Coupled stochastic variational principles for multiscale surface gravity waves -- Part I: theoretical framework
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Mathematical Physics. The day in mathematical physics: rigorous models of physical systems.
Kai: I'm Kai.
Mira: And I'm Mira, here with Lev.
Kai: We're looking at the paper: "Coupled stochastic variational principles for multiscale surface gravity waves -- Part I".
Mira: Here's the gist of the paper. This paper develops a comprehensive stochastic variational framework for surface gravity waves that bridges deterministic phase-resolved and stochastic phase-averaged descriptions The resulting coupled system, combining large-scale dynamics with correlation-mode evolution, provides a systematic framework for deriving stochastic models of ocean waves Theoretical Foundation and Hamiltonian Structure The framework starts from Luke’s variational principle for irrotational, incompressible free-surface flow This principle is extended by introducing a decomposition of the velocity potential into a large-scale deterministic component and a regularized stochastic noise term representing unresolved scales A pathwise variational principle yields stochastic counterparts of the Laplace, Bernoulli, and kinematic boundary conditions The resulting coupled system preserves the Hamiltonian structure of the Zakharov–Craig–Sulem formulation This preservation ensures that important invariants such as total energy are respected in a statistical sense.
Kai: Stochastic Modeling and Noise Regularization The velocity potential is assumed to decompose into a large-scale component and a fluctuating stochastic component The noise term is defined through a temporal mollification using a family of smooth functions hτ with compact support and unit integral This regularized noise process Wτt has variance E(Wτt)2= τ−1, so that Wτt = O(τ−1/2).
Mira: The system is then formulated using a second variational principle in expectation to provide evolution equations for the noise correlation functions This approach reconciles phase-averaged and phaseresolved descriptions by representing unresolved scales through a systematic stochastic closure.
Kai: Asymptotic Approximations and Physical Regimes The paper derives simplified models through asymptotic assumptions applied to the nondimensional equations These approximations are restricted to the shallow-water regime, characterized by the small shallowness parameter µ = H2/L2 ≪ 1 In the long-wave, small-amplitude limit, the theory reduces to a stochastic Airy-wave model in which unresolved fluctuations act as a superposition of random linear waves on a deterministic carrier For coastal regimes with weaker scale separation, stochastic Serre–Green–Naghdi and Boussinesq-type systems were obtained.
Paper summary: Kai: So, wrapping up this look at "Coupled stochastic variational principles for multiscale surface gravity waves -- Part I: theoretical framework", what's the big picture here? The authors are building a framework that uses coupled stochastic variational principles to handle the complexity of multiscale surface gravity waves.
Mira: They start with Luke’s principle and extend it by decomposing the velocity potential into a large-scale deterministic part and a regularized noise term for unresolved scales, then they use two coupled variational principles to derive evolution equations for those noise correlations.
Lev: The key structural thing they manage is preserving the Hamiltonian structure of the Zakharov–Craig–Sulem formulation throughout this process. That means total energy is respected statistically, which is a big deal when you're trying to model complex systems like waves.
Kai: And they show that by using asymptotic approximations—like Airy or Boussinesq—they can get concrete solutions for different physical regimes, showing how the unresolved scales behave differently depending on the conditions.
Mira: The implication is that we have a unified framework where unresolved dynamics become explicit dynamical variables rather than just prescribed stochastic forcings in this paper "Coupled stochastic variational principles for multiscale surface gravity waves -- Part I: theoretical framework".
Lev: And they conclude by showing that the noise correlation modes evolve as mean-reverting stochastic modes with a finite correlation time tau, which is what you need for practical modeling.
Kai: It establishes a foundation for these new stochastic water-wave models capable of consistently representing transport and uncertainty across scales. It’s about building this structure into the model itself rather than just adding noise on top.
Mira: And that geometric optics interpretation of the noise modes, where they are active geometrical objects, provides a direct connection with ray-tracing dynamics, which is useful for spectral numerical methods.
Lev: So, if we consider what it means for the world in terms of modeling ocean waves and climate processes, it suggests that stochastic transport by unresolved scales might contribute significantly alongside classical four-wave interaction mechanisms in realistic ocean conditions.
Kai: It’s a framework that takes uncertainty and wave interactions across scales and tries to represent them consistently through variational structure. That’s what they did in this paper "Coupled stochastic variational principles for multiscale surface gravity waves -- Part I: theoretical framework".
Conclusion: Kai: So we've been looking at this paper, "Coupled stochastic variational principles for multiscale surface gravity waves -- Part I: theoretical framework". Basically, they’re trying to build a system that handles both the big wave dynamics and the tiny noise scales in ocean waves using these coupled principles.
Mira: Right. What I found interesting is how they connect deterministic phase-resolved stuff with stochastic phase-averaged descriptions through this variational structure. It’s an attempt to bridge that gap systematically.
Lev: From my side, it's about how they manage the noise term, which they regularize using a temporal mollification—a smooth function with compact support and unit integral—to get variance of one over tau minus one over two. That means the noise isn't just white; it has a finite memory.
Kai: Exactly. And they show that this whole setup preserves the Hamiltonian structure of the Zakharov–Craig–Sulem formulation, which is crucial because it keeps things like total energy conserved in a statistical sense, even with all this noise involved.
Mira: That preservation of the structure is key because it means you don't just tack on random noise; you’re deriving the noise evolution from the underlying physics itself through these expectation equations. It’s a very consistent way to do it.
Lev: For us, that consistency is important because it suggests that we can move away from ad hoc noise models and instead have the unresolved scales evolving as something governed by these derived equations.
Kai: And they get pretty specific about the physical regimes, showing how these equations simplify down into things like Airy wave models in shallow water or Boussinesq-type systems in coastal regimes. It’s not just one simple equation for everything.
Mira: That's because the assumptions they make—like the small shallowness parameter mu being much less than one—restrict the applicability, but it gives us concrete results for those specific physical scenarios.
Lev: If we were trying to run this on real hardware, that means these simplified models are a good starting point before you try to incorporate all the full nonlinearities of a deep ocean scenario.
Kai: So what this paper really suggests is that by making the unresolved dynamics an explicit dynamical part of the system, rather than just an added forcing term, we get a new class of water-wave models that can handle uncertainty and transport across many scales consistently.
Mira: It’s about treating those unresolved scales as active geometrical objects with their own evolution dictated by these Hamiltonian equations. That's a pretty strong conceptual shift for how we visualize wave interactions.
Lev: And the next thing is looking at how this connects to things like ray-tracing dynamics, which is important because that gives us a way to use spectral methods to actually simulate what’s happening on the real ocean surface.
University of Rennes · inria · Odyssey · IRMAR-UMR
math-ph, math.MP, physics.ao-ph
Submitted: 2026-08-27
Updated: 2026-11-13
License: http://creativecommons.org/licenses/by-nc-nd/4.0/
Importance score: 82/100
The gist: This paper develops a comprehensive stochastic variational framework for surface gravity waves that bridges deterministic phase-resolved and stochastic phase-averaged descriptions The resulting…
Key concepts
- Stochastic Variational Framework
- This method uses a variational principle extended with stochastic terms to model wave dynamics. It bridges deterministic phase-resolved descriptions with stochastic phase-averaged descriptions by systematically incorporating noise representing unresolved scales into the governing equations.
- Hamiltonian Structure Preservation
- The framework ensures that key physical invariants, like total energy, are respected statistically by preserving the Hamiltonian structure of the Zakharov–Craig–Sulem formulation. This is crucial for maintaining consistency between large-scale dynamics and correlation-mode evolution.
- Geometric Optics Interpretation
- Unresolved noise correlation modes are treated as active geometrical objects governed by a Hamilton–Jacobi equation. This means they are not just passively advected fluctuations but have trajectories determined by the Hamiltonian, linking wave physics to ray-tracing dynamics.
Terminology
Summary
This paper develops a comprehensive stochastic variational framework for surface gravity waves that bridges deterministic phase-resolved and stochastic phase-averaged descriptions The resulting coupled system, combining large-scale dynamics with correlation-mode evolution, provides a systematic framework for deriving stochastic models of ocean waves
Theoretical Foundation and Hamiltonian Structure
The framework starts from Luke’s variational principle for irrotational, incompressible free-surface flow This principle is extended by introducing a decomposition of the velocity potential into a large-scale deterministic component and a regularized stochastic noise term representing unresolved scales A pathwise variational principle yields stochastic counterparts of the Laplace, Bernoulli, and kinematic boundary conditions The resulting coupled system preserves the Hamiltonian structure of the Zakharov–Craig–Sulem formulation This preservation ensures that important invariants such as total energy are respected in a statistical sense.
Stochastic Modeling and Noise Regularization
The velocity potential is assumed to decompose into a large-scale component and a fluctuating stochastic component The noise term is defined through a temporal mollification using a family of smooth functions hτ with compact support and unit integral This regularized noise process Wτt has variance E(Wτt)2= τ−1, so that Wτt = O(τ−1/2). The system is then formulated using a second variational principle in expectation to provide evolution equations for the noise correlation functions This approach reconciles phase-averaged and phaseresolved descriptions by representing unresolved scales through a systematic stochastic closure. p. 1 In this sense, the approach reconciles phase-averaged and phaseresolved descriptions, offering a unified framework in which the effects of unresolved scales are represented through a systematic stochastic closure. The formalism is fully stochastic and introduces
stochastic partial differential equations to describe wave evolution, thereby providing a principled
alternative to empirical source-term parameterizations.Where the paper says it
Asymptotic Approximations and Physical Regimes
The paper derives simplified models through asymptotic assumptions applied to the nondimensional equations These approximations are restricted to the shallow-water regime, characterized by the small shallowness parameter µ = H2/L2 ≪ 1 In the long-wave, small-amplitude limit, the theory reduces to a stochastic Airy-wave model in which unresolved fluctuations act as a superposition of random linear waves on a deterministic carrier For coastal regimes with weaker scale separation, stochastic Serre–Green–Naghdi and Boussinesq-type systems were obtained.
Geometric Optics Interpretation of Correlation Modes
The noise correlation modes satisfy a Hamilton–Jacobi equation with associated ray-tracing dynamics This complex Hamiltonian structure highlights that unresolved correlation modes are active geometrical objects rather than passively advected fluctuations The leading-order ray trajectories are determined by the real part of the Hamiltonian, which coincides with advection by the resolved horizontal flow.
Conclusion and Future Directions
The framework suggests that stochastic parameterization of unresolved scales do not need to be introduced ad hoc but are rather constrained by variational structure and conservation principles The geometric-optics interpretation of the noise modes provides a direct connection with ray-tracing and spectral numerical methods The companion paper suggests that stochastic transport by unresolved scales may contribute significantly alongside to classical four-wave interaction mechanisms in realistic ocean conditions. p. 1 We further analyze explicit solutions via WKB approximations
[55, 67] and show that the noise correlation functions satisfy a Hamilton-Jacobi equation with
ray-tracing dynamics, linking the stochastic framework to geometric optics and wave-packet propagation. A companion paper [49] develops, for a linear stochastic coupled system, the asymptotic
approximations, kinetic theory, and geophysical implications, including a detailed comparison with
Hasselmann-type nonlinearities using JONSWAP observations [34], and suggests that stochastic
transport by unresolved scales can dominate classical four-wave interactions across a broad range
of realistic ocean conditions. A companion paper [49] develops, for a linear stochastic coupled system, the asymptotic
approximations, kinetic theory, and geophysical implications, including a detailed comparison with
Hasselmann-type nonlinearities using JONSWAP observations [34], and suggests that stochastic
transport by unresolved scales can dominate classical four-wave interactions across a broad range
of realistic ocean conditions. This is reminiscent to the attempt of defining simpler stochastic linear
models in wavenumber space to represent wave interactions [3].Where the paper says it
Where the paper says it
Dirichlet-to-Neumann Operator and Shape Derivative
The Dirichlet-to-Neumann operator is defined as G[η]φ = ∂nϕz=η, where ∂n denotes the outward normal derivative at the free surface The shape derivative of the Dirichlet-to-Neumann operator is given by δG[η]φ · ζ = −Gη − ∇ · (ζV).
Hamiltonian Reduction under the Stochastic Bernoulli Constraint
The stochastic Bernoulli constraint is interpreted as a reduction constraint imposed on the admissible fluctuating potentials This restriction restricts the admissible stochastic coupling so that the decorrelation limit remains well defined. The resulting stochastic water-wave model therefore remains canonical, with interaction energy ZDhφτG[η]φ, dx. p. 1 The resulting coupled system preserves the Hamiltonian structure of the Zakharov–
Craig–Sulem formulation [13, 59, 69], ensuring that important invariants such as total energy are
respected in a statistical sense. We further analyze explicit solutions via WKB approximations
[55, 67] and show that the noise correlation functions satisfy a Hamilton-Jacobi equation with
ray-tracing dynamics, linking the stochastic framework to geometric optics and wave-packet propagation.Where the paper says it
Preservation of the Poisson Structure
Since the fluctuating potential is prescribed rather than promoted to an independent canonical variable, it does not introduce an additional canonical variable The resulting stochastic water-wave model therefore remains canonical, with interaction energy ZDhφτG[η]φ, dx and the original Poisson bracket on the large-scale variables (η, φ).
Final Stochastic Water Wave Models
The paper presents two main systems: one with finite decorrelation time and one in the time-decorrelated limit The system with time-decorrelated velocity noise retains the full nonlinear water-wave Bernoulli equation when correlation time is sufficiently large. The dynamics of the noise correlation modes are governed by a nonlinear stochastic partial differential equation rather than by a linear transport equation with random coefficient. p. 1 We further analyze explicit solutions via WKB approximations
[55, 67] and show that the noise correlation functions satisfy a Hamilton-Jacobi equation with
ray-tracing dynamics, linking the stochastic framework to geometric optics and wave-packet propagation. A companion paper [49] develops, for a linear stochastic coupled system, the asymptotic
approximations, kinetic theory, and geophysical implications, including a detailed comparison with
Hasselmann-type nonlinearities using JONSWAP observations [34], and suggests that stochastic
transport by unresolved scales can dominate classical four-wave interactions across a broad range
of realistic ocean conditions.Where the paper says it
Geometric-optics Solution for the Correlation Modes
The noise potential may be expanded as ϕi(x, z, t) = Z Rdϕbi(k, z, t)eik·x dk The complex dispersion (eikonal) relation is ωj = kj · uz=η − ikj ∂zϕz=η. The complex Hamiltonian structure highlights that unresolved correlation modes are active geometrical objects rather than passively advected fluctuations. p. 9 From a geometrical viewpoint, the condition
∂zϕ
τG[η]φ = 0,
expresses orthogonality between fluctuating vertical motions and resolved normal transport at the
free surface. It removes stochastic work in the normal direction, so unresolved fluctuations act
tangentially to the evolving interface and do not modify the large-scale Hamiltonian flux through
the boundary.Where the paper says it
Ornstein-Uhlenbeck Process for the Regularized Noise
A simple explicit closure is obtained by modeling the regularized noise through an Ornstein–Uhlenbeck (OU) process In this closure, the nonlinear dynamics of the unresolved fluctuations is replaced by a linear mean-reverting Ornstein–Uhlenbeck process.
Conclusion
Taken together, these results suggest that stochastic variational modelling provides a unified framework in which unresolved dynamics become explicit dynamical variables rather than prescribed stochastic forcings By combining Hamiltonian structure, multiscale asymptotics and dynamically evolving correlation modes within a single variational formulation, the present work establishes a foundation for a new class of stochastic water-wave models capable of consistently representing transport, uncertainty and wave interactions across scales.
Dirichlet-to-Neumann Operator and its Shape Derivative
The operator G[η] is self-adjoint and non-negative The variation of the quadratic form associated with the Dirichlet-to-Neumann operator satisfies δδη 1/2φG[η]φ = 1/2∇φ2 − 1/2G[η]φ + ∇η · ∇φ21 + ∇η2.
The condition ∂zϕτz=ηG[η]φ = 0 removes the coupling between fluctuating vertical motions and large-scale normal transport at the interface This condition restricts the admissible stochastic coupling without introducing an additional dynamical degree of freedom.
The stochastic Bernoulli constraint (168) and the compatibility condition (169) restrict the admissible prescribed fluctuations, but do not generate independent phase-space constraints Consequently, they do not deform the canonical bracket.
The system with time-decorrelated velocity noise yields a large-scale dynamic boundary condition that reads ∂tφ + η + ϵ2ν∇φ2 − ϵµ2ν1/µGµ,ϵ[η]φ + ϵ∇η · ∇φ21 + ϵ2µ∇η2 = 0. The dynamics of the noise correlation modes is now governed by a nonlinear stochastic partial differential equation rather than by a linear transport equation with random coefficient. p. 1 We further analyze explicit solutions via WKB approximations
[55, 67] and show that the noise correlation functions satisfy a Hamilton-Jacobi equation with
ray-tracing dynamics, linking the stochastic framework to geometric optics and wave-packet propagation. A companion paper [49] develops, for a linear stochastic coupled system, the asymptotic
approximations, kinetic theory, and geophysical implications, including a detailed comparison with
Hasselmann-type nonlinearities using JONSWAP observations [34], and suggests that stochastic
transport by unresolved scales can dominate classical four-wave interactions across a broad range
of realistic ocean conditions.Where the paper says it
The complex Hamiltonian structure highlights that unresolved correlation modes are active geometrical objects rather than passively advected fluctuations The first Hamilton equation dxjdt = u(xj) − ikjkj ∂zϕ(xj) drives both the transport and the complex modulation of the wave packets.
The unresolved fluctuations therefore evolve as mean-reverting stochastic modes with correlation time τ. Such OU closures have been widely used in ocean and climate modelling to represent unresolved processes with finite temporal memory. p. 1 A companion paper [49] develops, for a linear stochastic coupled system, the asymptotic
approximations, kinetic theory, and geophysical implications, including a detailed comparison with
Hasselmann-type nonlinearities using JONSWAP observations [34], and suggests that stochastic
transport by unresolved scales can dominate classical four-wave interactions across a broad range
of realistic ocean conditions. This is reminiscent to the attempt of defining simpler stochastic linear
models in wavenumber space to represent wave interactions [3].Where the paper says it
The framework also opens several directions for future work. The companion paper suggests that stochastic transport by unresolved scales may contribute significantly alongside to classical four-wave interaction mechanisms in realistic ocean conditions.
The condition ∂zϕτz=ηG[η]φ = 0 removes the coupling between fluctuating vertical motions and large-scale normal transport at the interface. This condition restricts the admissible stochastic coupling without introducing an additional dynamical degree of freedom.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Stochastic parameterization of unresolved scales can be constrained by variational structure and conservation principles, limiting spurious dissipation or artificial growth in models. This allows AI systems to generate more physically consistent stochastic models by deriving noise terms from underlying physical constraints rather than introducing them ad hoc.
-
The framework accommodates rotational stochastic forcing while preserving a potential-flow description for the large scales, suggesting a route toward coupling surface-wave dynamics and oceanic turbulence. AI systems can model complex ocean phenomena by incorporating rotational noise into wave models without losing the structure of the resolved flow equations.
-
The geometric-optics interpretation of noise modes provides a direct connection with ray-tracing and spectral numerical methods, opening a practical path toward efficient implementations. This allows AI to develop more efficient numerical solvers for stochastic water waves by leveraging geometric optics approximations for correlation modes.
-
The system can be solved explicitly in Fourier space (assuming lateral periodic condition), or equivalently through a slow-envelope approximation and a WKB expansion, yielding explicit solutions for wave amplitudes and potential fields in the long-wave limit. AI systems can utilize these explicit analytical solutions to rapidly predict wave evolution under simplified conditions.
-
The noise correlation modes continuously adapt their geometry and amplitude to the large-scale dynamics, described by a complex Hamiltonian structure involving ray trajectories. This allows AI models to represent unresolved fluctuations as dynamically evolving wave packets that are transported, stretched and modulated by the resolved flow, rather than just passively advected ones.
-
The closure for the regularized noise can be modeled through an Ornstein–Uhlenbeck process, providing a simple colored-noise closure for unresolved intermittency. AI systems can use this mean-reverting stochastic process to model unresolved intermittency with finite temporal memory, leading to more realistic representations of complex wave statistics.
-
The derivation of simplified models under different asymptotic assumptions yields specific dynamics, such as the stochastic Airy-wave model or the stochastic Saint-Venant equations depending on the nonlinearity parameter epsilon. AI systems can automatically select and apply these appropriate reduced models based on input parameters like wave amplitude and depth to optimize computational efficiency while maintaining physical accuracy.
Abstract
This paper develops a comprehensive stochastic variational framework for surface gravity waves. Starting from Luke's variational principle for irrotational, incompressible free-surface flow, we introduce a decomposition of the velocity potential into a large-scale deterministic component and a regularized stochastic noise term representing unresolved scales. A path-wise variational principle yields the stochastic counterparts of the Laplace, Bernoulli, and kinematic boundary conditions. To close the system, a second variational principle in expectation is formulated, providing evolution equations for the noise correlation functions. The resulting coupled system preserves the Hamiltonian structure of the Zakharov--Craig--Sulem formulation. We further analyze explicit solutions via WKB approximation and show that the noise correlation functions satisfy a Hamilton-Jacobi equation with ray-tracing dynamics. This framework provides a rigorous foundation for reduced stochastic models of ocean waves and for the kinetic theory developed in the companion paper.
Sources
Related papers
- Existence of a fluid phase for the two-dimensional one-component plasma
- The Huang-Yang conjecture for the low-density Fermi gas
- Anderson localisation in infinite and finite resonator crystals with long-range interactions
- Free energy and spectral edge of the SYK model
- Anderson localization for 1-d quasi-periodic Schr"odinger operators with degenerate weights
- Irregular SLE(4) martingales and isomonodromic deformations