Source-linked AI summary
The Sobolev stability threshold for 2D shear flows near Couette
Jacob Bedrossian, Vlad Vicol, Fei Wang
TL;DR
The paper investigates the Sobolev stability threshold for 2D shear flows near Couette as the inverse Reynolds number ν tends to zero. The analysis first treats Couette flow and then extends its mixing-based energy estimates to shears sufficiently close to Couette, using inviscid damping and enhanced dissipation. ε ≪ ν^1/2 yields global stability estimates for the perturbation, while enhanced dissipation accelerates nonzero-mode decay relative to the heat timescale.
Problem
The paper investigates the Sobolev stability threshold for 2D shear flows near Couette as the inverse Reynolds number ν tends to zero.
Method
The analysis first treats Couette flow and then extends its mixing-based energy estimates to shears sufficiently close to Couette, using inviscid damping and enhanced dissipation.
Results
ε ≪ ν^1/2 yields global stability estimates for the perturbation, while enhanced dissipation accelerates nonzero-mode decay relative to the heat timescale.
Takeaways & Limitations
Finite-regularity stability near Couette persists below a ν^1/2 threshold.
Takeaways & Limitations
The result requires Sobolev exponent N > 1 and does not improve the ν^1/2 threshold within the paper’s treatment.
Abstract
from arXiv · showhide
We consider the 2D Navier-Stokes equation on $\mathbb T \times \mathbb R$, with initial datum that is $\varepsilon$-close in $H^N$ to a shear flow $(U(y),0)$, where $\| U(y) - y\|_{H^{N+4}} \ll 1$ and $N>1$. We prove that if $\varepsilon \ll ν^{1/2}$, where $ν$ denotes the inverse Reynolds number, then the solution of the Navier-Stokes equation remains $\varepsilon$-close in $H^1$ to $(e^{t ν\partial_{yy}}U(y),0)$ for all $t>0$. Moreover, the solution converges to a decaying shear flow for times $t \gg ν^{-1/3}$ by a mixing-enhanced dissipation effect, and experiences a transient growth of gradients. In particular, this shows that the stability threshold in finite regularity scales no worse than $ν^{1/2}$ for 2D shear flows close to the Couette flow.
1. Introduction
The paper studies high-Reynolds-number stability of two-dimensional shear flows near Couette in Sobolev spaces. It identifies an ε ≪ ν^1/2 threshold and explains stability, enhanced dissipation, inviscid damping, and transient growth mechanisms.
- 1. Introduction: The central problem is determining how the nonlinear stability threshold for 2D shear flows scales with the inverse Reynolds number ν in Sobolev regularity.The paper targets perturbations of shear flows close to Couette and seeks the smallest exponent γ such that ε ≪ ν^γ ensures stability.
- 1. Introduction: For perturbations of shears close to Couette, the estimated Sobolev stability threshold is ε ≲ ν^1/2.This threshold is worse than the γ = 0 result available for sufficiently smooth Gevrey perturbations of 2D Couette flow.
- 1. Introduction: Under this threshold, the perturbation remains controlled globally in time around the heat-evolving shear, with estimates separating zero and nonzero x-Fourier modes.The theorem assumes N > 1, ν-independent shear regularity, and small initial perturbation; the nonzero modes satisfy an enhanced dissipation estimate.
- 1. Introduction: The result is weaker in threshold scaling than the corresponding 3D comparison and cannot presently be improved within the paper’s treatment without addressing nonlinear regularity losses.The analysis also requires N > 1, the smallest Sobolev exponent supporting local well-posedness for the underlying 2D Euler equation.
- 1. Introduction: The proof first treats Couette flow, where the shear coordinate change is trivial and the linearized dynamics exhibit inviscid damping and enhanced dissipation, then extends the ideas to nearby decaying shears.The general-shear case requires a more complicated time-dependent coordinate change and additional control of energy transfer and commutators.
- 1. Introduction: Inviscid damping drives the velocity field back toward a shear even at infinite Reynolds number, while the Orr mechanism produces transient growth and regularity loss.The associated gradient growth means uniform stability in every positive Sobolev space is impossible.
2. Stability threshold for the Couette flow
For Couette flow, the proof uses sheared coordinates and a time-dependent Fourier multiplier to establish global stability when the initial H^N perturbation satisfies ε ≪ ν^1/2.
- Proof strategy: The analysis changes coordinates to remove Couette mixing, exploiting the resulting simplifications in the background flow and Biot–Savart law.For Couette, the background shear remains y, the U'' term vanishes, and the Biot–Savart law becomes a Fourier multiplier.
- Stability result: ε ≪ ν^1/2 yields a unique global solution for Couette flow, with constants independent of ν and the initial datum.This is the stated stability threshold for N > 1.
- Proof strategy: A time-dependent Fourier multiplier captures transient unmixing effects and their influence on the nonlinear problem.The multiplier-based norm is equivalent to the standard H^N norm at initial time, simplifying the estimates relative to earlier approaches.
- Bootstrap argument: The argument closes a bootstrap estimate globally by improving its constant from 8 to 4, thereby proving the required a priori bounds for all time.Continuity then forces the maximal bootstrap time to be infinite.
- Bootstrap argument: The critical nonzero-mode estimate is the only step requiring ε ≪ ν^1/2, while the remaining estimates complete the proposition under this condition.The proof combines the nonzero- and zero-frequency estimates with commutator bounds and convolution estimates.
3. Shear flows close to Couette
The paper extends the Couette stability argument to shear flows sufficiently close to Couette by using coordinates adapted to the decaying background shear and proving global estimates under ε ≪ ν^1/2.
- 3. Shear flows close to Couette: The adapted coordinates follow the decaying background shear and preserve the relevant critical mixing times through the relation ∂y ↦ ∂y Ū(∂v − t∂z).The transformation is invertible for sufficiently small δ and is designed to account for the shear’s time dependence.
- 3. Shear flows close to Couette: In the new coordinates, vorticity obeys a transport-diffusion equation coupled to the stream function through the modified Laplacian.The equation contains the additional linear source term b∂zφ and a more complicated Biot–Savart relation than in the Couette case.
- 3.2. Equivalence of the two coordinate systems: The coordinate transformation can be inverted and Sobolev norms transferred between systems using implicit-function and composition estimates.These tools connect the transformed estimates to the original variables and complete the implication to Theorem 1.1.
- 3. Shear flows close to Couette: Under ε ≪ ν^1/2, the transformed perturbation satisfies global-in-time estimates with constants independent of ν and the initial datum.These estimates imply the main theorem after transferring bounds between the adapted and original coordinate systems.
- 3.4. Proof of Theorem 3.7: The proof closes by controlling transport, source, and diffusion-error terms in the vorticity energy estimate, including separate zero and non-zero Fourier modes.The bootstrap estimates improve by continuity, yielding global existence of the controlled interval.
Appendix A. Construction and properties of the multiplier M
Appendix A constructs the Fourier multiplier M used in the stability norm and verifies its required properties, including a sharp ν-dependent loss in one condition.
- Appendix A. Construction and properties of the multiplier M: A multiplier M satisfying conditions (2.6a)–(2.6e) is constructed as a product M = M1M2, with separate differential equations defining its non-zero-mode components.The construction builds M1 and M2 so that the required derivative and positivity properties can be verified.
- Appendix A. Construction and properties of the multiplier M: The multiplier’s properties are established by direct computations and by treating non-zero Fourier modes separately, using |k| ≥ 1.The appendix also relates the construction to multipliers used in earlier work.
- Appendix A. Construction and properties of the multiplier M: The ν^-1/6 prefactor in condition (2.6e) is sharp in the stated sense: 1/6 is the smallest sacrifice needed for a uniform lower bound.The remark explains that near |t − ξ/k| ≥ ν^-1/3, the derivative size of M2 is approximately ν^1/3.
- Appendix A. Construction and properties of the multiplier M: Additional lemmas provide commutator and zero-mode estimates, while showing that Δ_tΔ^-1 can be approximately treated as the identity for sufficiently small δ.These estimates support the energy argument in the main proof.