{"id":"322083f7-1f26-4d50-abd6-367c704f4d54","arxiv_id":"2607.13706","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Global strong solutions of the Constantin–Johnson morning-glory model exist uniquely for every H¹ initial velocity, without any smallness condition.","lead":"This mathematics paper proves that a two-dimensional model of the 'morning glory' cloud—a rolling cloud formation over northern Australia—has a unique, well-behaved solution that lasts forever, for any starting state with finite H¹ energy. It removes the smallness restriction that limited earlier strong-solution results, upgrading them from small-data or merely local statements.","discovery_kind":"extension","skeptic_critique":null,"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the Constantin-Johnson asymptotic model for the morning glory cloud. It claims global strong well-posedness for initial data v0 ∈ H¹ of arbitrary size, with forcing K ∈ L²_loc([0,∞);L²), improving on the small-data result of Alonso-Orán and Granero-Belinchón. The proof combines a contraction-mapping local existence theorem (Section 3) with global a priori estimates (Section 4): L∞_t L²_x ∩ L²_t H¹_x for v, L∞_t L⁴_x for v, L∞_t L²_x ∩ L²_t H¹_x for the vertical velocity w, and finally L∞_t H¹_x ∩ L²_t H²_x for v. The key novelty is deriving an evolution equation for w and using w-estimates to control the otherwise problematic boundary term in the H¹ estimate. The local well-posedness is obtained via a contraction mapping in X_T, and the global result follows by a maximal-time argument.","tokens_in":11926,"tokens_out":24904,"duration_ms":344348,"significance":"If correct, the result removes the smallness restriction in [1] and is a genuine contribution to the mathematical theory of this atmospheric model. The estimate chain is structurally plausible, and the paper is free of adjustable parameters. The method of introducing and estimating the w-equation is a useful technique that may transfer to related asymptotic models. However, the proof relies on unstated parabolic and elliptic regularity results for the mixed boundary-value problem; these must be supplied for the result to be fully rigorous. The paper is otherwise clearly written and the main argument is credible.","major_comments":[{"comment":"The proof invokes two linear regularity results without statement or proof: (i) at (3.3), 'standard parabolic theory' giving U ∈ C([0,T];H¹) ∩ L²(0,T;H²) with ∂tU ∈ L² for the heat equation with boundary conditions U|z=1 = ∂zU|z=0 = 0 and initial data v0 ∈ H¹; and (ii) in Prop 4.4, the 'elliptic estimate' used to convert ||Δv||₂ control into the full ||∇²v||₂ bound. Both are load-bearing: the first makes the contraction map F land in X_T, and the second closes the H¹ bootstrap. Since the boundary combination (periodic in x, Dirichlet at z=1, Neumann at z=0) is not the pure Dirichlet or pure Neumann case, the authors should state and prove, or give a precise reference for, these maximal-regularity and elliptic estimates. As written, a reader cannot tell whether the domain of the Laplacian and its H¹-interpolant have the required properties.","section":"§3 (Prop 3.1) and §4 (Prop 4.4)"},{"comment":"The evolution equation for w is stated after 'one can verify from (1.1) by direct calculations' but the calculation is not shown. This equation is central: all subsequent w-estimates (Prop 4.3) and therefore the H¹ bootstrap depend on its exact form. In particular, the derivation involves commuting ∂x with ∫_0^z, integration by parts in z, and the boundary term ∂zw|z=1 = 0 inherited from v|z=1 = 0. Please include the derivation, or at least a sufficiently detailed outline, so that the reader can check the signs and boundary terms.","section":"§4, Eq. (4.1)"},{"comment":"The space X_T is defined with the condition ∂zφ|z=0 = 0. For functions in L∞(0,T;H¹) ∩ L²(0,T;H²) the trace of ∂zφ on z=0 is meaningful only for a.e. t>0, not as a continuous condition up to t=0. The initial data v0 is only assumed to satisfy v0|z=1 = 0; if X_T imposes ∂zφ|z=0 = 0 at all times, the fixed point would force ∂zv0|z=0 = 0. The trace interpretation of the boundary conditions should be clarified (e.g., in the sense of L²(0,T;L²_x) for the H² part), and the local well-posedness statement should be framed so that no spurious compatibility condition on v0 is introduced.","section":"§3, definition of X_T"}],"minor_comments":[{"comment":"The abstract states K ∈ L²(0,T;L²) while Theorem 1.1 states K ∈ L²_loc([0,∞);L²). Make the statement consistent.","section":"Abstract and Theorem 1.1"},{"comment":"Typographical errors: 'repectively' in the notation paragraph, and 'soltions' in Remark 1.2. Also some equations have missing spacing.","section":"Section 1"},{"comment":"The proof of Lemma 2.1 is typeset densely and hard to read. Rewrite the interpolation chain with clear parentheses and line breaks.","section":"Lemma 2.1"},{"comment":"The step '≤ μ/2 ||∇²v||₂² + ...' implicitly uses the elliptic estimate on Δv. Add an explicit statement or equation number for this estimate so the reader can see where it enters.","section":"Section 4, Prop 4.4"},{"comment":"Reference [24] is an unpublished preprint. If it is essential for the w-estimate idea, consider providing a detailed statement of the result used or a more public reference.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper appears mathematically sound in its main argument, provided the omitted regularity lemmas are added. The main result would be a nice contribution to the field. The missing linear theory is standard, so the revision should be straightforward. I would encourage acceptance after revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this paper does what it says. It removes the smallness restriction from Alonso-Orán–Granero-Belinchón and upgrades Matioc–Roberti's local strong solutions to global ones, for all H1 initial data and L2 forcing. The strategy is a standard 2D parabolic bootstrap with one genuinely nonstandard ingredient: instead of trying to close H1 directly, they first derive and estimate an equation for the vertical velocity w (their (4.1)), which lets them control the boundary term ∫ w|∂zv|^2 dx at z=1 that would otherwise block the bootstrap. The w-equation is correctly derived, the boundary conditions are consistent (w=0 at z=0, ∂zw=0 at z=1), and the Grönwall coefficients are in L1(0,T). I checked the main calculations; the structure holds up.\n\nCredit where due: the theorem is new with respect to the cited literature, the proof architecture is genuinely different from [1], and the paper is honest about the source of the w-equation idea (Li–Wang [24]). No circularity, no fitted parameters. This is a straightforward, serious contribution to the primitive-equations energy-method literature.\n\nSoft spots, in order of importance. (1) The unstated linear regularity theory: the fixed-point map uses \"standard parabolic theory\" for the heat operator with mixed boundary conditions v|z=1 = ∂zv|z=0 = 0, and Prop 4.4 invokes an elliptic estimate to convert ||Δv||_2 control into full ||∇^2v||_2. These are true for this domain (smooth cylinder, complementary Dirichlet/Neumann), but they are load-bearing and should be stated, with references or proofs. (2) The most delicate absorption step — closing the L4 estimate in Prop 4.2 — is compressed. The final inequality appears to follow by Young with the right powers, but the text jumps from a long chain to the conclusion; a referee will need to see the details. (3) The contraction estimates in Propositions 3.2–3.3 are handled loosely (T^1/4 versus T^1/2 mix), though the logic seems sound. (4) The blow-up criterion (5.1) is asserted without proof; easy to justify, but it should be shown. These are addressable gaps, not demonstrated errors. They are all in the \"written proof needs expansion\" category.\n\nWho this is for: PDE people working on the Constantin–Johnson model or on 2D/3D primitive-equation-type systems. They will get a clear template for how to handle one-sided boundary conditions on the vertical velocity. The paper deserves a serious referee; the core argument is credible and the result, if confirmed, settles the natural open end of the problem.\n\nI recommend sending it to peer review. The gaps I found are fixable and the main theorem is likely correct.","headline":"Global strong well-posedness for large H1 data in the Constantin–Johnson morning-glory model; the proof strategy is credible and the result is new, but the write-up compresses several nontrivial estimate steps and leans on an unstated parabolic regularity theorem.","tokens_in":12528,"tokens_out":2005,"would_cite":false,"duration_ms":536582,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["26D10","35Q35","35Q86","76D03","76D05","86A05","86A10"],"pacs":[],"model":"deepseek-v4-flash","headline":"The morning glory cloud model admits unique global strong solutions for arbitrarily large initial velocity fields.","keywords":["morning glory cloud","global well-posedness","strong solutions","large initial data","a priori estimates","nonlinear wave propagation","primitive equations"],"falsifier":"A concrete way to test the central claim is to numerically solve the system in a periodic channel with initial data having a large H¹ norm (for example, a steep front). The theorem predicts that ||v(t)||_{H¹} remains bounded on any finite interval; observing a computed finite-time blow-up of this norm would contradict it. A more direct mathematical falsifier would be to find a counterexample to the elliptic estimate used in Proposition 4.4 for the mixed boundary conditions — for instance, exhibit a function in the domain for which the L² norm of Δv is finite but the full H² norm is not control","tokens_in":11802,"feed_emoji":"☁️","tokens_out":9073,"duration_ms":80553,"temperature":0.7,"pith_summary":"This paper proves that a model for the morning glory cloud — a system of PDEs describing nonlinear wave propagation in the troposphere — has a unique global strong solution for any initial horizontal velocity in H¹, as long as the thermodynamic forcing is square-integrable in space and time. Previously, global strong solutions were known only when the initial data satisfied a smallness condition. The proof works by establishing a priori estimates: first L² and L⁴ bounds on the horizontal velocity, then an L² bound on the vertical velocity via a derived evolution equation, and finally the crucial H¹ bound that closes the argument. If correct, the result removes the small-data restriction and shows that strong solutions do not blow up in finite time.","feed_headline":"Global strong solutions for large data in morning glory model","feed_subtitle":"New energy estimates remove the small-data restriction, so strong solutions of the morning glory model cannot blow up.","key_machinery":"The key mechanism is a two-step energy estimate. First, the vertical velocity w is expressed through the incompressibility condition as w = -∫₀^z ∂x v dξ, and an evolution equation for w is derived from the original system (equation (4.1)), with boundary conditions w|z=0 = 0 and ∂z w|z=1 = 0. Energy estimates on this w-equation yield sup_t ||w||₂² + ∫_0^T ||∇w||₂² dt ≤ C. This control is then used to bound the boundary term that appears when multiplying the v-equation by -Δv and integrating by parts: the term ∫ w ∂z v ∂z²v can be dominated by a product of the already-controlled quantities, allowing the H¹ norm of v to be bounded over any finite time interval. The local well-posedness is obta","core_discovery":"The central claim is Theorem 1.1: given any initial velocity v0 ∈ H¹(Ω) that is periodic in x and satisfies v0|z=1 = 0, and any forcing K ∈ L²_loc([0,∞);L²(Ω)), the initial-boundary-value problem for the morning glory model has a unique global strong solution. The solution belongs to C([0,T];H¹) ∩ L²(0,T;H²) with ∂t v ∈ L²(0,T;L²) for every finite T. This extends the earlier small-data theory to large data, and the central new step lies in the a priori H¹ estimate: instead of trying to bound ∂zv directly as in the primitive equations, the authors derive an evolution equation for the vertical velocity w and use its energy estimate to control the boundary term that had blocked previous attempt","pith_inferences":["The technique of deriving and estimating the vertical velocity equation could be transferred to other quasilinear systems with one-sided boundary conditions, such as certain atmospheric or oceanic models where a boundary term prevents standard energy estimates.","The paper does not address long-time behavior; for large data, questions of decay, attractors, or time-periodic forcing remain open, and the bounds here are only uniform on finite intervals.","The proof leans on a black-box maximal regularity result for the heat operator with mixed boundary conditions. An explicit verification of that regularity on the periodic channel would make the argument self-contained, though the regularity is expected to hold on the smooth cylinder.","A natural numerical check would be to simulate the model with steep initial fronts and track the H¹ norm; the theorem predicts it stays bounded, while a computed growth to infinity would refute the claim."],"forward_implications":["If Theorem 1.1 is correct, the small-data restriction in the existing literature is removed: strong solutions exist globally for arbitrary H¹ initial data, not just for small L∞ data.","The a priori estimate scheme is robust enough to cover the alternative boundary condition v|z=0 = 0 and the infinite-channel case over R × (0,1), as noted in the remarks.","Since the H¹ norm remains bounded on every finite interval, no finite-time blow-up of strong solutions can occur for this model under the stated assumptions.","The method of controlling a problematic boundary term through an auxiliary evolution equation for the vertical velocity is a generalizable technique for quasilinear systems with one-sided boundary conditions."],"fun_headline_variants":["Morning glory model: strong solutions for all large data","Large-data strong solutions exist for morning glory cloud model","Morning glory cloud: global strong solutions without small data","Strong solutions for morning glory model now proven for large data","No more blow-up: morning glory strong solutions for all data"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The proof assumes that the linear heat equation on the periodic channel with the mixed boundary conditions v|z=1 = ∂z v|z=0 = 0 has the standard L²-based maximal regularity — that is, that a solution with zero initial data and L² forcing lies in L²(0,T;H²) with time derivative in L².","fun_headline_variants_meta":{"raw":{"variants":["Morning glory model: strong solutions for all large data","Large-data strong solutions exist for morning glory cloud model","Morning glory cloud: global strong solutions without small data","Strong solutions for morning glory model now proven for large data","No more blow-up: morning glory strong solutions for all data"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000604,"raw_usage":{"total_tokens":2612,"prompt_tokens":658,"completion_tokens":1954,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":402,"completion_tokens_details":{"reasoning_tokens":1875}},"tokens_in":402,"tokens_out":1954,"duration_ms":13507,"temperature":1.0,"reasoning_tokens":1875,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T04:01:37.384986+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete way to test the central claim is to numerically solve the system in a periodic channel with initial data having a large H¹ norm (for example, a steep front). The theorem predicts that ||v(t)||_{H¹} remains bounded on any finite interval; observing a computed finite-time blow-up of this norm would contradict it. A more direct mathematical falsifier would be to find a counterexample to the elliptic estimate used in Proposition 4.4 for the mixed boundary conditions — for instance, exhibit a function in the domain for which the L² norm of Δv is finite but the full H² norm is not control","supporting_citations":[],"review_version":1}