🔢 mathematics

Nontrivial integrable weak stationary solutions to active scalar equations with non-odd drift

This paper constructs nontrivial weak stationary solutions to active scalar equations with non-odd drift in low regularity spaces by employing a convex integration scheme that utilizes highly oscillatory corrections with a variable degree of intermittency decreasing to zero across iteration stages.

Original authors: Nicholas Gismondi

Published 2026-01-22

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Nicholas Gismondi

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Finding a "Ghost" in the Machine

Imagine you are trying to balance a very complicated, wobbly stack of Jenga blocks. This stack represents a physical system (like a fluid or a gas) described by a specific set of rules called an Active Scalar Equation.

Usually, when scientists try to solve these equations, they look for "smooth" solutions—solutions that are neat, predictable, and don't have any jagged edges or sudden jumps. However, this paper asks a different question: Can we find a solution that is "rough" or "jagged" but still mathematically valid?

The author, Nicholas Gismondi, says "Yes." He constructs a solution that is technically a "weak solution." Think of this not as a solid block of wood, but as a cloud of dust that still obeys the laws of physics, even though it's messy and hard to pin down.

The Main Obstacle: The "Odd" vs. "Even" Problem

To understand the trick the author uses, imagine you are trying to cancel out a loud noise (an error) in a room.

The Old Way (Odd Multipliers): In many previous attempts to solve these equations, the "noise-canceling" tool the scientists used was like a mirror. If you pushed the mirror left, it pushed back right. In math terms, this is called being "odd." If the tool is a perfect mirror, any noise you try to cancel out with the average of the room just disappears into the void. You can't use the average to fix the problem because the mirror forces the average to be zero.
The New Way (Non-Odd Multipliers): This paper focuses on a specific type of system where the "mirror" is broken or bent. It's not a perfect mirror; it's "non-odd." Because the mirror is bent, the average of the room doesn't have to be zero. This gives the author a new tool: he can use the average of the system to cancel out the biggest, loudest errors.

The Analogy: Imagine trying to balance a seesaw. If the pivot point is perfectly in the middle (odd), you can't use the weight of the empty air to balance it. But if the pivot is slightly off-center (non-odd), the weight of the air itself helps you balance the seesaw. This paper uses that off-center pivot to build a solution where others couldn't.

The Construction: The "Intermittent Slab"

How does the author actually build this messy solution? He uses a method called Convex Integration.

Think of this like building a sculpture out of clay, but you have to do it in layers, and each layer has to be incredibly wiggly and fast-moving.

The Iteration: He starts with a rough guess. Then, he adds a tiny, super-fast vibration to fix the errors. Then he adds another even faster vibration to fix the new errors. He repeats this thousands of times.
The "Intermittent Slab": Usually, when you add these vibrations, they are spread out evenly, like a fine mist. But this author uses something called an "intermittent slab."
- The Metaphor: Imagine a sprinkler. A normal sprinkler sprays water everywhere evenly. An "intermittent slab" is like a sprinkler that only sprays water in very thin, sharp sheets, leaving huge gaps of dry air in between.
- Why it matters: By making the vibrations happen in these thin, sharp sheets rather than everywhere, the author can control the "roughness" of the final solution. He can make the solution "spiky" in a way that allows it to exist in a mathematical space that is usually considered too messy for solutions to live in.

The Result: A Solution That Exists (But is Rough)

The paper proves that for a wide class of these equations, you can construct a solution that:

Is non-trivial (it's not just zero; it actually does something).
Is stationary (it doesn't change over time; it's a frozen snapshot of a complex dance).
Is integrable (you can calculate its total "amount," even though it's messy).
Lives in a specific mathematical "neighborhood" called Besov spaces.

The "Neighborhood" Analogy:
Imagine a city.

Smooth solutions live in the fancy downtown area with paved roads and streetlights (high regularity).
This new solution lives in a rougher, more chaotic part of town. It's not a slum, but it's not downtown either. It's a place where the streets are bumpy and the buildings are a bit crooked. The author proves that a valid "house" (a solution) can exist in this rough neighborhood, provided the neighborhood isn't too chaotic.

What This Paper Does NOT Do

It is important to stick to what the paper actually claims:

It does not say this solution describes a real-world fluid you can see in a lab right now. It is a mathematical existence proof.
It does not solve the famous "Onsager Conjecture" for all cases (it specifically avoids the "odd" cases like the Surface Quasi-Geostrophic equation, which is a different, harder problem).
It does not provide a new way to predict weather or design engines.

Summary

Nicholas Gismondi has shown that if you have a specific type of mathematical rule (one that isn't perfectly symmetrical), you can build a "rough" solution that stays still in time. He does this by using a clever trick to cancel out errors using the system's average, and by building the solution out of "intermittent slabs"—thin, sharp layers of vibration—rather than smooth waves. This proves that these messy, jagged solutions are mathematically possible, expanding the map of where solutions to these equations can exist.

Technical Summary: Nontrivial Integrable Weak Stationary Solutions to Active Scalar Equations with Non-Odd Drift

Problem Statement
This paper addresses the existence of nontrivial weak solutions to the stationary active scalar equation on the torus $\mathbb{T}^d$ ( $d \geq 2$ ):
$\begin{cases} \nabla \cdot (\theta u) + \Lambda^\gamma \theta = 0 \\ u = T\theta \end{cases}$
where $\theta$ is a scalar field, $u$ is a divergence-free velocity field, $\Lambda^\gamma = (-\Delta)^{\gamma/2}$ is the fractional Laplacian with $0 < \gamma \leq 2$ , and $T$ is a convolution-type Calderón-Zygmund operator. The central structural constraint of this work is that the Fourier multiplier $m(\xi)$ of the operator $T$ is not odd.

The authors focus on constructing solutions that are integrable functions (specifically lying in $L^{2-\epsilon}(\mathbb{T}^d)$ and $\dot{B}^{-\epsilon}_{\infty,\infty}(\mathbb{T}^d)$ for $0 < \epsilon < 1$ ) rather than mere distributions. A significant challenge arises because for multipliers that are odd, the operator $T$ is skew-adjoint on $L^2$ , forcing the mean of the nonlinear term $\int_{\mathbb{T}^d} w T w$ to vanish for smooth perturbations $w$ . This vanishing prevents the cancellation of leading-order errors in standard convex integration schemes. The paper restricts its scope to non-odd multipliers to bypass this obstruction, noting that general skew-adjoint operators remain out of reach for current methods.

Methodology
The construction employs the method of convex integration, originally introduced by De Lellis and Székelyhidi. The approach involves an iterative procedure where a sequence of smooth approximate solutions $(\theta_q, u_q, R_q)$ is constructed, satisfying a relaxed equation with a stress field error $R_q$ . The goal is to drive $\|R_q\|_{\dot{H}^{-s}} \to 0$ as $q \to \infty$ .

Key technical components include:

Non-Odd Multiplier Exploitation: The non-odd nature of $m(\xi)$ allows the mean of the nonlinear term to be non-zero. Specifically, the construction utilizes the fact that the image of $m(\xi) + m(-\xi)$ contains a basis for $\mathbb{R}^d$ (established via Lemmas 2.9 and 2.10). This enables the cancellation of the leading-order stress error using the mean of the perturbation.
Intermittent Building Blocks: The perturbations $w_{q+1}$ are constructed using "intermittent slabs" (Lemma 2.12). These are highly oscillatory functions with a variable degree of intermittency that decreases to zero at higher stages of the iteration. This specific structure allows the authors to control the $L^p$ norms of the solution while maintaining the necessary regularity.
Negative Regularity Estimates: Unlike previous convex integration schemes for fluid equations that often estimate errors in $L^p$ or $C^0$ norms, this work estimates the stress field in negative Sobolev norms ( $\dot{H}^{-s}$ ). This choice is crucial as it avoids requiring derivatives to act directly on the high-frequency building blocks, thereby relaxing the requirement that the building blocks be stationary solutions to any PDE.
Weak Solution Definition: Since the constructed solutions lie in spaces with negative regularity where the product $\theta u$ is not classically defined, the authors adopt a definition of weak solutions based on paraproducts. They define the product via the absolute convergence of the paraproduct expansion in $\dot{H}^{-s}$ , following the framework of [1] and [21].

Key Contributions and Results
The main result (Theorem 1.3) establishes that for any $0 < \gamma \leq 2$ , there exist nontrivial weak solutions $(\theta, u)$ to the stationary active scalar equation such that:

$\theta, u \in \bigcap_{0<\epsilon<1} (\dot{B}^{-\epsilon}_{\infty,\infty}(\mathbb{T}^d) \cap L^{2-\epsilon}(\mathbb{T}^d))$ .
The product $\theta u$ is well-defined as a paraproduct in $\dot{H}^{-s}(\mathbb{T}^d)$ for sufficiently large $s$ .
The pair satisfies the equation in the sense of the weak paraproduct definition.

The paper provides a detailed inductive proof (Proposition 4.1) demonstrating that the oscillation, Nash, and dissipation errors can be made arbitrarily small in the $\dot{H}^{-s}$ norm by choosing sufficiently large frequency parameters $\lambda_q$ .

Significance and Scope
The authors claim this work provides the first construction of weak solutions to the stationary active scalar equation on the torus that are integrable functions rather than just distributions. This is highlighted in Remark 1.4.

The paper is modest regarding its limitations:

The solutions do not lie in $L^2(\mathbb{T}^d)$ , which remains a difficult open problem.
The results are restricted to multipliers that are not odd. The authors explicitly state that their result does not apply to the Surface Quasi-Geostrophic (SQG) equation, as its multiplier is odd.
The method relies on the multiplier being a Calderón-Zygmund operator. The paper notes that the Magneto-Geostrophic (MG) equation involves an unbounded multiplier, placing it outside the scope of the current method, though it suggests this as an interesting open question.
The construction utilizes a specific class of non-odd multipliers; the authors do not claim to resolve the existence of solutions for general skew-adjoint operators.

In summary, the paper extends the convex integration technique to a specific class of stationary active scalar equations by leveraging the non-odd symmetry of the drift operator to cancel leading-order errors, utilizing intermittent building blocks to achieve integrability, and employing negative Sobolev norms to define the weak solution.