🔢 mathematics

Instability of two-pulse periodic waves with long wavelength in some Hamiltonian PDEs

This paper establishes the spectral instability of two-pulse periodic waves with long wavelengths in various Hamiltonian PDEs by combining asymptotic analysis of the action integral's Hessian and the convergence of a renormalized periodic Evans function to the infinite-period limit.

Original authors: Thomas Courant

Published 2026-02-20

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Thomas Courant

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 you are watching a calm river. Usually, the water flows smoothly. But sometimes, under the right conditions, you see waves. Some waves are solitary—like a single, perfect hump of water traveling alone (a soliton). Others are periodic—like a train of identical waves rolling one after another.

This paper is about a very specific, weird kind of wave: a two-pulse periodic wave.

The Setup: The "Double-Act" Wave

Imagine a wave that repeats itself over and over. But instead of looking like a single bump, each "period" (each repeating unit) contains two distinct bumps:

A Bright Soliton: A tall, bright hump that stands out against the flat water.
A Dark Soliton: A dip or a "hole" in the water, where the water level drops down and then comes back up.

The authors are studying what happens when you stretch this wave out so that the distance between these repeating units becomes huge. In the limit, the wave looks like a bright hump and a dark dip sitting far apart, connected by a very long, flat stretch of water.

The Big Question: Is it Stable?

In physics, "stable" means: If I poke the wave slightly, will it return to its original shape, or will it fall apart?

The authors ask: If we have this giant, two-pulse wave, is it stable?

Their answer is a resounding, "No. It is always unstable."

How They Proved It: Two Different Tools

The authors used two different "flashlights" to look at the problem, because the wave behaves differently depending on the specific conditions of the water and the forces involved.

Tool 1: The "Energy Balance" Check (The Hessian Matrix)

Think of the wave as a ball sitting in a landscape of hills and valleys.

Stable means the ball is in a deep valley. If you nudge it, it rolls back to the center.
Unstable means the ball is on a hilltop or a saddle point. A tiny nudge sends it rolling away.

The authors calculated the "shape" of this energy landscape for their two-pulse wave. They found that if the two individual bumps (the bright one and the dark one) are both stable on their own, the combination creates a weird, unstable saddle point. It's like trying to balance a ball on a horse's back while the horse is running; the combined motion makes it impossible to stay balanced.

The Analogy: Imagine two tightrope walkers. If they walk alone, they are steady. But if they try to walk on the same rope, holding hands, the slight wobble of one inevitably throws the other off. The paper proves that for these specific waves, the "wobble" always wins.

Tool 2: The "Ghostly Echo" Check (The Evans Function)

What if the first tool didn't give a clear answer? (This happens if one of the bumps is already unstable on its own).

The authors used a second method called the Evans function. Think of this as a sophisticated sonar system.

They imagined the giant wave as two separate "ghosts" (the two solitons) that are far apart.
They asked: "If the ghost on the left is unstable, does its instability infect the whole wave?"
They proved that the "instability signal" from one of the solitons travels across the long gap and destabilizes the entire periodic wave.

The Analogy: Imagine two people standing far apart in a large field. If one person starts screaming (instability), the sound waves travel across the field. Even if the second person is quiet, the whole field is now filled with noise. The paper proves that if either of the two pulses is inherently "screaming" (unstable), the whole wave pattern collapses.

The Conclusion: The "Two-Pulse" Paradox

The most surprising part of the paper is the first case. Usually, in physics, if you take two stable things and put them together, you might expect a stable result.

But here, the authors show that even if both the bright pulse and the dark pulse are perfectly stable on their own, putting them together in a long-period wave makes the whole thing fall apart.

It's like taking two perfectly stable chairs and tying them together with a very long, elastic rope. Even though the chairs are fine, the tension in the rope creates a new kind of instability that makes the whole setup wobble and collapse.

Why Does This Matter?

This isn't just about water waves. These equations describe:

Fluid dynamics: How compressible fluids (like air or gas) move.
Quantum mechanics: The behavior of light in fibers or atoms in Bose-Einstein condensates (via the Nonlinear Schrödinger equation).

The paper tells us that in the real world, if you try to create a pattern of light or fluid that looks like a "bright-dark" pair repeating over a long distance, nature will not let it stay that way. It will inevitably break apart or change shape. This helps scientists understand the limits of what kinds of waves can exist in nature and which ones are just mathematical curiosities that can't survive in the real world.

In short: Nature hates this specific "double-pulse" arrangement. No matter how you set it up, it's destined to be unstable.

1. Problem Statement

The paper investigates the spectral stability of two-pulse periodic traveling waves in a broad class of quasilinear Hamiltonian Partial Differential Equations (PDEs). These equations include:

Quasilinear generalizations of the Korteweg–de Vries (KdV) equation.
Dispersive perturbations of the Euler equations for compressible fluids (in both Eulerian and Lagrangian coordinates).
Hydrodynamic formulations of the Nonlinear Schrödinger (NLS) equations.

The Specific Regime:
The study focuses on the long-wavelength limit (large period). In this regime, the periodic wave profile on each period consists of two distinct pulses:

One pulse converging to a bright soliton (a localized pulse on a zero background).
One pulse converging to a dark soliton (a localized dip on a non-zero background).
Both solitons share the same asymptotic endstate. The central question is whether these "two-pulse" periodic waves are spectrally stable or unstable as the period tends to infinity.

2. Methodology

The author employs a dual-approach strategy to cover all possible stability scenarios, combining asymptotic analysis of action integrals with spectral perturbation theory using Evans functions.

Approach A: Asymptotic Expansion of the Action Integral

Framework: Based on the stability criteria for periodic waves established in previous works (specifically [BMR16] and [BMR20]).
Mechanism: The stability of periodic waves is linked to the signature of the Hessian matrix of the abbreviated action integral $\Theta(c, \lambda, \mu)$ .
Calculation: The author performs a detailed asymptotic expansion of the Hessian $\nabla^2 \Theta$ as the period $X \to \infty$ (equivalently, as the energy parameter $\epsilon = \sqrt{\mu - \mu_s} \to 0$ ).
Logic: By analyzing the negative signature $n(\nabla^2 \Theta)$ , the author applies Theorem 2.3.1 (from [BMR16]) to determine instability. Specifically, if the number of negative eigenvalues of the Hessian satisfies certain parity conditions relative to the dimension of the system, the wave is spectrally unstable.

Approach B: Convergence of the Periodic Evans Function

Motivation: The action integral approach (Approach A) only detects instability when the sum of the second derivatives of the Boussinesq momenta of the two solitons is positive. It cannot detect instability if both solitons are individually stable but the sum is negative (or zero).
Mechanism: The author adapts the strategy from [YZ19] (originally for single-pulse limits) to the two-pulse case.
Process:
1. Define the periodic Evans function $E^\epsilon(z, \gamma)$ for the two-pulse wave.
2. Prove that the periodic wave profile converges exponentially fast to the superposition of the two limiting solitons (one bright, one dark) as the period increases.
3. Show that a suitably renormalized (rescaled) periodic Evans function converges exponentially fast to the product of the Evans functions of the two individual solitons.
Logic: If either of the limiting solitons is spectrally unstable (i.e., its Evans function has a zero in the right half-plane), the product will have a zero, implying the periodic wave is also unstable.

3. Key Contributions

Generalization of Stability Criteria: The paper extends the stability analysis from single-soliton limits (studied in [BMR20]) to two-pulse limits involving distinct soliton types (bright and dark) on the same period.
Unified Instability Proof: It establishes that all two-pulse periodic waves with sufficiently large periods are spectrally unstable, regardless of the stability of the individual limiting solitons.
Technical Innovation in Evans Functions: The paper successfully generalizes the Evans function convergence results from [YZ19] to a scenario where the periodic wave converges to two different homoclinic orbits simultaneously, rather than just one.
Resolution of the "Stable Soliton" Case: It resolves the case where both limiting solitons are spectrally stable. Previous methods (modulational instability or action Hessian) might fail here; the paper proves that even in this case, the synchronization of the two pulses breaks down, leading to instability.

4. Main Results

The main theorem (Theorem 2.3.3) states that for a two-pulse periodic wave of sufficiently large period:

Case 1: Sum of Momentum Derivatives is Positive.
If $d_c^2 M_\ell + d_c^2 M_r > 0$ (where $M$ is the Boussinesq momentum and $d_c$ is the derivative with respect to speed), the wave is spectrally unstable. This is proven via the asymptotic expansion of the action Hessian (Theorem 3.4.1).
Case 2: At Least One Soliton is Unstable.
If either the left-hand (dark) soliton or the right-hand (bright) soliton is spectrally unstable, the two-pulse periodic wave is spectrally unstable. This is proven via the convergence of the Evans function (Corollary 4.4.2).
Case 3: Both Solitons Stable but Sum Negative.
Even if both solitons are stable and $d_c^2 M_\ell + d_c^2 M_r < 0$ (a case where the action Hessian is negative-definite and modulational instability is absent), the wave is spectrally unstable. This is a non-trivial result derived from the Evans function analysis, showing that the interaction between the two distant pulses creates an instability.

Conclusion: Two-pulse periodic waves in this class of Hamiltonian PDEs are always spectrally unstable in the long-wavelength limit, except for the degenerate case where the solitons are stable and the sum of momentum derivatives is exactly zero.

5. Significance

Physical Relevance: The results apply to fundamental models in fluid dynamics (Euler-Korteweg systems) and nonlinear optics (NLS equations). The finding that "two-pulse" structures are inherently unstable suggests that such configurations are unlikely to persist as stable coherent structures in physical systems, even if the constituent solitons are stable.
Theoretical Advancement: The paper bridges a gap in the theory of Hamiltonian PDEs. While the stability of solitary waves and single-pulse periodic waves is well-understood, the stability of multi-pulse periodic waves in the long-wavelength limit was an open problem. The work demonstrates that the "superposition" of stable solitons does not guarantee the stability of the resulting periodic wave.
Methodological Rigor: The combination of variational methods (action integrals) and spectral methods (Evans functions) provides a robust framework for analyzing complex asymptotic regimes in quasilinear dispersive PDEs, where standard linearization techniques often fail due to the lack of global existence results in low-regularity spaces.

In summary, the paper proves that the synchronization required to maintain a two-pulse periodic wave is fragile; any perturbation that decouples the velocities or phases of the two constituent solitons leads to exponential growth, rendering the wave spectrally unstable.