Shielding of breathers for the focusing nonlinear Schrödinger equation

We study a deterministic gas of breathers for the Focusing Nonlinear Schrödinger equation. The gas of breathers is obtained from a $N$-breather solution in the limit $N\to \infty$.\ The limit is performed at the level of scattering data by letting the $N$-breather spectrum to fill uniformly a suitable compact domain of the complex plane in the limit $N\to\infty$. The corresponding norming constants are interpolated by a smooth function and scaled as $1/N$. For particular choices of the domain and the interpolating function, the gas of breathers behaves as finite breathers solution. This extends the shielding effect discovered in “M. Bertola, T. Grava, and G. Orsatti - Physical Review Letters, 130.12 (2023): 1” for a soliton gas also to a breather gas.

💡 Research Summary

The paper introduces a deterministic “breather gas” for the focusing nonlinear Schrödinger (FNLS) equation, extending the concept of a soliton gas and the recently discovered soliton shielding phenomenon to breather solutions. Starting from the well‑known N‑breather construction via the inverse scattering transform (IST) for FNLS with non‑zero boundary conditions, the authors recall that each breather is encoded by a set of four poles in the complex spectral plane (one free pole ζ in the upper half‑plane outside the unit disk, together with three symmetry‑related poles) and associated norming constants C. The scattering data are packaged into a 2×2 matrix M(z;x,t) which satisfies a Riemann–Hilbert (RH) problem (residue conditions at the poles, asymptotics at infinity and at the origin). The physical field ψ(x,t) is recovered from the (1,2) entry of M.

The central contribution is the passage to the limit N→∞ while the discrete spectrum becomes a continuous distribution. The authors let the N free poles ζ₁,…,ζ_N fill uniformly a compact domain D₁⊂D⁺₁ (the upper‑half‑plane outside the unit circle). The norming constants are interpolated by a smooth function ρ(z) and scaled as C_j = ρ(ζ_j)/N. In this scaling the residues, originally O(1), become O(1/N) and the sum of residues turns into an integral over D₁. By removing the poles from M and introducing jump contours γ₊₁,γ₊₂,γ₋₁,γ₋₂ that encircle D₁, D₂ (its symmetry) and their complex conjugates, the RH problem is reformulated for a new matrix Y_N(z;x,t) with a jump matrix J_N(z;x,t). J_N contains exponential phase factors e^{±2iθ(z;x,t)} (θ is the usual linear phase) multiplied by a product over (z−ζ_j)⁻¹ weighted by ρ(z). As N→∞ the product converges to an exponential of a Cauchy integral, and the jump matrix converges pointwise to a continuous J(z;x,t).

The authors prove the existence and uniqueness of the limiting RH problem using standard L²‑theory for Cauchy operators, assuming ρ is smooth and bounded on the closure of D₁. They then reconstruct ψ(x,t) from the limiting solution, showing that for particular choices of D₁ and ρ the resulting field coincides exactly with a finite‑N breather solution. Two explicit examples are presented:

1. D₁ is a disk and ρ is constant. The breather gas reduces to a single Kuznetsov–Ma breather; the infinite collection of breathers “shields” each other, leaving only the profile of one breather.

2. D₁ is an “8‑shaped” domain (two touching disks) with an appropriate ρ. The gas reproduces a two‑breather solution, demonstrating a higher‑order shielding effect.

These examples mirror the soliton shielding observed in Bertola, Grava, and Orsatti (2023) and confirm that the phenomenon is not limited to solitons but also applies to oscillatory breather structures. The paper concludes by discussing the physical relevance of breather gases as statistical ensembles for large‑scale nonlinear wave phenomena, such as rogue waves in optics or water, and suggests that the methodology can be adapted to other integrable equations (e.g., mKdV, KdV) with non‑zero boundary conditions.