{"id":"2d00aa8d-6e97-4dd5-9003-c6a146c7e51f","arxiv_id":"2507.18359","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A variational method computes minimal absorbing zones for plane Couette and Poiseuille flows; the zone centers do not reproduce turbulent mean profiles, but the zones give rigorous attractor-containment bounds.","lead":"Scientists use a new optimization method to find the smallest region of flow state space that traps all possible long-term behaviors for two standard channel flows. The center of that region does not match the actual turbulent average, but the region itself offers rigorous bounds for proving when the smooth laminar flow is globally stable.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The computed absorbing-zone radius may be too small because the finite-amplitude critical perturbation is restricted to spanwise-uniform modes; oblique modes are not excluded by the k=0 right-hand side.","rationale":"The reader identified the same weakest assumption; my stress-test agrees. The paper's central claim that the computed minimal zone is an absorbing zone hinges on correctly evaluating the maximum growth rate at every energy. The spanwise-uniform reduction in Section 3 is not justified by the k=0 right-hand side alone, because homogeneous oblique solutions exist at eigenvalue crossings. The cited Joseph and Carmi analogy applies to the laminar base flow, not to the optimized non-laminar shift flows; the paper itself only reports that k1=0 was 'consistently found' in its computations. Since the critical finite perturbation in the reported method is forced to k1=k3=0 even when the infinite-amplitude eigenfunction is spanwise-varying, the finite-energy branch may be different. This concern is concrete and testable with a moderate spectral computation. If the test confirms the one-dimensional result, the central claim stands; if not, the radius is underestimated and the absorbing-zone property fails for the reported zones. Other issues, such as the e_AZ,BF = e_AZ,opt + e_d geometry error and the local-optimality of 'minimal', are real but secondary: they affect the global-stability application or the wording 'minimal' without invalidating the construction of a valid absorbing zone around the shift flow. Thus I recommend no change to the reader's CONDITIONAL verdict: the paper should be accepted only after the reduced-dimension assumption is verified or the claims appropriately qualified.","tokens_in":12038,"tokens_out":14221,"duration_ms":156249,"concrete_test":"Take one optimized shift-flow profile, e.g. plane Couette at Re=100 and Re=300 and Poiseuille at Re=1000, and recompute the finite-amplitude stability threshold without the k1=k3=0 reduction. Represent perturbations in Chebyshev x2 plus Fourier modes with |k1|<=K1 and |k3|<=K3 for K1 and K3 around 8, and for each energy beta maximize dE/dt using the same spectral discretization with an iterative eigensolver or optimization over the divergence-free constraint; sweep beta to locate the largest zero of the maximum. Compare with e_AZ from Fig. 3. Independently, compute the largest eigenvalue of (3.3) on a (k1,k3) grid for these profiles to check whether the maximum occurs at k1=0 and whether any oblique eigenvalue lies close to the critical mu_L. If the full critical energy is larger than the reported e_AZ, the one-dimensional result is not a rigorous absorbing-zone radius.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The reported radius e_AZ is claimed to bound a rigorous absorbing zone: for every perturbation with energy e>e_AZ, dE/dt<0. This requires the global maximum of dE/dt at each energy level to be negative outside the ball. Section 3 restricts the finite-amplitude problem to perturbations with k1=k3=0, arguing that because the right-hand side of (2.10) is non-oscillatory, all oscillatory Fourier components vanish. That argument is only correct when mu_L is not an eigenvalue of the operator in (2.8) for any oblique wavenumber. At resonance, the homogeneous version of (2.10) has nonzero oblique solutions; such modes can be added to the k=0 particular solution while preserving the Euler-Lagrange equations, giving additional stationary points of the constrained problem. The paper's one-dimensional branch may therefore not be the global maximum, and the 'consistent' finding that the largest infinite-amplitude growth rate occurs at k1=0 does not settle the finite-energy zero crossing. If an oblique branch yields a larger critical energy, trajectories starting just outside the reported ball can have dE/dt>0, so the computed zone is not absorbing. Since the central claim and the global-stability application depend on this radius, this is the most load-bearing weakness.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a variational method for computing 'minimal absorbing zones' in the state space of incompressible shear flows. A shift flow U is taken as the centre of a ball measured by perturbation kinetic energy; using the Reynolds–Orr identity, the author shows that if the largest eigenvalue of the infinite-amplitude variational problem is negative, the perturbation energy decays outside a sufficiently large ball. The finite-amplitude Euler–Lagrange equations are then solved to find the critical energy e_AZ at which the maximum growth rate vanishes. A gradient-based optimisation over one-dimensional shift flows is performed for plane Couette and Poiseuille flows, and the resulting optimised profiles are compared with DNS mean profiles and the Spalding law; the comparison is unfavourable, leading the author to reject the random-wandering hypothesis for the turbulent mean. The paper argues that an absorbing zone around the laminar base flow can be constructed and used for global-stability proofs.","tokens_in":12343,"tokens_out":25848,"duration_ms":283608,"significance":"If the computed zones are indeed absorbing, the method provides a rigorous, fully nonlinear containment set for all attractors in wall-bounded shear flows, complementing edge-state and unstable-periodic-orbit analyses. A clear strength is that no parameters are fitted to turbulent-mean data: the shift flow is optimised against the zone-size objective, and the DNS/Spalding comparison is a post-hoc benchmark. The explicit test and rejection of the turbulent-mean hypothesis is scientifically honest and useful. However, the numerical value of e_AZ and the base-flow-zone construction rest on several unproved technical assumptions that must be resolved before the central claims can be accepted.","major_comments":[{"comment":"The reduction of the finite-amplitude problem to spanwise-uniform perturbations (k1=k3=0) is not rigorously justified. The claim that a non-oscillatory right-hand side forces all oscillatory Fourier components to vanish is only valid when μ_L is not an eigenvalue of the operator in (2.8) for any oblique wavenumber pair. At resonance, homogeneous oblique solutions can be added to the particular solution while preserving the Euler–Lagrange equations, and such branches may yield a larger critical energy. The statement that the maximum infinite-amplitude growth rate is 'consistently found' at k1=0 does not settle the finite-energy zero crossing. The author should either prove that the critical μ_L lies outside the spectrum for all k1,k3 or perform a full wavenumber search over the finite-amplitude branches.","section":"§3, Eq. (3.5)"},{"comment":"The radius of the base-flow-centred absorbing zone is not e_AZ,opt + e_d. In the L2 norm associated with kinetic energy, if a ball of radius sqrt(2e_AZ,opt) around U_SF is contained in a ball of radius R around U_BF, the triangle inequality gives R ≥ sqrt(2e_AZ,opt) + sqrt(2e_d), so the required kinetic-energy radius is (sqrt(e_AZ,opt)+sqrt(e_d))^2, not e_AZ,opt+e_d. The printed formula understates the size of the base-flow zone, and the subsequent conclusion that the minimal zone contains the laminar state is not established by the inequality e_d ≤ e_AZ,BF (which is tautological). This affects the quantitative global-stability discussion and should be corrected.","section":"§4, Eq. (4.1)"},{"comment":"Selecting the largest-energy stationary solution with dE/dt=0 does not by itself identify the absorbing boundary, because the Euler–Lagrange equations are only necessary conditions for a constrained extremum. The envelope theorem implies that for the global-maximizer branch the derivative dF/dβ equals μ_L; the reported critical value μ_L=+0.024 would imply that the maximum growth rate increases with energy at the crossing, so that F>0 just outside the claimed zone, contradicting the absorbing-zone property. This suggests either that the selected branch is not the global maximizer or that a sign error is present (Eq. (2.10) has −μ_L while Eq. (3.5) has +μ_L). The author should clarify the sign convention and verify directly that max dE/dt < 0 for all β > e_AZ over a fine scan of μ_L and wavenumbers.","section":"§2.2, Figure 2"},{"comment":"The proof that trajectories 'eventually enter' the absorbing zone requires more than strict negativity of dE/dt outside the ball. If the maximum growth rate approaches zero as the energy approaches the boundary from above, trajectories may approach the zone asymptotically without entering it in finite time. To establish an absorbing set in the standard sense, one needs a uniform negative bound on dE/dt for energies above some slightly enlarged radius, or an explicit differential inequality guaranteeing finite-time entry. The current argument only gives monotone decrease of energy outside the zone, which is insufficient for the 'eventually enters' claim as stated.","section":"§2.1–§2.2"},{"comment":"The optimisation is performed with a gradient-based local method (fmincon) initialised from a continuation path; no evidence is given that the computed shift flow is the global minimiser of e_AZ over all admissible shift flows. Since the objective is non-convex, the reported zone should be described as a locally minimal absorbing zone unless a global search or convexity argument is supplied. The abstract's phrase 'minimal radius' is stronger than what is demonstrated.","section":"§2.3"}],"minor_comments":[{"comment":"There is a sign inconsistency between Eq. (2.10), which contains −μ_L u_i on the left-hand side, and Eq. (3.5), which contains +μ_L u_1. The derivation in the text should be checked and the sign convention stated consistently throughout.","section":"§3, Eq. (3.5)"},{"comment":"The sentence 'it can be seen that e_d ≤ e_AZ,BF ... therefore, the minimal absorbing zone includes the laminar state' is logically insufficient: containing the laminar state requires e_d ≤ e_AZ,opt, not e_d ≤ e_AZ,BF. The comparison in Figure 7 should be made against e_AZ,opt.","section":"§4, p. 11"},{"comment":"The axis labels and the sign of μ_L in Figure 2 need clarification, especially because the caption reports a positive critical μ_L while the text and Eq. (3.5) suggest opposite conventions.","section":"Figure 2"},{"comment":"The reference 'Olivier Dauchot & Paul Manneville 1997' should be formatted consistently with the journal style, and the definition of 'eventually enters' should be made precise.","section":"§1 and §4"},{"comment":"The abrupt change in behaviour near Re_E and the statement about sharp velocity profiles at larger Reynolds numbers could be documented more quantitatively (e.g., by reporting the wall-normal grid resolution tests in a table).","section":"§4, Figure 3"}],"recommendation":"major_revision","confidential_remarks":"The central computational claim rests on the identification of the global maximum of dE/dt on each energy sphere. The sign and branch issues in §2.2 and Figure 2 are serious and need to be resolved before the numerical values of e_AZ can be trusted. If the author can show that the computed branch is indeed the global maximiser (e.g., by direct maximisation over wavenumbers and μ_L) and correct the base-flow-zone radius formula, the paper would be a valuable contribution. The rejection of the turbulent-mean hypothesis is handled honestly and should not be held against the manuscript."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read Nagy's paper. The real contribution is the variational minimization of the absorbing-zone radius over shift flows, and the paper is commendably honest that the zone centroid does not predict turbulent mean profiles. That negative result is well supported and worth publishing.\n\nThe derivation from the Reynolds-Orr identity is standard, but the finite-amplitude Euler-Lagrange formulation and the optimization are new. The numerics look careful—grid convergence checks, no fitted parameters, and the comparison to DNS and the Spalding profile is done after the fact. I believe the negative result.\n\nTwo soft spots, one major, one minor.\n\nMajor: the computed e_AZ is not established as a rigorous absorbing radius. The finite-amplitude problem is solved only for k1=k3=0. The text's reasoning—that the RHS of (2.10) is non-oscillatory so all oscillatory components vanish—is a particular-solution argument. If mu_L coincides with an oblique eigenvalue of the homogeneous operator, oblique modes can be added, giving other stationary points with possibly larger energy. The 'consistent' observation that the infinite-amplitude eigenvalue is largest at k1=0 does not rule this out at finite energy. So the reported zone may be too small, and the central claim that it is absorbing is not proven.\n\nMinor: equation (4.1) sets e_AZ,BF = e_AZ,opt + e_d. To guarantee a base-flow-centered ball lies inside the shift-flow ball, the correct energy radius is (sqrt(e_d)+sqrt(e_AZ,opt))^2, which is larger. So the laminar-centered zone is underestimated by a cross term. This is a geometric slip, not a conceptual one.\n\nThe citation pattern is fair, the key prior work is credited, and the author explicitly lists the limitations of the mean-flow hypothesis. I see no fitted parameters hiding anywhere.\n\nWho is it for: people working on rigorous bounds and global stability in shear flows. As it stands, the method is a promising foundation but not yet a rigorous one. With the oblique-mode gap closed or at least bounded, it would be a solid JFM paper.\n\nRecommendation: accept for peer review. The concerns are addressable, and the core idea deserves refereeing.","headline":"Novel variational method for minimal absorbing zones in shear flows, with an honest negative result on mean-flow prediction, but the rigorous-radius claim is undercut by the one-dimensional restriction.","tokens_in":12780,"tokens_out":4240,"would_cite":false,"duration_ms":43281,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["76D05","76E30"],"pacs":[],"model":"deepseek-v4-flash","headline":"For plane Couette and Poiseuille flow, the paper computes the minimal absorbing zone: a state-space sphere that every trajectory must eventually enter and never leave.","keywords":["absorbing zone","Reynolds-Orr identity","shear flow stability","variational optimisation","plane Couette flow","plane Poiseuille flow","turbulent mean flow","global stability"],"falsifier":"For one of the reported optimal shift flows, say Poiseuille flow at $Re=2000$, solve the infinite-perturbation eigenvalue problem on a grid of nonzero streamwise and spanwise wavenumbers. If any oblique wavenumber pair yields a growth rate larger than the $k_1=k_3=0$ value used in the paper, the minimal zone is not absorbing. A complementary dynamical test is to initialise a resolution-converged simulation just outside the claimed zone and check that the kinetic energy never increases above its initial value before entering; one crossing would contradict the zone's definition.","tokens_in":11820,"feed_emoji":"🌊","tokens_out":10545,"duration_ms":109922,"temperature":0.7,"pith_summary":"The paper sets out to show that incompressible shear flows possess a computable absorbing zone: a hypersphere in state space, centred on an arbitrary 'shift flow', such that every trajectory starting outside the sphere eventually enters it and then stays inside. Because the zone attracts all trajectories, it necessarily contains every attractor, unstable periodic orbit, and edge state of the flow. Existence rests on the Reynolds–Orr identity, which removes the nonlinear advection terms from the kinetic-energy equation and makes energy a Lyapunov-like functional everywhere outside the zone. The paper turns this existence statement into a numerical method, using gradient-based optimisation to shrink the zone to minimal radius for plane Couette and plane Poiseuille flows. The hoped-for bonus, that the zone's centre approximates the turbulent mean flow, is tested against direct numerical simulation data and found quantitatively wanting, but the minimal zone itself survives as a rigorous attractor-containing set and a tool for global-stability arguments.","feed_headline":"Smallest state-space trap for all shear-flow trajectories computed","feed_subtitle":"This minimal absorbing zone bounds every attractor and opens a route to global-stability proofs.","key_machinery":"The load-bearing identity is the Reynolds–Orr kinetic-energy balance: after decomposing the velocity into a stationary shift flow and fluctuations, the nonlinear self-interaction of the fluctuations makes no net contribution to $de/dt$, so outside a sufficiently large sphere the energy derivative is dominated by negative viscous dissipation. Existence of the zone is governed by an infinite-perturbation eigenvalue problem whose largest eigenvalue must be negative; the zone boundary is then set by the largest energy level at which the constrained Euler–Lagrange equations admit a solution with $de/dt=0$, the critical finite perturbation. Minimality is achieved by gradient-based optimisation, where the sensitivity of the zone radius to the shift flow is computed by solving an adjoint linear system; in one-dimensional wall-normal profiles the eigenvalue problem is simplified by a Fourier ansatz in the two homogeneous directions.","core_discovery":"The central claim is that, for any shift flow whose largest infinite-perturbation growth rate is negative, there is a critical energy level at which the maximal kinetic-energy growth rate changes sign; the energy sphere at that level is an absorbing zone. The boundary is found by solving a constrained Euler–Lagrange problem with a prescribed energy level, and the true boundary is the maximal energy at which the energy derivative vanishes. Varying the shift flow under the constraint that the infinite-perturbation eigenvalue remain negative, the paper minimises the zone radius and obtains explicit minimal absorbing zones for Couette and Poiseuille flow as functions of Reynolds number. On the paper's own evidence, the optimal shift-flow profiles are not quantitatively accurate turbulent mean profiles: they have sharper near-wall gradients and, in viscous units, sit well above the law of the wall. What the computation does deliver is a finite, explicit, provable set that contains all attractors, plus an absorbing zone around the laminar state whenever the laminar state lies inside the minimal zone.","pith_inferences":["If the minimal zone radius is read as a lower bound for energy-driven transition, any perturbation with energy below that radius that still becomes turbulent must rely on non-normal transient growth; comparing the zone radius with known minimal seed energies would calibrate how much of transition is linear transient growth rather than direct energy growth.","The zone's failure to track the turbulent mean may be partly a metric artefact: replacing kinetic energy with a weighted norm that still cancels nonlinear terms would yield a different, possibly tighter zone, and the paper's own discussion invites this test.","Applying the same variational construction to pipe flow or boundary layers would show whether the two patterns seen here, near-wall steepening and a plateau below the law of the wall, are universal properties of energy-based absorbing zones or artefacts of plane-channel geometry."],"forward_implications":["Below the classical energy-stability limits ($Re_E=20.6625$ for Couette and $Re_E=49.6035$ for Poiseuille), the minimal absorbing zone collapses to the laminar point, recovering the classical result that no other attractor can exist.","Above those limits the zone has finite, computable radius comparable to the laminar energy, so all turbulence-related attractors are guaranteed to live inside an explicitly known finite-energy set.","Any larger set containing the minimal zone is also absorbing; in particular one can build an absorbing zone around the laminar base flow, and if that zone sits inside the laminar region of attraction, global stability of the base flow is proven.","At high Reynolds number the optimal shift-flow profiles become nearly Reynolds-number-independent and resemble dissipation-maximising variational profiles, but they do not match measured turbulent mean profiles; the paper attributes the mismatch to the use of plain kinetic energy and to non-uniform exploration of state space by turbulent trajectories.","Directly minimising the laminar-centred absorbing zone rather than the shift-flow zone gives about 5 percent smaller guaranteed bounds for Poiseuille flow, while the zone around the shift flow grows by about 10 percent."],"supporting_citations":[{"why":"Introduced the absorbing-zone concept and the energy-based argument that all trajectories enter it.","marker":"Olivier Dauchot & Paul Manneville (1997)"},{"why":"Provides the classical energy-stability analysis and the analogy that locates the maximum growth at zero streamwise wavenumber.","marker":"Joseph & Carmi (1969)"},{"why":"Supplies the dissipation-maximising Couette profile used for comparison with the optimal shift flow.","marker":"Plasting & Kerswell (2003)"},{"why":"Provides the plane-Couette DNS mean profile against which the turbulent-mean hypothesis is tested and rejected.","marker":"Cavalieri & Nogueira (2022)"},{"why":"Supplies the conditional-Lyapunov-function construction used to propose a global-stability proof from the laminar-centred absorbing zone.","marker":"Nagy (2025)"}],"fun_headline_variants":["Tightest provable sphere that traps all shear-flow trajectories","All shear-flow attractors confined to a computable minimal sphere","Variational method yields smallest provable attracting set for shear flows","All shear-flow dynamics eventually trapped in one minimal zone","Explicit minimal absorbing zone bounds every attractor without empiricism"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that for every optimized shift flow the fastest-growing infinite perturbation has no variation in the flow direction and the critical finite perturbation is uniform in the spanwise direction; no proof is given for arbitrary optimized shift flows, and if an oblique perturbation grew faster, the computed zone radius would be too small.","fun_headline_variants_meta":{"raw":{"variants":["Tightest provable sphere that traps all shear-flow trajectories","All shear-flow attractors confined to a computable minimal sphere","Variational method yields smallest provable attracting set for shear flows","All shear-flow dynamics eventually trapped in one minimal zone","Explicit minimal absorbing zone bounds every attractor without empiricism"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000809,"raw_usage":{"total_tokens":3580,"prompt_tokens":1006,"completion_tokens":2574,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":622,"completion_tokens_details":{"reasoning_tokens":2490}},"tokens_in":622,"tokens_out":2574,"duration_ms":20385,"temperature":1.0,"reasoning_tokens":2490,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T18:14:59.339106+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For one of the reported optimal shift flows, say Poiseuille flow at $Re=2000$, solve the infinite-perturbation eigenvalue problem on a grid of nonzero streamwise and spanwise wavenumbers. If any oblique wavenumber pair yields a growth rate larger than the $k_1=k_3=0$ value used in the paper, the minimal zone is not absorbing. A complementary dynamical test is to initialise a resolution-converged simulation just outside the claimed zone and check that the kinetic energy never increases above its initial value before entering; one crossing would contradict the zone's definition.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the classical energy-stability analysis and the analogy that locates the maximum growth at zero streamwise wavenumber."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the dissipation-maximising Couette profile used for comparison with the optimal shift flow."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the plane-Couette DNS mean profile against which the turbulent-mean hypothesis is tested and rejected."},{"cited_title":"Journal of Fluid Mechanics 1014, A25","cited_arxiv_id":null,"evidence_quote":"Supplies the conditional-Lyapunov-function construction used to propose a global-stability proof from the laminar-centred absorbing zone."}],"review_version":1}