🌀 nonlinear sciences

Integrable perturbation theory for dark solitons of the defocusing nonlinear Schrödinger equation

This paper revisits and refines integrable perturbation theory for dark solitons on a nonzero background in the defocusing nonlinear Schrödinger equation by establishing a complete eigenfunction expansion that correctly predicts the slow-time evolution of soliton parameters, the formation of radiation shelves, and phase dynamics, all of which are validated through numerical simulations.

Original authors: Nicholas J. Ossi, Barbara Prinari, Jianke Yang

Published 2026-03-03

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Nicholas J. Ossi, Barbara Prinari, Jianke Yang

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, dark lake. Suddenly, a ripple moves across the surface, but instead of a wave cresting up, it's a dip—a localized hole in the water that travels smoothly while the rest of the lake remains flat. In the world of physics, this is called a dark soliton. It's a "hole" in a wave that refuses to spread out or disappear, behaving like a particle.

For decades, scientists have had a perfect mathematical toolkit to describe these holes when the lake is perfectly still and undisturbed. This toolkit is called the Inverse Scattering Transform (IST). It's like having a magic decoder ring that can take a complex wave pattern and instantly tell you exactly where the "holes" are, how fast they are moving, and how deep they are.

The Problem: The Lake Isn't Perfect
In the real world, the lake isn't perfectly still. There is wind, rain, or perhaps a slight current (mathematicians call these "perturbations"). When you add these small disturbances to the dark soliton, two things happen:

The hole slowly changes its speed, depth, and position.
A strange, flat "shelf" of water starts to build up on the sides of the hole, like a ramp leading up to the dip.

Previous attempts to predict exactly how these holes behave in a disturbed lake were flawed. Some scientists tried to use the old "magic decoder ring," but they kept getting the wrong answers. They missed the "ramp" (the shelf) and got the speed of the hole wrong. They essentially tried to listen to a song with a broken speaker and couldn't hear the bass notes.

The Solution: Fixing the Decoder Ring
This paper, by Nicholas Ossi, Barbara Prinari, and Jianke Yang, is about fixing the decoder ring.

Here is the simple breakdown of what they did:

1. The "Broken" Parts of the Map

The old mathematical maps (called "completeness relations") used to describe these waves had a blind spot. They ignored the very edges of the spectrum—the points where the water depth changes abruptly (mathematically known as "branch points").

The Analogy: Imagine trying to map a coastline but ignoring the jagged cliffs at the very end. You might get the general shape of the beach, but you'll miss the dangerous rocks.
The Fix: The authors realized that these "cliffs" (singularities) are actually the source of the "shelf" phenomenon. By carefully accounting for the math at these specific edge points, they could finally explain why the ramp forms.

2. The "Shelf" Phenomenon

When the dark soliton moves through a slightly disturbed medium, it leaves a wake.

The Analogy: Think of a boat moving through water. It leaves a wake behind it. But a dark soliton is unique; it leaves a "shelf" on both sides. It's as if the boat is dragging a flat platform of water with it as it moves.
The Discovery: The authors proved that this shelf isn't a mistake in the math; it's a real physical feature caused by the "cliffs" they fixed in their map. They derived formulas to predict exactly how high this shelf is and how fast it moves.

3. Predicting the Future

With their corrected map, the authors could now predict exactly how the dark soliton would behave over time when disturbed. They looked at four common types of disturbances:

Linear Damping: Like friction slowing the boat down.
Nonlinear Damping: Like the water getting thicker as the boat moves faster.
Dissipation: Energy loss, like heat.
Self-steepening: The wave trying to sharpen its own edge.

For each of these, they calculated:

How the speed of the hole changes.
How the depth of the hole changes.
How the position of the hole shifts.
The height and speed of the shelf growing around it.

4. The "Time Travel" Issue

There was one tricky part. Their predictions were perfect for a while, but if you watched the simulation for a very long time, the predicted position of the hole started to drift away from the actual computer simulation.

The Analogy: It's like predicting a car's position based on its current speed. For the next hour, you are spot on. But if the car's speed is slowly changing every second, your prediction will eventually be off because you didn't account for the change in the change.
The Explanation: The authors realized that to be perfect over very long times, you need to account for a "super-slow" time scale (like watching a glacier move). Their current math handles the "slow" changes, but the "super-slow" changes require a more complex, second-layer calculation. They explained why their method hits this wall, which is a huge step forward in understanding the limits of the theory.

The Bottom Line

This paper is a masterclass in mathematical housekeeping. The authors went back to the foundation of the theory, found a few loose screws (the singularities at the edge of the spectrum), tightened them up, and suddenly, the whole machine worked perfectly again.

They proved that:

The "shelf" is real and predictable.
The old theories were missing this shelf because they ignored the edges.
Their new formulas match computer simulations almost perfectly for a long time.

In short, they took a broken compass, fixed the needle, and showed us exactly how to navigate the dark, rippling waters of the nonlinear world.

1. Problem Statement

The paper addresses the challenge of applying integrable perturbation theory to dark solitons of the defocusing Nonlinear Schrödinger (NLS) equation on a nonzero background.

Context: While perturbation theory for solitons decaying at infinity (bright solitons) is well-established, the non-vanishing background of dark solitons introduces significant mathematical difficulties.
The Gap: Previous attempts to extend eigenfunction-expansion-based perturbation theory to dark solitons (e.g., by Konotop, Vekslerchik, Chen et al., Ao & Yan) suffered from critical flaws. Specifically, they failed to correctly account for the singularities of the scattering data at the branch points ( $z = \pm q_0$ ) of the continuous spectrum.
Consequences of Previous Errors:
- Incorrect discrete eigenmodes were used in orthogonality conditions.
- This led to erroneous predictions for the adiabatic evolution of soliton parameters (velocity, amplitude, phase, and center).
- Crucially, previous eigenfunction-based methods failed to predict the radiation shelf (a moving plateau of radiation) that forms on the sides of the perturbed soliton, a phenomenon confirmed by numerical simulations and direct perturbation theory (e.g., Ablowitz et al.).
Goal: To rigorously prove the completeness of squared eigenfunctions for the defocusing NLS with nonzero background, correct the treatment of branch point singularities, and derive accurate evolution equations for all soliton parameters and the radiation shelf.

2. Methodology

The authors employ a rigorous Inverse Scattering Transform (IST) based framework combined with eigenfunction expansion.

A. Completeness of Squared Eigenfunctions

Variational Approach: The authors derive the relationship between variations in the potential ( $\delta q$ ) and variations in the scattering data ( $\delta \rho$ ) using the Lax pair formalism.
Branch Point Singularities: A key innovation is the explicit handling of the singularities at $z = \pm q_0$ $z = \pm q_{0}$ (where $\gamma(z) = 1 - q_0^2/z^2 = 0$ $γ (z) = 1 - q_{0}^{2} / z^{2} = 0$ ).
- In reflectionless (pure soliton) cases, the scattering coefficients are finite at these points, creating simple poles in the integrand of the closure relation.
- The authors prove that the closure relation must include residue contributions from these branch points.
- They show that the integral over the continuous spectrum must be treated as a Cauchy Principal Value, and the contributions from the poles at $\pm q_0$ must be extracted and added as discrete terms.
General Closure Relation: They derive a general closure relation (Eq. 3.30) that includes:
1. Principal value integrals over the continuous spectrum.
2. Residue terms from discrete eigenvalues (solitons).
3. Residue terms from the branch points ( $\pm q_0$ ), which were previously neglected or mishandled.

B. Linearization and 1-Soliton Closure

The linearization operator $L$ is defined, and its eigenfunctions are identified as the squared eigenfunctions of the IST.
For a single dark soliton, the authors explicitly construct the discrete eigenmodes ( $Z_1, \tilde{Z}_1$ ) and the generalized eigenmodes arising from the branch points.
They define modified discrete eigenfunctions ( $\tilde{Z}_1, \tilde{\Upsilon}_1$ ) that incorporate the branch point contributions. This modification is essential to ensure the orthogonality conditions correctly suppress secular growth.

C. Perturbation Expansion

The perturbed NLS equation is analyzed using a multiple-scale expansion with a slow time variable $T = \epsilon t$ .
The solution is expanded as $q \approx e^{i\sigma_1} [u(\xi) + \epsilon \tilde{q}]$ , where $u$ is the unperturbed soliton and $\tilde{q}$ is the first-order correction.
Secular Growth Suppression: By expanding the source term $W$ (derived from the perturbation) and the correction $\tilde{q}$ in the basis of squared eigenfunctions, the authors enforce orthogonality conditions ( $\langle \Upsilon, W \rangle = 0$ ) to eliminate secular terms that would otherwise cause the perturbation expansion to break down.

3. Key Contributions

Rigorous Proof of Completeness: The paper provides a complete proof of the closure relation for squared eigenfunctions on a nonzero background, explicitly accounting for the singularities at the branch points $z = \pm q_0$ .
Correction of Discrete Eigenmodes: The authors demonstrate that previous works used incorrect discrete eigenmodes because they ignored the branch point residues. They introduce modified generalized eigenfunctions ( $\tilde{Z}_1, \tilde{\Upsilon}_1$ ) that correctly incorporate these residues.
Prediction of the Radiation Shelf:
- The first-order correction integral $\tilde{q}$ contains a pole due to the branch point singularities.
- Analyzing this integral reveals that it generates a moving radiation shelf on both sides of the soliton.
- The authors derive explicit formulas for the height ( $h_\pm$ ) and phase gradient ( $\phi_\xi^\pm$ ) of the shelf, showing they are directly linked to the singularity strength at $\pm q_0$ .
Unified Evolution Equations: The method yields a closed system of differential equations for all four soliton parameters:
- Velocity ( $k_1$ )
- Amplitude/Depth ( $\Lambda_1$ )
- Center ( $x_1$ )
- Phase ( $\sigma_1$ )
- Crucially, a new condition linking the evolution of the phase and the center is derived, which was missing in previous eigenfunction-based approaches.

4. Results and Applications

The theory is applied to four specific perturbations and validated against direct numerical simulations (using the ETDRK4 method):

Linear Damping ($F = -iq$):
- Predicts exponential decay of background amplitude, velocity, and depth.
- Predicts a linear shift in the soliton center.
- Result: Matches numerical simulations and the direct perturbation theory of Ablowitz et al. [2] for parameters up to $O(\epsilon)$ .
Nonlinear Damping ( $F = -i|q|^2q$ ):
- Predicts specific decay rates for parameters.
- Result: Confirms the formation of a raised shelf and accurate parameter evolution.
Dissipation ( $F = iq_{xx}$ ):
- Predicts constant background and amplitude, but a drift in velocity and center.
- Result: The linear prediction for the center matches numerics initially but deviates at long times ( $t \sim 1/\epsilon$ ). The authors explain this is due to the need for an $O(\epsilon^2)$ correction (quadratic in time) which requires a second time scale ( $\epsilon^2 t$ ). This aligns with findings in [2].
Self-Steepening ( $F = -i(|q|^2q)_x$ ):
- Unique Finding: For this perturbation, the inner product $\langle \Upsilon_{\pm 0}, W \rangle = 0$ .
- Result: No radiation shelf is generated. The soliton amplitude and velocity remain constant; only the center shifts linearly. This contradicts earlier eigenfunction-based predictions (e.g., [10]) but is confirmed by numerics.

5. Significance

Resolution of the "Shelf" Conjecture: The paper resolves the conjecture in [2] that squared eigenfunctions are an insufficient basis for dark soliton perturbations. The authors prove that with the correct treatment of branch point singularities, the eigenfunction expansion is sufficient and accurately predicts the shelf.
Correction of Prior Literature: It identifies and corrects specific mathematical errors in previous works (e.g., [3–5, 9, 10, 25, 39]) regarding the orthogonality conditions and the definition of discrete eigenmodes.
Methodological Advancement: The technique of extracting branch point residues and modifying the discrete eigenbasis to account for them provides a robust framework for perturbation theory in integrable systems with nonzero backgrounds.
Limitations and Future Work: The authors note that while their $O(\epsilon)$ predictions for the soliton center and phase are accurate for moderate times, they eventually deviate from numerics at $t \sim 1/\epsilon$ . This is attributed to the need for higher-order corrections ( $O(\epsilon^2)$ ) involving slower time scales, a limitation shared by both integrable and direct perturbation theories.

In summary, this work establishes a mathematically rigorous and physically accurate integrable perturbation theory for dark solitons, successfully unifying the prediction of soliton parameter evolution and the emergence of radiation shelves, while correcting decades of previous inconsistencies in the field.