{"id":"282fb735-99da-4230-a352-96087e9f5dc2","arxiv_id":"2505.19822","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"For 3D MHD with a rationally aligned background magnetic field, the sharp stability threshold around Couette flow is gamma=1, with inviscid damping and a nu^{-1/3} magnetic amplification.","lead":"This paper proves that a strongly magnetized plasma flow near Couette shear remains stable for a rationally aligned magnetic field when the initial disturbance is smaller than the viscosity, and it identifies the exact stability threshold. It also shows the magnetic field can temporarily amplify by a factor of viscosity to the minus one-third, which differs from the irrational case.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The advertised γ=1 threshold for rationally aligned fields is only proven under |α|>8p; for weaker fields the OLS boundary estimates in Section 3.1 cannot be closed, so the abstract overclaims the rational-alignment result.","rationale":"The reader's weakest assumption correctly identifies the strong-field condition |α|>8p. My independent reading of Section 3.1 confirms that this condition enters through the OLS terms: the spectral gap |σk+l|≥1/p for non-homogeneous modes is used after integration by parts, and the factor p/(2|α|) must be small to absorb the resulting boundary and bulk contributions. Without |α|>8p, the displayed OLS1+OLS2 bound no longer fits into the bootstrap, and no alternative mechanism is presented. This is the most load-bearing limitation because it separates the abstract's unqualified claim (γ=1 for rational σ) from the theorem's actual parameter regime. I do not find an internal inconsistency in the proof under the stated assumptions: the homogeneous-mode estimates, the M3-based zero-mode estimates, and the power counting in the displayed nonlinear bounds appear consistent, and the omitted local well-posedness lemma is a standard ingredient rather than a source of doubt. Therefore the reader's CONDITIONAL verdict is appropriate, and my read does not change it.","tokens_in":33242,"tokens_out":28942,"duration_ms":273515,"concrete_test":"Recompute the OLS1+OLS2 bound in Section 3.1 with |α|=4p, so that p/(2|α|)=1/8, and check whether the boundary terms can still be absorbed by the left-hand side of (3.1); if the factor is not smaller than the available dissipative constants, the bootstrap closes only for |α|>8p, confirming that the abstract must be weakened. Alternatively, with |α|=p/2, the displayed OLS1+OLS2 estimate gives a factor p/(2|α|)=1, so the claimed ε^2/2 absorption in Section 3.1 fails without a new estimate.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim as advertised, namely the γ=1 threshold for σ∈Q, is not what Theorem 1.1 proves. Theorem 1.1 assumes |α|>8p, where p is the denominator of σ=q/p, and this condition is load-bearing. In the OLS estimate of Section 3.1, the non-homogeneous frequency gap |σk+l|≥1/p enters after integration by parts in time; the boundary and bulk terms OLS1+OLS2 are bounded by p/(2|α|)(∥∂X|∇L|MW^2_{≠NH}∥^2_{L∞H^N}+∥∂XXMW^2_{≠NH}∥^2_{L2H^N}). The factor p/(2|α|)<1/16 is what makes these terms absorbable. If |α|≤8p, this factor is not small in the required sense, and the paper supplies no alternative estimate; the bootstrap for (2.9a)-(2.9d) then has no way to control the linear pressure-type terms. Thus the result is a strong-field result for each rational σ, with the required field strength growing with the denominator. The abstract and title, which state the γ=1 threshold for rationally aligned fields without the α restriction, overstate the proven theorem. This is a scope limitation rather than an internal contradiction; if Theorem 1.1 is read literally with |α|>8p, no fatal flaw is apparent. The omitted Lemma 2.1 is standard for 3D MHD and is not the main source of uncertainty.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies the stability threshold for the 3D incompressible MHD system on T x R x T near Couette flow (y,0,0) with a uniform magnetic field alpha(sigma,0,1), where sigma = q/p is rational and nu = mu. The main theorem (Theorem 1.1) asserts that if |alpha| > 8p and the initial perturbation has H^{N+2} norm epsilon <= epsilon_0 nu with N > 9/2, then global stability holds. The proof establishes enhanced dissipation of order nu^{1/3}, inviscid damping of U^2_neq, a nu^{-1/3} amplification of nonzero-mode magnetic fields, and suppression of the lift-up effect for the zero mode, encoded in the estimates (1.3a)-(1.3f). The strategy splits frequencies into homogeneous and non-homogeneous modes with respect to |sigma k + l|, introduces good unknowns W^+- = T^t_{+-alpha}(U +- B), and closes a bootstrap (Proposition 2.1) using the multipliers M1, M2, M3 and energy estimates for the non-homogeneous, homogeneous, and zero modes. The title and abstract advertise the threshold gamma = 1 for rationally aligned fields, but the theorem itself requires the strong-field condition |alpha| > 8p, a point that is load-bearing in the proof.","tokens_in":33589,"tokens_out":9585,"duration_ms":98302,"significance":"If fully verified, the result would improve the known threshold for rational sigma from gamma = 4/3 to gamma = 1, complementing the Diophantine irrational result of Liss and the recent extension by Rao, Zhang, and Zi, and it would provide a new nu^{-1/3} magnetic amplification in low Sobolev regularity. The paper's conceptual tools are well chosen: separating homogeneous modes sigma k + l = 0, using the M3 multiplier of Wei-Zhang, and carefully trading regularity for growth estimates are natural and potentially useful for later work. The main estimates are stated in explicit, checkable form, and the bootstrap does not fit any parameter to the claimed conclusion, which is a genuine strength. However, the proof as written leaves several load-bearing estimates as sketches, and the advertised scope of the result is broader than what Theorem 1.1 actually proves; these issues must be addressed before the result can be fully evaluated.","major_comments":[{"comment":"The abstract and title claim the gamma = 1 threshold for rationally aligned magnetic fields, but Theorem 1.1 assumes |alpha| > 8p, and Section 3.1 shows this assumption is load-bearing. In the OLS estimate following Eq. (3.1), the boundary and bulk terms give OLS1 + OLS2 <= p/(2|alpha|)(||dX|grad L|M W^2_{neq NH}||^2_{L^infty H^N} + ||dXX M W^2_{neq NH}||^2_{L^2 H^N}), and the proof absorbs them by invoking |alpha| > 8p, which makes p/(2|alpha|) < 1/16. For |alpha| <= 8p the paper supplies no alternative estimate, and the bootstrap hypothesis (2.9a) cannot be closed. The proven statement is therefore a strong-field theorem for each rational sigma, with the required field strength growing with the denominator. Please revise the abstract and title to state this restriction explicitly, and add to Remark 1.2 that for a fixed alpha the result covers only denominators p < |alpha|/8.","section":"Sec. 1, Theorem 1.1; Sec. 3.1"},{"comment":"Lemma 2.1, local well-posedness, is stated with the sentence 'we state the following lemma without showing more details' and with no reference. This lemma provides the existence interval [0,2t0] and the initial bounds at t0 that are needed for the continuity argument in Proposition 2.1, so it is part of the proof of the main theorem. Please include a proof, or at least a precise statement of the standard theorem being invoked together with the exact Sobolev regularity and the dependence of t0 on alpha and sigma. As written, the bootstrap has no demonstrated starting point.","section":"Sec. 2.4, Lemma 2.1"},{"comment":"Several estimates that are load-bearing for Proposition 2.1 are omitted as 'similar' or 'left to the reader'. Examples include the remaining contributions of NLT and NLP in Section 3.1, the other contributions of NLS1, NLS2, and NLP in Section 3.2, the NLS2 and NLT terms in Section 4.2, the nonlinear terms in Sections 4.3 and 4.5, and the LU3 and remaining terms in Section 5.1. Some of these involve the same homogeneous-homogeneous interactions and zero-mode terms that the paper identifies as the main difficulty. To make the proof verifiable, please provide complete estimates in an appendix, or give a precise accounting of why each omitted case reduces to a displayed estimate. The current level of detail is not sufficient for a journal referee to certify that the bootstrap closes at the stated nu-scales.","section":"Secs. 3.1, 3.2, 4.2, 4.3, 4.5, 5.1"}],"minor_comments":[{"comment":"The abstract states the threshold 'in H^N (N > 13/2)', while Theorem 1.1 states N > 9/2 and measures the initial data in H^{N+2}. These are consistent after reindexing, but the mismatch should be explained explicitly to avoid an apparent contradiction.","section":"Abstract; Sec. 1"},{"comment":"The theorem states sigma = q/p in Q without requiring gcd(q,p) = 1. Since the condition |alpha| > 8p depends on the chosen representative, please add the reduced-form assumption or state that any denominator p works with the correspondingly weaker bound.","section":"Sec. 1, Theorem 1.1"},{"comment":"In the first energy estimate for dY^L M B^2_neq H, the right-hand side displays dX M U^3_neq H in the initial term and in the L^2 H^{N'} term; these should be dY^L M B^2_neq H. Please correct the typo.","section":"Sec. 4.4"},{"comment":"In the energy estimate for dX^s M B^3_neq H, the initial term is written as dX^s M U^3_neq H(t0); it should be dX^s M B^3_neq H(t0).","section":"Sec. 4.5"},{"comment":"There are several typographical errors: 'magnetic fleld' in the abstract, 'magnetic flied' in the introduction, 'Turning to' spelled 'Turing to' in Section 3.3, and 'simliar' in Section 3.4. Please proofread the text.","section":"Global"}],"recommendation":"major_revision","confidential_remarks":"The central bootstrap is plausible and the strong-field restriction is explicit in Theorem 1.1, so this is not a case for rejection on internal inconsistency. However, the abstract overclaims the rational-alignment result, and the number of omitted estimates is large enough that the paper cannot currently be certified as correct without substantial additional work by the authors. I would be willing to review a revised version that fixes the scope statement and supplies the missing estimates."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Three things to know. First, the core result is a genuine step: for rational σ=q/p, the paper pushes the stability threshold for 3D MHD near Couette from γ=4/3 to γ=1 and answers a question Liss raised in [27]. The new piece is the decomposition into non-homogeneous (σk+l≠0) and homogeneous (σk+l=0) modes, and the ν^{-1/3} amplification of B^1 from exactly resonant modes; that mechanism is not in [27] or [30]. Second, the theorem is not as broad as the abstract sounds. Theorem 1.1 assumes |α|>8p, and that condition is load-bearing—the OLS boundary estimates in Section 3.1 need the factor p/(2|α|) to be small to absorb the linear pressure terms. The abstract and title omit this caveat, so as written they overclaim the rational-alignment case. For weak fields, the bootstrap given here does not close. This is a scope limitation, not an internal contradiction; the proof as stated is plausible under the hypothesis. Third, the paper is honest about its dependencies: Lemma 2.1 on local well-posedness is stated without proof, and many nonlinear estimates are waved off as 'similar' to earlier ones. That is normal for this literature but it means the paper is a high-confidence proof sketch of a large estimate package rather than a fully verified chain. The omissions look routine, not central.\n\nWhat is good: the decomposition is clever, the bootstrap hypotheses are carefully chosen to balance ν-powers and regularity losses, and the explanation of why U^2_H and B^2_H behave differently is clear. I do not see circularity or fitted parameters. The self-citation [31] is to a different 2D problem and is not load-bearing.\n\nWho this is for: anyone working on stability thresholds for MHD or on enhanced dissipation near Couette. The paper deserves a serious referee. A referee should ask for the abstract to be corrected to state the |α|>8p condition, and for Lemma 2.1 to be supplied or cited. I would not desk-reject it.","headline":"Proves γ=1 for rational σ only under a strong-field condition |α|>8p that the abstract omits; the resonant-mode analysis is genuinely new and worth reviewing.","tokens_in":34148,"tokens_out":3033,"would_cite":true,"duration_ms":30998,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35Q35","76E25","76W05"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that for 3D magnetohydrodynamic flow near Couette shear, a rationally aligned magnetic field still yields the optimal stability threshold gamma=1, with nonzero magnetic modes amplifying at most like nu^{-1/3}.","keywords":["stability threshold","3D MHD equations","Couette flow","rational magnetic field","enhanced dissipation","inviscid damping","Sobolev spaces","Fourier multiplier method"],"falsifier":"Run the Fourier-truncated 3D MHD system at a rational $\\sigma$=q/p with $\\alpha$=9p (inside the theorem's range) and initial data of size epsilon_0 nu; if any nonzero magnetic mode grows faster than C $nu^{{-1/3}}$ or the nonzero velocity modes fail to show the $nu^{{1/3}}$ enhanced-dissipation rate for arbitrarily small epsilon_0, the threshold claim fails. A cheaper consistency check is the boundary constant itself: in the energy estimate (3.1) the oscillatory term is bounded by p/(2|$\\alpha$|) times the bootstrap quantities, and |$\\alpha$|>8p is exactly the margin that makes that factor harmless.","tokens_in":33031,"feed_emoji":"🧲","tokens_out":10940,"duration_ms":85938,"temperature":0.7,"pith_summary":"The paper proves that for the 3D incompressible MHD equations near Couette flow, when the background magnetic field is rationally aligned, the stability threshold is gamma=1: initial perturbations of size epsilon_0 nu remain globally controlled, with velocity fluctuations decaying through enhanced dissipation at rate $nu^{{1/3}}$ and the magnetic field's nonzero modes amplifying by at most $nu^{{-1/3}}$. This covers all rational slopes $\\sigma$=q/p under the strong-field condition |$\\alpha$|>8p, and it answers an open question raised in earlier work for generically irrational fields. The proof's central move is to separate Fourier modes where the rational alignment makes the background magnetic term vanish exactly (homogeneous modes) from all other modes, and to show that a loss of one derivative of regularity suffices to control their interaction. If correct, the same gamma=1 threshold holds for rational and Diophantine-irrational alignments alike, in contrast to the weaker gamma=4/3 known for arbitrary $\\sigma$.","feed_headline":"Rational magnetic fields still give the γ=1 MHD Couette threshold","feed_subtitle":"For rational slopes σ=q/p, perturbations of size ν remain stable and magnetic non-zero modes grow only by ν^{-1/3}.","key_machinery":"The proof is carried by two mechanisms working together. First, the rational slope $\\sigma$=q/p forces the resonance set $\\sigma$ k + l = 0 to be a genuine sublattice, so every nonzero mode splits into 'homogeneous' modes ($\\sigma$ k + l = 0, where the background field exerts no restoring force) and 'non-homogeneous' modes (|$\\sigma$ k + l| >= 1/p, where the oscillatory multiplier T^t_{\\pm\\$\\alpha$}=$e^{{\\mp i\\alpha(\\sigma k+l)t}}$ can be integrated by parts). Second, the good unknowns $W^{{\\pm}}$=$T^{{\\pm}}$(U\\pm B) and the vorticity-type variables Q=\\Delta_L U, G=\\Delta_L B reveal that the linearized homogeneous modes decouple: $Q^{2}$_{\\neq H} obeys a purely damped equation, while $G^{2}$_{\\neq H} picks up the stretching term 2\\partial^L_{XY}\\$Delta_L^{{-1}}$$G^{2}$_{\\neq H}, producing the \\$nu^{{-2/3}}$ linear amplification that becomes \\$nu^{{-1/3}}$ after nonlinear coupling. The Fourier multiplier M=$e^{{\\delta_0\\nu^{1/3}}$t}M_1M_2 (and M_3 for the zero mode) converts these mechanisms into the exponential decay rate \\delta_0\\$nu^{{1/3}}$ and the \\$nu^{{-1/6}}$ $L^{2}$-time growth seen in the estimates.","core_discovery":"The paper's central claim is Theorem 1.1: with nu=mu, rational $\\sigma$=q/p, |$\\alpha$|>8p, and N>9/2, any initial datum satisfying ||(u_in,b_in)||_{$H^{{N+2}}$} <= epsilon_0 nu leads to a global solution whose sheared-coordinate profiles satisfy the decay estimates (1.3a)-(1.3f). Concretely, the nonzero velocity modes lose energy at the enhanced rate delta_0 $nu^{{1/3}}$, the $U^{2}$_neq component exhibits nonlinear inviscid damping, the nonzero magnetic modes are amplified by at most $nu^{{-1/3}}$ (with partial_X $B^{1}$_neq reaching $nu^{{-1/2}}$), and the zero modes stay uniformly bounded, so the lift-up mechanism is suppressed. The new content is that these bounds hold for rational $\\sigma$, where the generic Diophantine condition used in prior work fails; the proof replaces it with the exact spectral gap |$\\sigma$ k + l| >= 1/p and isolates the homogeneous modes $\\sigma$ k + l = 0 as the only place where the field's restoring force disappears.","pith_inferences":["If the gap condition |sigma k + l| >= 1/p is the real organizing mechanism, the same gamma=1 threshold should hold for all rational sigma with |alpha|>8p, and possibly for badly approximable irrationals once a suitable quantitative gap is assumed; this is a testable conjecture, not a claim of the paper.","The predicted nu^{-1/3} magnetic amplification at rational sigma could be checked by direct numerical simulation of the linearized system (or of the full MHD equations at small epsilon); seeing a steeper growth of partial_X B^1_neq would indicate a missing nonlinear stabilization channel.","The strong-field cutoff |alpha|>8p is set by the proof's integration-by-parts constant; one would expect the threshold to change character near alpha approx 8p, where the boundary term p/(2|alpha|) crosses the bootstrap margin — this boundary regime is left open."],"forward_implications":["For every rational alignment sigma=q/p with a sufficiently strong field, perturbations of size epsilon_0 nu remain globally stable, matching the gamma=1 threshold already known for Diophantine irrational sigma.","The nonzero magnetic field modes grow by at most nu^{-1/3} (and partial_X B^1_neq by nu^{-1/2}), so the magnetic field does not stay small on the enhanced dissipation time scale; this growth is part of the stable picture, not a sign of instability.","Velocity modes with nonzero Fourier frequency enjoy inviscid damping: U^2_neq decays uniformly in H^{N-2} while the sheared-gradient contribution has the nu^{1/3} enhanced-dissipation rate; this is the first nonlinear inviscid damping statement for u^2_neq in this rational setting.","The zero modes (U^0,B^0) remain bounded in H^N, so the lift-up mechanism that would linearly amplify the streamwise velocity is suppressed at this perturbation size.","As the authors note, the same proof strategy with unequal viscosities nu != mu gives a corresponding statement for data of size min{nu,mu} when |alpha| >= 8p(mu+nu)/sqrt(mu nu)."],"supporting_citations":[{"why":"Supplies the oscillation multiplier T^t_a, the non-homogeneous mode analysis, and the open question for rational sigma that this paper answers.","marker":"[27]"},{"why":"Extends the Sobolev threshold framework to nu != mu and gives the gamma=4/3 general-sigma result that the rational gamma=1 result improves.","marker":"[30]"},{"why":"Provides the sharpened 3D Couette transition-threshold technique, especially using dissipation of partial_X W^{\\pm,3} to control linear stretching.","marker":"[32]"},{"why":"Introduced the multiplier M_3 used here to control the zero-mode nonlinear cascade.","marker":"[33]"},{"why":"Formulated the stability-threshold problem and proved the baseline gamma=3/2 for 3D Couette Navier-Stokes that the MHD thresholds are measured against.","marker":"[3]"}],"fun_headline_variants":["Rational magnetic slopes tighten Couette threshold to γ=1","MHD Couette: rational σ achieves γ=1, magnetic growth capped at ν^{−1/3}","γ=1 stability for MHD Couette with rationally aligned B","Couette flow: rational magnetic fields yield optimal threshold","Nonlinear damping and ν^{−1/3} bounds for rational MHD"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes the background magnetic field is strong enough, |alpha|>8p for sigma=q/p, and without this inequality the oscillatory estimates in Section 3.1 do not close, so the claimed gamma=1 threshold is only proven in the strong-field regime.","fun_headline_variants_meta":{"raw":{"variants":["Rational magnetic slopes tighten Couette threshold to γ=1","MHD Couette: rational σ achieves γ=1, magnetic growth capped at ν^{−1/3}","γ=1 stability for MHD Couette with rationally aligned B","Couette flow: rational magnetic fields yield optimal threshold","Nonlinear damping and ν^{−1/3} bounds for rational MHD"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000336,"raw_usage":{"total_tokens":1882,"prompt_tokens":988,"completion_tokens":894,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":604,"completion_tokens_details":{"reasoning_tokens":795}},"tokens_in":604,"tokens_out":894,"duration_ms":7965,"temperature":1.0,"reasoning_tokens":795,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T14:06:16.791079+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the Fourier-truncated 3D MHD system at a rational $\\sigma$=q/p with $\\alpha$=9p (inside the theorem's range) and initial data of size epsilon_0 nu; if any nonzero magnetic mode grows faster than C $nu^{{-1/3}}$ or the nonzero velocity modes fail to show the $nu^{{1/3}}$ enhanced-dissipation rate for arbitrarily small epsilon_0, the threshold claim fails. A cheaper consistency check is the boundary constant itself: in the energy estimate (3.1) the oscillatory term is bounded by p/(2|$\\alpha$|) times the bootstrap quantities, and |$\\alpha$|>8p is exactly the margin that makes that factor harmless.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the oscillation multiplier T^t_a, the non-homogeneous mode analysis, and the open question for rational sigma that this paper answers."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Extends the Sobolev threshold framework to nu != mu and gives the gamma=4/3 general-sigma result that the rational gamma=1 result improves."},{"cited_title":"Wei and Z","cited_arxiv_id":null,"evidence_quote":"Provides the sharpened 3D Couette transition-threshold technique, especially using dissipation of partial_X W^{\\pm,3} to control linear stretching."},{"cited_title":"Wei and Z","cited_arxiv_id":null,"evidence_quote":"Introduced the multiplier M_3 used here to control the zero-mode nonlinear cascade."},{"cited_title":"Bedrossian, P","cited_arxiv_id":null,"evidence_quote":"Formulated the stability-threshold problem and proved the baseline gamma=3/2 for 3D Couette Navier-Stokes that the MHD thresholds are measured against."}],"review_version":1}