{"id":"51a0d7e9-3700-4786-a20b-f8657479be8e","arxiv_id":"2412.07283","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For steady Navier-Stokes flow in a 2D wedge, the paper proves exact angle and flux thresholds for each flow type, including a new non-uniqueness range and corrected asymptotics.","lead":"This mathematics paper fully classifies self-similar viscous flows in a wedge-shaped region with no-slip walls, giving sharp conditions on wedge angle and total flow rate for each possible flow pattern. It also discovers flux ranges where two different flows exist for the same data, corrects numerical tables from 1940, and identifies the far-field flow pattern in an aperture domain.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Non-uniqueness for all angles and m≥2 rests on explicitly omitted 'similar' extension of a local computation at one special angle; the two-branch argument is proven only for the special angle, m=1.","rationale":"I read the paper in good faith and found the reduction to the ODE f'' = −f^2 −4f + b, the elliptic-integral reformulations, and the special-angle computations to be careful and internally consistent. The threshold formulas are derived, not fitted, and the comparison with Rosenhead's table is concrete—for example, Φmax(π/2)=0 versus the printed 0.5 is a genuine, checkable correction. I also agree that there is no circularity: the external monotonicity result of Guillod–Wittwer and the aperture existence result of Galdi–Padula–Solonnikov are cited appropriately. The most load-bearing weakness is exactly the one identified by the Reader: the non-uniqueness theorem, which is the paper's advertised new phenomenon, is proved in detail only at one special angle and for m=1, with the general angle and all m≥2 deferred by 'follows the same lines'. Since the sign of the branch-point slope is a delicate analytic fact—Lemma 6.6 relies on explicit cancellations at e1=1 and does not visibly extend to arbitrary e*1(α)—this is not a harmless omission. The reader's conditional verdict is appropriate; my stress-test does not change that verdict, so I recommend UNCHANGED rather than a move to ACCEPT or REJECT.","tokens_in":38537,"tokens_out":4944,"duration_ms":50557,"concrete_test":"Recompute the omitted branch-point expansion for general α. Let e*1(α) solve I+(e,0)=α and let e2(e1) be the unique solution of I1,2(e1,e2)=α from Lemma 6.2. Derive c(α)=lim_{e1→e*1(α)+} [J1,2(e1,e2(e1)) − Φmax(1,1)(α)]/(e1−e*1(α)) using the elliptic-integral representations (30)–(31), and verify c(α)>0 for all α∈(0,π/2), with checks near α→0 and α→π/2. Equivalently, evaluate c(α) by high-precision quadrature at α=0.01, 0.1, 0.5, 1.0, 1.4; if any value is ≤0, the interval non-uniqueness statement for m=1 needs revision. For m=2, repeat the level-set construction for I2,3(e1,e2)=α and check the same two-branch structure before Theorem 1.3(2) is accepted.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central new claim—the existence of a second branch of type (m,m+1) flows for every Φ in the stated interval—is not actually established in the text for general α or general m. In Section 6.1, the proof of Proposition 6.1(2) is carried out only for the special half-angle ¯α = I+(1,0) < π/2. After Lemma 6.6 the paper states that 'the general case α∈(0,π/2) is similar and omitted', and at the end of Section 6.1 it states: 'The proof of Proposition 6.1 when m≥2 follows the same lines, so we omit the details.' The omitted step is not cosmetic: the two-branch conclusion requires showing that, along the level set I1,2(e1,e2)=α, the flux J1,2(e1,e2(e1)) starts with positive slope at the branch point (e*1(α),0) and then tends to −∞, via Lemmas 6.6 and 6.3. Lemma 6.6 computes the slope only at e1=1 using explicit cancellations (equations (52)–(54)); for general α the branch point e*1(α) is implicitly defined by I+(e*1(α),0)=α, and the sign of this slope is not shown. For m≥2, the analogous level-set geometry for Im,m+1 is not displayed at all, and Theorem 7.1(5), which identifies the endpoint solution at maximal flux, is also explicitly omitted. Thus, as written, the rigorous proof of non-uniqueness covers the special angle for m=1; the full statement of Theorem 1.3(1)–(2) is conditional on an unverified 'same lines' argument.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies self-similar (Jeffery-Hamel) solutions of the steady Navier-Stokes equations in a two-dimensional sector with no-slip boundary conditions. After reducing the PDE to the ODE f'' = -f^2 - 4f + b with boundary and flux constraints, the authors express the angle and flux as elliptic integrals of the roots e1,e2,e3 and classify existence by flow type (m+,m-). The main results are: necessary and sufficient flux thresholds Φmax(m+,m-)(α) for existence; uniqueness of pure outflow, pure inflow, and type (m,m) flows; non-uniqueness of type (m,m+1) flows for fluxes between the (m,m) and (m,m+1) thresholds; endpoint characterizations at maximal flux; asymptotics of the thresholds; and an application identifying the leading-order terms of the small-flux aperture-domain solution as type (0,1) upstream and type (1,2) downstream. The proofs rely on monotonicity properties of complete and incomplete elliptic functions, many of which are established in the paper.","tokens_in":38716,"tokens_out":7594,"duration_ms":71149,"significance":"If the full statements hold, the paper gives the first rigorous classification of no-slip Jeffery-Hamel flows in a sector, corrects Rosenhead's numerical table (e.g., Φmax(1,1)(π/2)=0 rather than 0.5), and identifies the previously unspecified aperture-domain asymptotics. The derivations are parameter-free, with independent numerical benchmarks, and the analytic machinery (monotonicity of H(γ), level-set analysis) is carefully developed for the cases that are proved. However, the headline non-uniqueness result is, as written, rigorously established only for one special angle and m=1; the extension to all α and m≥2 is asserted by 'similar' arguments that are not supplied. This is the main weakness.","major_comments":[{"comment":"The non-uniqueness proof is carried out only for the special angle ¯α = I+(1,0). Lemma 6.6 computes the derivative of J1,2 along the level set at e1=1 using the explicit cancellations in (52)-(54), and the text then states 'The general case α∈(0,π/2) is similar and omitted.' This is load-bearing for Theorem 1.3(1): for a general α the branch point e1*(α) is defined implicitly by I+(e1*(α),0)=α, and the sign of the analogous derivative at (e1*(α),0) is precisely what guarantees that every Φ in (Φ(1,1)max(α), Φ(1,2)max(α)) is attained on two distinct branches. Without this computation the non-uniqueness statement is not established for α≠¯α.","section":"Section 6.1, proof of Proposition 6.1(2) (m=1)"},{"comment":"The assertion that the case m≥2 in Proposition 6.1 'follows the same lines' is load-bearing for Theorem 1.3(2) and Theorem 1.2(3). No analogue of Lemmas 6.2-6.6 is stated for Im,m+1, and the topology of the level sets {Im,m+1=α} is not described. In particular, the inequalities Φ(m,m)max(α) < Φ(m,m+1)max(α) ≤ (m/(m+1))Φ(m+1,m+1)max(α) are asserted for m≥2 without proof. The authors should either provide the omitted details or state a version of the theorem limited to the cases actually proved.","section":"Section 6.1, final paragraph"},{"comment":"The dichotomy in Part (1) depends on the assertion that the attainable fluxes along the level set form an interval (-∞, Φ(1,2)max(α)]. The proof gives an upper bound (49) and the limit -∞ from Lemma 6.3, but it does not explicitly state the continuity of J1,2(e1,e2(e1)) on (e1*(α),∞) nor the resulting intermediate-value argument. This is standard and likely repairable, but it should be part of the proof.","section":"Section 6.1, proof of Proposition 6.1(1)"},{"comment":"The endpoint cases at maximal flux are used in the conventions of Theorem 1.3 and in Remark 1.5 to correct Rosenhead's table. Part (5) (type (m,0) solution at Φ=Φ(m,m)max for m≥2) is dismissed with 'similar to Part (4)', and Part (7) is likewise not proved. These are not mere repetitions: the proof of Part (4) relies on the special structure of e1=0 for α≥π/2, which does not directly transfer to the m≥2 case. Please provide the proof or explicitly label these statements as conjectural.","section":"Section 7, Theorem 7.1(5) and (7)"}],"minor_comments":[{"comment":"The word 'abd' in 'Using (33), (34), abd (35)' should be 'and'.","section":"Section 4, proof of Proposition 4.1"},{"comment":"The expression 'α∈(0, pi/2)' uses 'pi' instead of the symbol π.","section":"Remark 7.1"},{"comment":"The displayed condition 'I2,1(e1,e♯ 2(e1) = inf' is missing a closing parenthesis and should read 'I2,1(e1,e♯2(e1)) = inf'.","section":"Section 6.2, proof of Lemma 6.9"},{"comment":"The reference '(65) and (6.2)' appears to be a typo; equation (6.2) is not defined, and the intended reference is likely to the displayed inequality for ∂²I+/∂e2² above (66).","section":"Section 6.2, proof of Proposition 6.7"},{"comment":"The phrases 'The general case α∈(0,π/2) is similar and omitted' and 'follows the same lines, so we omit the details' should be replaced by actual proofs or precise reductions, for the reasons given in the major comments.","section":"Section 6.1"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is honest about its omissions, which are explicitly flagged ('similar and omitted', 'proof ... omitted'), so this is a completeness problem rather than a hidden error. However, because the omitted material is exactly what carries the headline non-uniqueness statement, the paper cannot be accepted in its current form. The authors should be encouraged to supply the missing arguments; the rest of the paper, particularly Sections 3-5 and 8, appears solid."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [Colleague],\n\nThis paper is a serious attempt to close the gap between Rosenhead's numerics and rigorous theory for Jeffery-Hamel flows in a wedge. The core reduction to the ODE f''=-f^2-4f+b is standard, and the thresholds, asymptotics, and the aperture-domain application are genuinely new. The handling of complete and incomplete elliptic integrals is careful, and the comparisons to Rosenhead's table—e.g., Phi_max(pi/2)=0 rather than 0.5—are convincing.\n\nThe parts that are fully written out hold up: pure outflow/inflow classification, the (1,1) and (m,m) periodic flows, and the type (2,1) existence for alpha <= pi/2. The uniqueness proofs are detailed and use the right monotonicity of elliptic integrals. No circularity: the external facts from Guillod-Wittwer and Galdi-Padula-Solonnikov are cited properly.\n\nThe problem is the non-uniqueness claim, Theorem 1.3(1)-(2) and Proposition 6.1(2). The proof is only carried out for the special angle \\bar{\\alpha}=I_+(1,0) and for m=1. The assertions that 'the general case is similar' and that m>=2 'follows the same lines' are not backed by a displayed argument. This is load-bearing: the existence of a second branch for every alpha in (0,pi/2) and every m>=2 is exactly the advertised new phenomenon. A referee should ask for those details. The omission is explicitly acknowledged—the proof of Theorem 7.1(5) is even marked 'omitted'—so the authors are not hiding the gap, but as written the theorem overstates what is proven. This is a conditional accept at best, not a rejection: the missing pieces are likely routine, but they are not routine enough to hand-wave.\n\nResearchers working on Jeffery-Hamel flows, wedge asymptotics, or rigorous treatment of numerical fluid results will get value from it. I'd send it to a serious referee. The value of the classification and the numerics-vs-rigor corrections outweigh the gaps, and the authors should be held to completing the missing arguments. For a reading group, it is a good example of how numerical tables get rigorous confirmations—and how 'similar and omitted' can conceal a nontrivial gap.","headline":"Rigorous classification of Jeffery-Hamel flows in a wedge is mostly solid, but the headline non-uniqueness theorem is only proved for one special angle and m=1, with the general cases explicitly omitted.","tokens_in":39436,"tokens_out":3979,"would_cite":true,"duration_ms":36527,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35Q30","76D05","33E05","34B15","35B40"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that in a wedge of half-angle $\\alpha$ every no-slip self-similar flow type exists exactly up to a critical flux, computes the critical flux from elliptic integrals, proves uniqueness for pure outflow, pure inflow, and…","keywords":["self-similar solutions","Navier-Stokes equations","Jeffery-Hamel flows","no-slip boundary condition","elliptic integrals","flux threshold","non-uniqueness","aperture domain"],"falsifier":"For a wedge half-angle such as $\\alpha=\\pi/4$ that is not the special angle $I_+(1,0)$, trace the level set $I_{1,2}(e_1,e_2)=\\alpha$ numerically and plot the flux $J_{1,2}$ along it: if for some $\\alpha$ the second branch ends before $\\Phi_{\\max}^{(1,2)}(\\alpha)$, the claimed non-uniqueness interval for type $(1,2)$ shrinks; if the same failure occurs for some $m\\ge2$, the non-uniqueness statement would need revision.","tokens_in":38161,"feed_emoji":"🌊","tokens_out":8460,"duration_ms":77899,"temperature":0.7,"pith_summary":"The paper gives a complete existence classification for self-similar radial flows in a two-dimensional wedge with no-slip walls and a prescribed flux. For each flow type $(m_+,m_-)$ it proves a maximum flux $\\Phi_{\\max}^{(m_+,m_-)}(\\alpha)$ exists, with solutions of that type below the threshold and none above. It proves uniqueness of the pure outflow, pure inflow, and $(m,m)$ flows, and it uncovers a new non-uniqueness: fluxes between the $(m,m)$ and $(m,m+1)$ thresholds carry at least two $(m,m+1)$ solutions. The results confirm parts of the 1940 numerical study [23] and correct others, for example the maximum flux for type $(1,1)$ at $\\alpha=\\pi/2$ is $0$, not $0.5$. As an application, the small-flux solution in an aperture domain is shown to have a type $(0,1)$ leading term upstream and a type $(1,2)$ leading term downstream.","feed_headline":"Critical flux decides which wedge flows exist","feed_subtitle":"Full classification of self-similar no-slip flows fixes a 1940 numerical table and names far-field jets.","key_machinery":"The argument rests on the reduction of the Navier-Stokes system to the boundary-value problem $f''=-f^2-4f+b$, $f(-\\alpha)=f(\\alpha)=0$, with the flux as the integral of $f$; the azimuthal velocity vanishes and the pressure is eliminated. Multiplying by $f'$ gives the first integral $(f')^2=Q(f)$, where $Q$ is the cubic $-\\frac23(f-e_1)(f-e_2)(f-e_3)$ with roots summing to $-6$. The angle and flux become elliptic integrals $I(e_1,e_2)$ and $J(e_1,e_2)$ over the interval between consecutive roots, and for $(1,1)$ flows they combine into the identity $\\alpha^2+\\alpha\\Phi/4=H(\\bar\\gamma)$ with $H(\\bar\\gamma)=[(\\bar\\gamma^2-2)K(\\bar\\gamma)+3E(\\bar\\gamma)]K(\\bar\\gamma)$. The strict monotonicity of $H$ and of the root maps reduces existence and uniqueness to a one-parameter level-set analysis, and non-uniqueness appears when the level set of the angle integral has two branches.","core_discovery":"The central claim is that the set of no-slip self-similar solutions in a sector is organised by flux thresholds: for every admissible type with $|m_+-m_-|\\le 1$ the admissible fluxes form a half-line bounded above by $\\Phi_{\\max}^{(m_+,m_-)}(\\alpha)$, and the extremal solution often degenerates to a neighbouring type. The paper establishes the threshold values through complete and incomplete elliptic integrals, proves that the maximum-flux curves are monotone in the angle in the stated ranges, and shows that the flux interval $(\\Phi_{\\max}^{(m,m)}(\\alpha),\\Phi_{\\max}^{(m,m+1)}(\\alpha))$ contains two distinct solutions of type $(m,m+1)$. This gives the first rigorous justification of some entries in [23] and shows that other tabulated entries are numerically inaccurate.","pith_inferences":["Beyond the paper, the same level-set calculation could be run numerically for any intermediate angle to test whether the two-branch structure for $(1,2)$ flows persists; this would convert the asserted 'similar' extension at the end of Section 6.1 from an assumption into a checked fact.","If the threshold mechanism is robust, it should apply to other scale-invariant settings, such as rotated self-similar solutions or sector flows with Navier-slip boundary conditions, where the ODE gets an extra parameter.","The numerical corrections flagged in [23] suggest that other entries of the 1940 tables, especially fluxes at which outflow regions merge, could be recomputed with the same elliptic-integral expressions.","The two solutions sharing one flux are a candidate for flow hysteresis in wedge-shaped channels: the same imposed flux may be reached by different velocity profiles depending on the path of the flux."],"forward_implications":["For half-angle $\\alpha\\ge\\pi/2$, no pure outflow exists; for $\\alpha<\\pi/2$ the pure outflow is unique and its maximum flux satisfies $\\Phi_{\\max}^{(1,0)}(\\alpha)=8(\\pi/2-\\alpha)+o(\\pi/2-\\alpha)$ near the borderline.","Pure inflow exists for every negative flux when $\\alpha\\le\\pi/2$, and only up to a negative threshold when $\\alpha>\\pi/2$; that threshold equals the $(1,1)$ maximum flux.","The $(m,m)$ maximum flux is $m\\Phi_{\\max}^{(1,0)}(\\alpha/m)$, so periodic flows inherit the half-angle restriction of the pure outflow problem.","For every $m\\ge1$ and every flux in $(\\Phi_{\\max}^{(m,m)}(\\alpha),\\Phi_{\\max}^{(m,m+1)}(\\alpha))$, there are at least two $(m,m+1)$ self-similar solutions.","In the aperture domain with small flux, the far-field leading term is a type $(0,1)$ inflow upstream and a type $(1,2)$ outflow downstream, fixing the missing type information in the prior existence theorem [9]."],"supporting_citations":[{"why":"The 1940 numerical study is the baseline table the paper confirms and corrects, including the value $\\Phi_{\\max}^{(1,1)}(\\pi/2)=0.5$ that should be $0$.","marker":"[23]"},{"why":"Supplies the existence and uniqueness of the small-flux aperture-domain solution whose leading-order terms Theorem 1.4 identifies.","marker":"[9]"},{"why":"Supplies the strict monotonicity of $H(\\gamma)=[(\\gamma^2-2)K(\\gamma)+3E(\\gamma)]K(\\gamma)$ used to read off the flux thresholds.","marker":"[10]"},{"why":"Hamel's explicit elliptic-function solutions are the starting point for the full-plane classification extended here to sectors.","marker":"[12]"},{"why":"Jeffery's derivation of the two-dimensional flow family provides the classical solutions being classified.","marker":"[13]"},{"why":"Provides the classification of self-similar solutions in $\\mathbb{R}^2\\setminus\\{0\\}$ and the scaling framework used throughout.","marker":"[24]"},{"why":"Supplies the derivative formula $dK/d\\gamma$ used in the monotonicity argument for the root $e_2(\\bar\\gamma)$.","marker":"[26]"}],"fun_headline_variants":["Flux thresholds dictate wedge flow existence","Self-similar wedge flows: flux rules them all","New flux limits fix 1940 wedge-flow table","Non-unique wedge flows emerge at critical flux","Wedge flow existence pinned by flux bounds"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The non-uniqueness part of the classification assumes that the two-branch level-set analysis carried out at one special wedge angle repeats for every angle in $(0,\\pi/2)$ and for every $m\\ge2$ without essential change; the paper states this extension but does not display the general argument.","fun_headline_variants_meta":{"raw":{"variants":["Flux thresholds dictate wedge flow existence","Self-similar wedge flows: flux rules them all","New flux limits fix 1940 wedge-flow table","Non-unique wedge flows emerge at critical flux","Wedge flow existence pinned by flux bounds"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00064,"raw_usage":{"total_tokens":2919,"prompt_tokens":888,"completion_tokens":2031,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":504,"completion_tokens_details":{"reasoning_tokens":1961}},"tokens_in":504,"tokens_out":2031,"duration_ms":83501,"temperature":1.0,"reasoning_tokens":1961,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T18:55:55.568832+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For a wedge half-angle such as $\\alpha=\\pi/4$ that is not the special angle $I_+(1,0)$, trace the level set $I_{1,2}(e_1,e_2)=\\alpha$ numerically and plot the flux $J_{1,2}$ along it: if for some $\\alpha$ the second branch ends before $\\Phi_{\\max}^{(1,2)}(\\alpha)$, the claimed non-uniqueness interval for type $(1,2)$ shrinks; if the same failure occurs for some $m\\ge2$, the non-uniqueness statement would need revision.","supporting_citations":[{"cited_title":"Rosenhead","cited_arxiv_id":null,"evidence_quote":"The 1940 numerical study is the baseline table the paper confirms and corrects, including the value $\\Phi_{\\max}^{(1,1)}(\\pi/2)=0.5$ that should be $0$."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the existence and uniqueness of the small-flux aperture-domain solution whose leading-order terms Theorem 1.4 identifies."},{"cited_title":"Guillod and P","cited_arxiv_id":null,"evidence_quote":"Supplies the strict monotonicity of $H(\\gamma)=[(\\gamma^2-2)K(\\gamma)+3E(\\gamma)]K(\\gamma)$ used to read off the flux thresholds."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Hamel's explicit elliptic-function solutions are the starting point for the full-plane classification extended here to sectors."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Jeffery's derivation of the two-dimensional flow family provides the classical solutions being classified."},{"cited_title":"ˇSver´ ak","cited_arxiv_id":null,"evidence_quote":"Provides the classification of self-similar solutions in $\\mathbb{R}^2\\setminus\\{0\\}$ and the scaling framework used throughout."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the derivative formula $dK/d\\gamma$ used in the monotonicity argument for the root $e_2(\\bar\\gamma)$."}],"review_version":1}