🔢 mathematics

Numerical study of the 2D Kaup-Broer-Kuperschmidt Boussinesq system

This paper presents a numerical study of the two-dimensional Kaup-Broer-Kuperschmidt Boussinesq system, demonstrating that its soliton-type solutions, line solitons, and general localized initial data are all unstable and prone to either dispersion or singularity formation.

Original authors: Théo Gaudry, Christian Klein, Jean-Claude Saut, Nikola Stoilov

Published 2026-05-07

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Théo Gaudry, Christian Klein, Jean-Claude Saut, Nikola Stoilov

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

Imagine a vast, calm ocean where waves usually travel smoothly, spreading out and fading away. Now, imagine a special kind of wave equation—a set of mathematical rules—that describes how water moves. In one dimension (a single line), these rules are well-behaved and predictable; they allow for "solitons," which are like perfect, self-sustaining surfer waves that can travel forever without losing their shape.

This paper investigates what happens when we take those same rules and expand them into two dimensions (a full surface, like a real pond). The researchers, acting as digital oceanographers, built a computer simulation to see if these perfect waves could survive in a 2D world.

Here is what they found, explained through simple analogies:

1. The "Perfect" Stationary Wave (The Static Solution)

First, the team tried to build a "stationary" wave—a wave that sits perfectly still in one spot, like a lighthouse beam frozen in time.

The Experiment: They used a computer to construct this perfect, round, stationary wave.
The Result: It exists, but it is incredibly fragile. Think of it like a house of cards built on a table.
- If you make the wave slightly weaker (less energy), it doesn't hold together; it simply dissolves and spreads out into the ocean, vanishing into nothingness (dispersion).
- If you make the wave slightly stronger (more energy), it doesn't just grow; it collapses in on itself violently and instantly. In math terms, this is called a "blow-up," where the wave becomes infinitely tall and infinitely narrow in a split second.

2. The "Line" Wave (The Long Strip)

Next, they looked at "line solitons." Imagine a wave that is a straight, infinite line stretching across the ocean, like a long ribbon. In a 1D world, this is stable. But in 2D, the ribbon can wiggle sideways.

The Experiment: They took this long ribbon and gave it a tiny nudge (a perturbation) to see if it would straighten out or break.
The Result: The ribbon is strongly unstable. No matter how small the nudge, the line eventually breaks apart. Instead of staying a straight line, it pinches off into two distinct, sharp peaks. These peaks then rush toward each other and collapse into a singularity (a point of infinite intensity). It's like trying to balance a long, straight stick on your finger; the slightest wobble causes it to snap and fall.

3. Random Waves (Gaussian Data)

Finally, they didn't start with perfect shapes at all. They threw random, bell-shaped "splashes" of water into the system to see if nature would naturally form a stable structure.

The Experiment: They simulated various random splashes.
The Result: Nothing stable formed.
- Some splashes just spread out and faded away.
- Others grew for a moment, only to eventually collapse and blow up.
- Crucially: They found no stable structures in 2D. There is no "magic shape" that can sit in the middle of the ocean and stay there forever.

The Big Picture: The "Supercritical" Collapse

The paper draws a parallel to a famous type of equation in physics called the "Nonlinear Schrödinger equation."

In some versions of this equation, waves can collapse slowly.
In this specific 2D KBK system, the collapse is "supercritical." This means the wave doesn't just get big; it gets big faster and faster as it approaches the end, similar to a black hole forming. The math suggests the wave shrinks in size while its height shoots up to infinity in a predictable, self-similar pattern (like a fractal zooming in).

Summary

In the world of this specific 2D water wave equation:

Perfect waves exist but are impossible to keep stable.
Too little energy makes them fade away.
Too much energy makes them explode into a singularity.
Line waves (ribbons) are just as unstable as the round ones.
Nature offers no safe harbor: There are no stable, localized islands of water that can survive in this 2D system. Everything either fades into the background or collapses into a point.

The authors conclude that while the 1D version of this system is a well-behaved, integrable system (like a perfect clock), the 2D version is chaotic and prone to violent collapse, with no stable structures to be found.

Technical Summary: Numerical Study of the 2D Kaup-Broer-Kuperschmidt Boussinesq System

Problem Statement
This work investigates the well-posed version of the two-dimensional (2D) Kaup-Broer-Kuperschmidt (KBK) system, a specific case of the Boussinesq 'abcd' systems. While the 1D KBK system is completely integrable and possesses known solitary wave solutions with established orbital stability, the qualitative properties of its 2D generalization remain largely unexplored. The 2D system is not integrable, and rigorous results regarding long-time behavior, global existence, or finite-time blow-up are absent. The primary objective is to numerically determine whether stable, localized structures (such as 2D solitons or line solitons) exist in the 2D KBK system and to characterize the evolution of general localized initial data.

Methodology
The authors employ a high-precision numerical approach combining spectral methods in space with exponential time differencing (ETD) schemes in time.

Formulation: The system is reformulated using a velocity potential $V$ (where $v = \nabla V$ ) and transformed into a system of fractional nonlinear Schrödinger (NLS) equations in Fourier space. This formulation facilitates the use of exponential integrators, specifically the fourth-order ETD scheme by Cox and Matthews, which is necessary to handle the stiffness of the system arising from high-order dispersion.
Static Solutions: To construct localized static solutions, the authors first solve the associated radial ordinary differential equation (ODE) in polar coordinates using a Chebyshev spectral discretization and a Newton-Krylov method with relaxation. This 1D radial solution is then interpolated onto a 2D Fourier grid to serve as an initial guess for solving the full 2D nonlinear system.
Time Evolution: The time evolution is simulated on a doubly periodic domain (2-torus) to approximate rapidly decaying functions. The authors monitor the relative energy conservation ( $\Delta E$ ) to ensure numerical accuracy, aiming for errors significantly below $10^{-3}$ .
Blow-up Analysis: To investigate potential singularities, the authors utilize a dynamical rescaling technique. This involves rescaling spatial variables and time to track the behavior of solutions near a potential blow-up time $t^*$ . The scaling factor $L(t)$ and the exponents of the $L^\infty$ and $L^2$ norms are fitted to power laws of the form $(t^* - t)^\alpha$ to determine the nature of the blow-up (algebraic vs. exponential).

Key Contributions and Results

Existence of a Static Solution: The authors successfully construct a radially symmetric, localized static solution to the 2D KBK system. This solution corresponds to a ground state of a vector NLS equation.
Instability of Static Solutions: Numerical perturbations of this static solution reveal a dichotomy in behavior:
- Dispersion: Perturbations with a mass (or $L^2$ norm) slightly smaller than the static solution lead to the dispersion of the wave to infinity.
- Finite-Time Blow-up: Perturbations with a slightly larger mass result in a rapid self-focusing process leading to finite-time blow-up. The numerical data suggests the blow-up is self-similar, with the scaling factor $L$ behaving as $L \propto (t^* - t)^{1/2}$ , and the norms growing as $\|\eta\| \sim (t^* - t)^\alpha$ with $\alpha \approx -1$ . This behavior is analogous to the focusing regime in $L^2$ -supercritical NLS equations.
Transverse Instability of Line Solitons: The study examines 1D line solitons (solutions independent of the transverse coordinate $y$ ) extended trivially to 2D. These structures are found to be strongly unstable against transverse perturbations. Both positive and negative perturbations lead to finite-time blow-up, characterized by the formation of sharp peaks and self-similar scaling similar to the 2D static solution case. Notably, blow-up occurs even when the perturbed mass is smaller than that of the constructed static solution, a behavior distinct from the $L^2$ -critical NLS equation.
General Localized Data: Simulations with general Gaussian initial data (where no specific static solution is used as a base) show that the system does not support stable localized structures. Depending on the parameters, the data either disperses or undergoes blow-up.

Significance and Claims
The paper concludes that the 2D KBK system does not support stable localized structures. The authors assert that:

Solutions for localized smooth initial data can blow up in finite time via a self-similar mechanism, mirroring the behavior of supercritical NLS equations.
Line solitary waves are strongly unstable and prone to blow-up under transverse perturbations.
There are no stable structures localized in two dimensions for this system.

The authors explicitly state that the precise criteria for data to blow up and the exact profiles of the blow-up solutions ( $U^\infty$ and $V^\infty$ ) remain unknown. The work serves as a numerical characterization of the system's dynamics, highlighting the absence of stability in 2D despite the integrability and stability properties of its 1D counterpart.

1. The "Perfect" Stationary Wave (The Static Solution)

2. The "Line" Wave (The Long Strip)

3. Random Waves (Gaussian Data)

The Big Picture: The "Supercritical" Collapse

Summary

More like this