Ocean engineering relies on a quantitative understanding of fluid mechanics, structural dynamics, and sediment transport. At the heart of this discipline lies the application of differential equations—mathematical tools that describe how physical quantities change in space and time. From the propagation of swell across entire ocean basins to the turbulent impact of a breaking wave on a seawall, ordinary and partial differential equations (ODEs and PDEs) provide the language for predicting and managing these phenomena. This article outlines the fundamental role of differential equations in wave modeling and coastal protection, detailing the specific equations used, the numerical methods required to solve them, and their application in designing the next generation of resilient coastal infrastructure.

Foundations of Wave Theory: From Linear to Nonlinear Models

The mathematical modeling of ocean waves begins with the fundamental laws of fluid dynamics. Engineers typically treat water as a homogeneous, incompressible fluid with negligible viscosity for many large-scale wave propagation problems. The starting point for most models is the Navier-Stokes equations, but significant simplifications are made to arrive at tractable models suitable for engineering design.

Linear (Airy) Wave Theory

The most basic theoretical model is linear wave theory, often called Airy wave theory. It assumes waves have a small amplitude (H) relative to their wavelength (L) and water depth (h). This allows engineers to neglect nonlinear convective acceleration terms in the governing equations. The flow is assumed to be irrotational and incompressible, allowing the definition of a velocity potential, φ. The governing equation reduces to Laplace's equation:

∇2φ = 0

This elliptic PDE is one of the simplest yet most powerful in classical field theory. Its solutions, combined with appropriate boundary conditions at the seafloor and the free surface, yield a complete description of the wave kinematics and dynamics. The linearized free surface boundary conditions are a type of initial-boundary value problem. Solving these leads to the dispersion relation, which links wave frequency (ω), wave number (k), and water depth (h):

ω2 = g k tanh(kh)

This relation is fundamental. It shows that wave speed (c = ω / k) depends on wavelength and water depth. In deep water, the relation simplifies to c = gT/(2π), meaning longer waves travel faster. In shallow water, it becomes c = √(gh), where wave speed depends only on depth. This simple PDE model allows engineers to transform wave parameters from deep water to the shoreline, a process essential for design wave estimation. The total wave energy per unit area, E = (1/8) ρgH2, is also derived from this linear framework.

Stokes, Cnoidal, and Solitary Wave Theories

Linear theory breaks down as waves steepen or approach the shore. When wave amplitude is not negligible compared to wavelength, nonlinear effects become important. Stokes wave theory adds higher-order correction terms to the linear solution. The second-order Stokes solution, for instance, shows that wave crests are sharper and troughs are flatter than linear theory predicts, and it generates a net mass transport (Stokes drift).

In very shallow water, waves become highly nonlinear. Here, Cnoidal wave theory provides a better description. These solutions are expressed in terms of Jacobian elliptic functions and approach solitary waves in the extreme limit. Solitary waves are stable humps of water that travel long distances without changing shape. The mathematical formulation of solitary waves was a key historical success of nonlinear wave theory. The Korteweg-de Vries (KdV) equation describes the evolution of weakly nonlinear, weakly dispersive waves in shallow water:

∂u/∂t + α u ∂u/∂x + β ∂3u/∂x3 = 0

This PDE is famous for its exact solution—the soliton—which balances nonlinear wave steepening with dispersion. Understanding these equations is vital for modeling tsunami propagation and wave run-up on coastal structures.

Boussinesq-Type Equations for Nearshore Processes

For engineering applications in the nearshore zone (where depth varies and waves transform through shoaling, refraction, and diffraction), Boussinesq equations are widely used. These PDEs incorporate the vertical structure of the flow, accounting for the effects of steep slopes and rapidly varying bathymetry. They can simulate wave groups, harmonic generation, and wave-induced currents that are primary drivers of sediment transport and coastal morphology. Modern Boussinesq models can accurately predict wave heights, periods, and directions at a specific coastal site, providing direct input for the design of breakwaters, jetties, and beach nourishment projects.

Numerical Methods for Solving Ocean Wave PDEs

Analytical solutions to the wave equations discussed are rare and limited to highly idealized geometries and boundary conditions. In engineering practice, solving these differential equations requires numerical discretization. Three main classes of numerical methods dominate coastal and ocean engineering.

Finite Difference Methods (FDM)

FDM is the most intuitive approach. It replaces the continuous derivatives in a PDE with algebraic difference quotients evaluated on a structured grid. For example, the second derivative in the one-dimensional wave equation can be approximated by a central difference scheme. FDM is straightforward to implement and works well for simple, rectangular domains. The numerical stability of FDM schemes is governed by the Courant-Friedrichs-Lewy (CFL) condition, a constraint on the time step relative to the grid spacing and wave speed. While efficient, FDM struggles with complex geometries like around a breakwater or an irregular coastline.

Finite Element Methods (FEM)

FEM is the standard tool for structural analysis and is also heavily used in computational fluid dynamics. It divides the computational domain into a mesh of small elements (triangles or quadrilaterals in 2D). The solution is approximated by a set of shape functions within each element. FEM excels at handling complex boundaries and graded meshes, where resolution can be refined in areas of interest (e.g., near a seawall) and coarsened elsewhere. This makes it ideal for modeling wave interaction with coastal structures. The Galerkin method is a typical approach for deriving the algebraic equations from the PDEs in FEM.

Boundary Element Methods (BEM)

For problems governed by Laplace's equation (like potential flow wave theory), BEM offers a computationally efficient alternative. It transforms the PDE in the volume into an integral equation on the boundary. Only the boundary of the domain needs to be discretized, which reduces the dimensionality of the problem by one. BEM is very effective for modeling wave diffraction and radiation around harbors and large offshore structures. However, it becomes more complex for nonlinear problems or when bulk properties like turbulence are important.

Applying Differential Models to Coastal Protection Systems

Coastal protection structures—seawalls, revetments, breakwaters, and dikes—are designed to resist wave forces, reduce erosion, and prevent flooding. The design process relies on differential equations to calculate design loads and predict performance.

Shallow Water Equations for Storm Surge and Tsunami Modeling

The shallow water equations (SWE) are a set of hyperbolic PDEs derived by depth-integrating the Navier-Stokes equations. They describe the evolution of water depth and horizontal flow velocities. They are the backbone of storm surge modeling. The National Oceanic and Atmospheric Administration (NOAA) uses the SLOSH model to predict hurricane storm surge. This model solves the SWE on a polar grid, incorporating wind stress, atmospheric pressure gradients, and the Coriolis effect. These simulations are used to generate evacuation maps and inform emergency management. The two-dimensional SWE can be written as:

∂h/∂t + ∂(hu)/∂x + ∂(hv)/∂y = 0 (Continuity)

∂(hu)/∂t + ∂(hu2 + gh2/2)/∂x + ∂(huv)/∂y = Sx (Momentum)

where h is total water depth, u and v are depth-averaged velocities, and Sx represents source terms like wind stress, bottom friction, and the Coriolis effect.

Wave Run-up, Overtopping, and Reflection

Engineers use empirical formulas derived from physical and numerical experiments to compute wave run-up on dikes and seawalls. These formulas are often based on the Iribarren number (ξ), a dimensionless parameter relating wave steepness to structure slope. Wave run-up (Ru) is proportional to the significant wave height (Hs) and the Iribarren number. The differential equations governing wave energy flux are used to compute wave heights at the toe of the structure. Overtopping discharge (q) is then estimated using formulas from the EurOtop Manual. These flow rates are critical for determining the required crest height of a seawall or dike to prevent flooding.

Sediment Transport and Morphological Modeling

Coastal protection is not just about barriers; it also involves managing sediment. The Exner equation is a conservation law that describes the evolution of the seabed elevation (zb) due to spatial gradients in sediment transport rate (qs):

(1-p) ∂zb/∂t = -∂qs/∂x

where p is the sediment porosity. This PDE links coastal hydrodynamics to morphology. To solve it, engineers must provide a sediment transport formula (e.g., Engelund-Hansen, Soulsby-van Rijn) that relates qs to the bed shear stress computed from the wave and current fields. These coupled models solve the SWE along with the Exner equation to predict beach erosion, shoal formation, and the response of the coastline to human interventions like dredging or breakwater construction. A decade-long simulation of coastline evolution requires solving these equations for millions of time steps.

Major Engineering Case Studies

Real-world projects demonstrate the power and necessity of these mathematical models.

Maeslantkering Storm Surge Barrier, Netherlands

The Maeslantkering is a massive storm surge barrier that protects Rotterdam. It consists of two floating arms, 210 meters long, that pivot into position across the Nieuwe Waterweg waterway. The decision to close is based on a real-time PDE solver that runs the shallow water equations on a supercomputer. The solver assimilates data from tide gauges, weather buoys, and wind forecasts. The computational domain covers the entire North Sea. If the model predicts a water level exceeding 3 meters above NAP (Amsterdam Ordnance Datum) in Rotterdam, the barrier is closed. This automated process is a direct, high-stakes application of differential equations to civil infrastructure. Without these predictive models, the barrier could not operate safely.

The Delta Works and Eastern Scheldt Barrier

The Netherlands' Delta Works is a series of dams, sluices, locks, dikes, and storm surge barriers. The Eastern Scheldt barrier is a unique structure designed to remain open under normal conditions but close during storms. Its design was heavily influenced by advanced mathematical modeling (using Delft3D) that predicted the impact of the barrier on the tidal regime and ecology of the estuary. Engineers solved the shallow water equations to determine the size and placement of the 65 concrete piers and 62 steel gates. The barrier's design is an iconic example of using PDEs to balance safety with environmental sustainability. Ensuring the long-term stability of these structures continued refinement of morphological models to predict scour around the piers.

As climate change leads to rising sea levels and potentially more intense storms, the demand for accurate, high-resolution, and fast predictive models continues to grow. Traditional PDE solvers can be computationally intensive, limiting their use in real-time control or probabilistic risk assessment.

Physics-Informed Neural Networks (PINNs)

A transformative approach is the use of Physics-Informed Neural Networks (PINNs). These are deep learning models that are trained to satisfy the governing PDEs. The neural network learns to map inputs (e.g., location, time, wave conditions) to outputs (e.g., wave height, velocity). The loss function of the network includes the residual of the PDE itself. This allows the model to learn from data while strictly respecting the laws of physics. PINNs can solve PDEs much faster than traditional numerical methods once trained, enabling real-time simulation. They are being explored for applications like digital twins of coastal zones and rapid uncertainty quantification.

Uncertainty Quantification and Probabilistic Design

Engineering design is moving from deterministic factors of safety to probabilistic frameworks. Engineers must quantify the probability of failure. This requires solving PDEs for thousands of different input scenarios (e.g., different storm intensities, sea-level rise projections, material strength variations). The use of surrogate models (like Gaussian processes or neural networks) built from high-fidelity PDE simulations is becoming standard. This allows the propagation of uncertainties through the model, providing engineers and stakeholders with a clear picture of the risks involved. The American Society of Civil Engineers (ASCE) standards increasingly emphasize reliability-based design, directly driving the need for advanced computational PDE solvers.

Conclusion

Differential equations are the language of ocean engineering. They translate the physical laws of fluid dynamics and structural mechanics into predictive models that directly inform the design and operation of coastal protection systems. From the simplest linear wave theory to complex nonlinear models and advanced computational methods like FEM and PINNs, these mathematical tools allow engineers to understand, predict, and manage the ocean's powerful forces. As climate change reshapes our coastlines, the importance of these models will only increase. The synergy between physics, mathematics, and computer science will continue to drive innovation, ensuring that coastal communities can adapt and remain resilient in the face of rising seas and intensifying storms.