Refined global well-posedness for the periodic modulated Korteweg-de Vries equation

Technical Summary: Refined Global Well-Posedness for the Periodic Modulated Korteweg-de Vries Equation

Problem Statement
The paper addresses the global well-posedness of the modulated Korteweg-de Vries (KdV) equation on the circle $\mathbb{T} = \mathbb{R}/(2\pi\mathbb{Z})$:
$$\partial_t u + \partial_x^3 u \cdot \partial_t w = \partial_x (u^2), \quad u|_{t=0} = u_0,$$
where $u$ is the unknown and $w: \mathbb{R} \to \mathbb{R}$ is a continuous, potentially highly irregular modulation function. The equation models weakly nonlinear long waves in inhomogeneous waveguides.

Previous work (Chouk, Gubinelli, Li, Li, and Oh, 2024) established global well-posedness in negative Sobolev spaces $H^s(\mathbb{T})$ for the modulated KdV when the modulation $w$ is sufficiently irregular (specifically $(\rho, \gamma)$-irregular). However, that result was restricted to the scaling subcritical regime $s > -3/2$. This barrier arose from the reliance on the classical KdV scaling symmetry, which limits the applicability of the $I$-method (method of almost conservation laws) to regularities above the scaling critical index $s_{cr} = -3/2$. The standard KdV equation is known to be ill-posed for $s < -1$, and the modulated version was previously only proven globally well-posed down to the scaling critical threshold, regardless of the irregularity of $w$.

Methodology
The authors develop a refined approach combining three main tools:

1. Pathwise Approach via Nonlinear Young Integration: Following the framework of Chouk and Gubinelli, the equation is treated pathwise. The time derivative of the modulation $\partial_t w$ is avoided by using a Duhamel formulation involving the modulated linear propagator $U_w(t) = e^{-w(t)\partial_x^3}$. The solution is analyzed in a "modulated interaction representation," transforming the equation into a nonlinear Young differential equation (YDE) of the form $u = u_0 + I_X(u)$, where $I_X$ is a nonlinear Young integral defined via the sewing lemma.
2. The $I$-Method: To extend local solutions to global ones, the authors employ the $I$-method. This involves defining a smoothing operator $I$ (a Fourier multiplier) that maps $H^s$ to $L^2$ for $s < 0$. The goal is to control the growth of the modified energy $\|Iu\|_{L^2}^2$.
3. Novel Scaling Transforms: The core innovation is the introduction of a new one-parameter family of scaling transforms for the pair $(u, w)$. Unlike the classical KdV scaling where $u_\lambda(t, x) = \lambda^{-2}u(\lambda^{-3}t, \lambda^{-1}x)$, the authors define:
   $$u_\lambda(t, x) = \lambda^{-b+1}u(\lambda^{-bt}, \lambda^{-1}x), \quad w_\lambda(t) = \lambda^3 w(\lambda^{-bt}),$$
   where $b \in \mathbb{R}$ is a free parameter. This exploits the additional degree of freedom provided by the modulation term $w$. By choosing $b$ appropriately (specifically $b > 3/2 - s$), the authors can render the problem scaling-subcritical even for $s \le -3/2$.

Key Contributions and Results
The primary contribution is the proof of global well-posedness for the modulated KdV in $H^s(\mathbb{T})$ for any $s \in \mathbb{R}$, provided the modulation $w$ is sufficiently irregular. This breaks the scaling critical barrier $s = -3/2$.

The main result (Theorem 1.5) establishes global well-posedness under specific conditions on the irregularity parameters $\rho$ and $\gamma$ of $w$ and the regularity $s$ of the initial data. The conditions are categorized as follows:

• For $1/2 < \rho \le \frac{3\gamma}{6\gamma-2}$, global well-posedness holds for $s > (\frac{3}{2} - 3\rho)\gamma$.
• For $\frac{3\gamma}{6\gamma-2} < \rho \le \frac{3}{2}$, global well-posedness holds for $s \ge -\rho$.
• For $\rho \ge \frac{3}{2}$, the range of $s$ depends on $\gamma$. Notably, if $\gamma$ is sufficiently small (specifically $1/2 < \gamma < \frac{\sqrt{5}-1}{2}$) and $\rho$ is large enough, global well-posedness is achieved for any $s \in \mathbb{R}$.

The proof involves:

• Applying the new scaling to the modulated KdV to obtain a scaled equation on a dilated circle $\mathbb{T}_\lambda$.
• Using the $I$-method on the scaled equation to derive an almost conservation law for the modified energy.
• Establishing a commutator estimate between the $I$-operator and the nonlinear Young integral driver.
• Iterating the local existence time. The new scaling allows the local existence time to be extended to cover any finite time interval by choosing the scaling parameter $\lambda$ and the smoothing parameter $N$ suitably, overcoming the "loss" of powers of $\lambda$ that would occur with the classical scaling when $s < -3/2$.

Significance
The paper claims to establish a "strong regularization-by-noise phenomenon" for the global well-posedness of dispersive PDEs. While previous results showed that irregular modulation could regularize the equation locally or globally only up to the scaling critical regularity, this work demonstrates that the modulation can effectively lower the well-posedness threshold arbitrarily far below $s = -3/2$, and indeed to any $s \in \mathbb{R}$, provided the modulation is sufficiently irregular.

This result significantly improves upon the previous global well-posedness theorem (Theorem A in [16]), which was restricted to $s > -3/2$. The authors emphasize that this improvement is specific to the periodic setting; they note that on the real line, the "high $\times$ high into low" frequency interactions prevent the solution map from being $C^2$ for $s < -3/2$, restricting local well-posedness to the scaling subcritical regime even with modulation.

The work relies on the interplay between the pathwise analysis of irregular drivers (using the sewing lemma and nonlinear Young integration) and the harmonic analysis techniques of the $I$-method, unified through a novel scaling argument that leverages the structure of the modulation term.