhanging interfaces and stagnation points. The new idea is real: they compose conformal maps with a harmonic horizontal distortion to get a local elliptic system, and then run Buffoni–Toland global bifurcation. The proof is extensive and mostly checks out structurally: elliptic estimates, nodal analysis ruling out loops, and compactness via C^ell bounds on the EM sets.\n\nWhat is new: Theorem 1.1 gives a continuous curve of solutions for any H, k, omega0, omega1 with omega0 - omega1 = 1, starting at shear flow and terminating with either stagnation on the interface or loss of conformal equivalence to a strip. That is a genuinely new global existence theorem. The reformulation itself is likely to be useful elsewhere, especially for multi-layer problems with density stratification.\n\nCredit where due: the paper is honest about limits. It says the numerical results are qualitative, not proofs. It does not oversell. The related work section is thorough, and the dispersion relation matches [54], which they check.\n\nSoft spots: the load-bearing Lopatinskii computation in Proposition 3.1. They state that the determinant factorization and lower bounds on the Lopatinskii constants are done with computer algebra, but they do not ship the code or even intermediate matrices. For a referee, that is a real verifiability gap. If those constants were wrong, the Fredholm step would fail and Theorem 2.4 could not be used. I see no reason to think they are wrong—the surrounding structure is coherent and the index computation gives a sensible result—but it is not checkable from the paper alone. That is a moderate issue, not a fatal one. Also, the limiting behavior is classified only in the sense of the quantity (1.3) tending to zero; they do not construct actual limiting solutions such as corners or wall-touching profiles. That is fine for a global bifurcation theorem, but readers should not expect the numerical Type I/II boundaries to be proven.\n\nOverall: this deserves a serious referee. I would send it out, with a request that the authors provide the computer algebra scripts or at least enough detail to reproduce the key minors. My own verdict would be conditional accept until that verification is available, but the main theorem is likely correct and important.","headline":"A substantial new global existence theorem for two-layer constant-vorticity Euler waves via a genuinely novel local elliptic formulation; the main proof is sound in structure, but a key computer-algebra step is not reproducible from the paper.","tokens_in":51888,"tokens_out":1559,"would_cite":true,"duration_ms":16633,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35Q31","35B32","76B15","35J47"],"pacs":[],"model":"deepseek-v4-flash","headline":"Two-layer constant-vorticity waves exist along a global curve that ends in stagnation on the interface or breakdown of a layer's strip structure.","keywords":["steady Euler equations","piecewise constant vorticity","interfacial waves","global bifurcation","elliptic systems","conformal mappings","stagnation points","free boundary"],"falsifier":"Numerically continue the first-mode branch for a parameter set such as $(k,H,\\omega_0)=(\\pi,0.45,0)$ until the branch can no longer be extended, and measure the three quantities in (1.3): the reciprocal of the maximum conformal gradient on the interface, the minimum conformal gradient, and the minimum interfacial speed $U^2+V^2$. The theorem claims at least one of these tends to 0; a terminal state with all three bounded strictly away from 0 — for instance a smooth overhanging interface with no stagnation and finite conformal gradient — would refute it.","tokens_in":50902,"feed_emoji":"🌊","tokens_out":16758,"duration_ms":164677,"temperature":0.7,"pith_summary":"The paper proves that in a two-dimensional channel filled with two fluid layers of constant vorticity (uniform local rotation), the lower rotating at rate $\\omega_0$ and the upper at $\\omega_1$ with $\\omega_0-\\omega_1=1$, separated by an unknown interface, there is always a continuous family of periodic travelling waves that starts from a flat shear flow and continues until it approaches a limiting behaviour. The only possible terminal behaviours are a stagnation point on the interface (where the fluid velocity vanishes, the likely site of a corner on the limiting wave) or a breakdown of the angle-preserving (conformal) parameterisation of one layer by a strip, which numerically appears as the layer pinching to zero width or the interface touching a wall. The proof works by reformulating the free-boundary problem as a local elliptic system on a fixed domain, the first such local formulation for a multi-layer problem that allows both overhanging profiles and internal stagnation points, and then applying global bifurcation theory. A sympathetic reader should care because the result establishes large-amplitude existence for all parameter values and pins down the possible limiting behaviours, while the reformulation itself is a tool likely to transfer to other interfacial free-boundary problems in fluid mechanics.","feed_headline":"Two-layer vortex waves run to stagnation or layer collapse","feed_subtitle":"A harmonic stitch between two conformal maps makes the free-boundary problem elliptic and globally tractable.","key_machinery":"The load-bearing object is the reformulated system (2.4): nine unknown functions on the fixed strip $D=\\mathbb{T}\\times(0,1)$ — four velocity components, four coordinate-map components, and the harmonic horizontal distortion $s(x,y)$ — together with two scalar constraints, two scalar parameters $h_0,h_1$, and the bifurcation parameter $q$. The harmonic distortion is the key: composing the upper layer's conformal map with $(x+s(x,y),h_0+h_1-h_1y)$ makes the two layer parameterisations agree along the interface, so the interface conditions become local pointwise equations rather than non-local compositions with an unknown one-dimensional diffeomorphism. In these coordinates the linearised operator is L-elliptic in the standard systems sense (Agmon–Douglis–Nirenberg), giving Schauder estimates, Fredholm properties, and a computable Fredholm index of 0; the abstract analytic global bifurcation theorem stated in the paper (Theorem 2.4) then extends the local branch to a global curve.","core_discovery":"The central discovery is an exact reformulation of the two-layer free-boundary problem as a local elliptic system. Each layer is parameterised by a conformal map from the fixed strip $D=\\mathbb{T}\\times(0,1)$: the lower map $(X_0,Y_0)$ is sense-preserving and the upper map $(X_1,Y_1)$ is sense-reversing, with the interface corresponding to $y=1$ in both. Because two conformal maps generally disagree on the interface, the upper map is composed with an unknown harmonic distortion $S_1(x,y)=(x+s(x,y),h_0+h_1-h_1y)$, whose free function $s(x,y)$ becomes part of the solution; this makes the two interface parameterisations agree pointwise and turns the previously non-local interface conditions into local boundary conditions. The resulting system for the nine unknowns $u_0,v_0,\\chi_0,\\eta_0,u_1,v_1,\\chi_1,\\eta_1,s$ with parameters $h_0,h_1,q$ is shown to be L-elliptic in the Agmon–Douglis–Nirenberg sense at every solution in an open set $U$ of non-degenerate states, so its linearisations are Fredholm of index 0. Global bifurcation theory then produces a solution curve emerging from the shear flow, the nodal property $v_0<0$ (obtained from maximum principles) rules out loops, and the compactness estimates force the termination quantity (1.3) to tend to 0: either $\\sup|\\nabla\\hat Y_i|$ blows up, $\\inf|\\nabla\\hat Y_i|$ vanishes, or the interfacial speed $\\inf(U^2+V^2)$ vanishes.","pith_inferences":["The harmonic-distortion construction should transfer to other two-layer free-boundary problems — internal waves with a density jump, or vortex patches — as long as the interfacial condition can be written as a pointwise differential relation; the paper states this expectation but does not carry it out, so this transfer is an open testable extension rather than a proved result.","If a Type I limiting wave keeps its interface away from the walls, the local singularity theory for rotating vortex patches suggests the corner angle at the stagnation point would be $\\pi/2$; the paper notes the analogy but neither proves the limiting wave exists nor computes its angle.","The numerical phase diagram suggests that the boundary in $(H,\\omega_0)$ between corner-terminating and wall-touching branches might be predicted analytically from the conjugate-flow invariants the paper already uses to compute bore amplitudes; deriving such a criterion is a natural next step the authors leave implicit."],"forward_implications":["For every $H\\in(0,1)$, every wavenumber $k$, and every vorticity pair with $\\omega_0-\\omega_1=1$, a continuous branch of non-trivial periodic solutions exists and is not confined to small perturbations of shear flow.","The interface on this branch is strictly monotone in each half-period, so the constructed large-amplitude waves are not overhanging, even though the formulation admits overhanging profiles.","The only possible terminal states are stagnation somewhere on the interface or a $C^1$ breakdown of the conformal equivalence between a layer and a strip; numerical continuation shows both types occur, as a corner at crest or trough (Type I) or as the interface approaching a wall (Type II).","Near the shear flow the solutions develop cat's-eye stagnation structures, and in the long-wavelength limit they connect to periodic bores and to the companion solitary-wave solutions, with bore amplitudes matching the conjugate-flow formula."],"supporting_citations":[{"why":"Supplies the abstract analytic global bifurcation theorem (stated here as Theorem 2.4) that extends the local branch to a global curve of solutions.","marker":"[9]"},{"why":"Provides the Agmon–Douglis–Nirenberg boundary estimates for elliptic systems used to prove the Fredholm property and uniform regularity.","marker":"[2]"},{"why":"Defines the L-ellipticity framework and the theorem quoted to show the linearised nine-function operator is Fredholm.","marker":"[76]"},{"why":"Supplies the elliptic-system Fredholm theory and the hypotheses used to verify the Lopatinskii condition in Section 3.","marker":"[72]"},{"why":"The single-layer elliptic formulation that this paper extends to two layers; its weaker Fredholm hypothesis is adapted here as Remark 2.5.","marker":"[43]"}],"fun_headline_variants":["Two-layer vortex waves reach stagnation or collapse","Global curves of two-layer vortex waves end in stagnation or pinch","Elliptic reformulation proves full range of two-layer vortex waves","Two-layer vorticity waves: from shear to stagnation or thin-layer limit"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof stands on the reformulated elliptic system (2.4) being exactly equivalent to the physical free-boundary problem throughout the open set $U$ of states whose interface is non-stagnant, non-self-intersecting, and strictly between the walls.","fun_headline_variants_meta":{"raw":{"variants":["Two-layer vortex waves reach stagnation or collapse","Global curves of two-layer vortex waves end in stagnation or pinch","Elliptic reformulation proves full range of two-layer vortex waves","Two-layer vorticity waves: from shear to stagnation or thin-layer limit"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000396,"raw_usage":{"total_tokens":2134,"prompt_tokens":1063,"completion_tokens":1071,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":679,"completion_tokens_details":{"reasoning_tokens":1002}},"tokens_in":679,"tokens_out":1071,"duration_ms":9436,"temperature":1.0,"reasoning_tokens":1002,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T19:12:27.857507+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Numerically continue the first-mode branch for a parameter set such as $(k,H,\\omega_0)=(\\pi,0.45,0)$ until the branch can no longer be extended, and measure the three quantities in (1.3): the reciprocal of the maximum conformal gradient on the interface, the minimum conformal gradient, and the minimum interfacial speed $U^2+V^2$. The theorem claims at least one of these tends to 0; a terminal state with all three bounded strictly away from 0 — for instance a smooth overhanging interface with no stagnation and finite conformal gradient — would refute it.","supporting_citations":[{"cited_title":"Princeton Series in Applied Mathematics","cited_arxiv_id":null,"evidence_quote":"Supplies the abstract analytic global bifurcation theorem (stated here as Theorem 2.4) that extends the local branch to a global curve of solutions."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the L-ellipticity framework and the theorem quoted to show the linearised nine-function operator is Fredholm."},{"cited_title":"Volume 1: Fredholm theory of elliptic problems in unbounded domains, volume 101 ofMonographs in Mathematics","cited_arxiv_id":null,"evidence_quote":"Supplies the elliptic-system Fredholm theory and the hypotheses used to verify the Lopatinskii condition in Section 3."},{"cited_title":"Haziot and Miles H","cited_arxiv_id":null,"evidence_quote":"The single-layer elliptic formulation that this paper extends to two layers; its weaker Fredholm hypothesis is adapted here as Remark 2.5."}],"review_version":1}