Soliton Turbulence in Shallow Water Ocean Surface Waves

TL;DR

This study provides experimental evidence of soliton turbulence in shallow ocean waves, revealing that low frequency energy is dominated by dense, non-Gaussian soliton gas behavior, confirmed through advanced spectral and theoretical analysis.

Contribution

First experimental confirmation of soliton turbulence in ocean waves using finite gap theory and spectral analysis, linking observed wave behavior to integrable soliton systems.

Findings

Low frequency spectra follow a ~ω^{-1} power law.

Solitons are dense, non-Gaussian, and have random phases.

Data analysis confirms soliton dominance in wave trains.

Abstract

We analyze shallow water wind waves in Currituck Sound, North Carolina and experimentally confirm, for the first time, the presence of $soliton$ $turbulence$ in ocean waves. Soliton turbulence is an exotic form of nonlinear wave motion where low frequency energy may also be viewed as a $dense$ $soliton$ $gas$, described theoretically by the soliton limit of the Korteweg-deVries (KdV) equation, a $completely$ $integrable$ $soliton$ $system$: Hence the phrase "soliton turbulence" is synonymous with "integrable soliton turbulence." For periodic/quasiperiodic boundary conditions the $ergodic$ $solutions$ of KdV are exactly solvable by $finite$ $gap$ $theory$ (FGT), the basis of our data analysis. We find that large amplitude measured wave trains near the energetic peak of a storm have low frequency power spectra that behave as $\sim\omega^{-1}$. We use the linear Fourier transform to estimate this power law from the power spectrum and to filter $densely$ $packed$ $soliton$ $wave$ $trains$ from the data. We apply FGT to determine the $soliton$ $spectrum$ and find that the low frequency $\sim\omega^{-1}$ region is $soliton$ $dominated$. The solitons have $random$ $FGT$ $phases$, a $soliton$ $random$ $phase$ $approximation$, which supports our interpretation of the data as soliton turbulence. From the $probability$ $density$ $of$ $the$ $solitons$ we are able to demonstrate that the solitons are $dense$ $in$ $time$ and $highly$ $non$ $Gaussian$.