Lemma 2.1 — exact affine plane waves over a linear background
OpenTorusEuler.PaperAffine.exactWaveLemma 2.1 of the frozen manuscript (cc:lem:wave). Let be an order-connected interval on which derivatives are unique (UniqueDiffOn). Let with differentiable and everywhere. Let be a matrix-valued function with within- derivative such that is symmetric and on . Let the wave vector and amplitude satisfy the Craik–Criminale equations on (WaveOn):
Then:
- Persistence of transverse data: if at some the vectors are nonzero and perpendicular, they remain nonzero and perpendicular for every ;
- Exact Euler solution: if for all , then the affine wave velocity is divergence-free and solves the Euler equations on with pressure , where ; all time derivatives are taken within .
These exact waves (Craik–Criminale) are the building blocks of the multiscale construction: each step adds one such wave over the linear background formed by the preceding scales.
Formalization Note The amplitude equation is stated with the pressure coefficient already substituted. The Euler conclusion is conditional on transversality holding on all of ; by clause 1 this holds whenever it holds at one time. The profile assumptions used in the paper's construction (odd, -periodic, , mean-zero primitive) are not needed for this local computation and are supplied separately by Lemma 3.1.
import Mathlib import Definitions.Def_TorusEuler_Statement import Definitions.Def_TorusEuler_Affine open Set open scoped InnerProductSpace
namespace TorusEuler.PaperAffine
/-- Lemma 2.1, `cc:lem:wave`. The Euler equations and preservation
of nonzero transverse data hold for the same affine-wave solution. -/
theorem exactWave {J : Set ℝ} (hJ : J.OrdConnected) (hDiff : UniqueDiffOn ℝ J)
{F P : ℝ → ℝ} (hF : Differentiable ℝ F) (hP : ∀ z, HasDerivAt P (F z) z)
{D D' : ℝ → Vec3 →L[ℝ] Vec3} {k a : ℝ → Vec3}
(hD : ∀ t ∈ J, HasDerivWithinAt D (D' t) J t)
(hS : ∀ t ∈ J, IsSelfAdjoint (D' t + (D t).comp (D t)))
(htr : ∀ t ∈ J, LinearMap.trace ℝ Vec3 (D t).toLinearMap = 0)
(hWave : WaveOn J D k a) :
(∀ ⦃t₀ : ℝ⦄, t₀ ∈ J → k t₀ ≠ 0 → a t₀ ≠ 0 → ⟪k t₀, a t₀⟫_ℝ = 0 →
∀ t ∈ J, k t ≠ 0 ∧ a t ≠ 0 ∧ ⟪k t, a t⟫_ℝ = 0) ∧
((∀ t ∈ J, ⟪k t, a t⟫_ℝ = 0) → ∀ t ∈ J, ∀ x,
derivWithin (fun r => affineWaveVelocity F D k a r x) J t +
fderiv ℝ (affineWaveVelocity F D k a t) x (affineWaveVelocity F D k a t x) +
gradient (affineWavePressure P (D' t + (D t).comp (D t))
(affineWavePressureCoeff (D t) (k t) (a t)) (k t)) x = 0 ∧
divergence (affineWaveVelocity F D k a t) x = 0) := by sorry
end TorusEuler.PaperAffineConfirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.