Wave propagation through periodic arrays of freely floating rectangular floes

This paper investigates the two-dimensional propagation of small-amplitude waves through an infinite periodic array of freely floating rectangular floes by combining Bloch-Floquet theory with Galerkin's method to derive a dispersion relation that captures hydrodynamic coupling and fluid resonance, ultimately providing accurate explicit approximations for wave behavior in broken ice fields.

The Gist

Imagine the ocean is covered in a giant, broken sheet of ice, like a puzzle that has been shattered into thousands of floating rectangular pieces. When a wave rolls through this broken ice, it doesn't just push the ice forward; it makes the ice pieces bob up and down, slide back and forth, and even tilt side-to-side.

This paper asks a very specific question: If we pretend the ice pieces are glued together with no gaps between them, and they can only bob up and down, do we get a good enough answer to predict how the waves move?

The authors, Lloyd Dafydd and Richard Porter, say: "It's complicated, but mostly yes, with some surprising twists."

Here is the breakdown of their study using simple analogies:

1. The Setup: A Dance Floor of Ice

Think of the ocean surface as a dance floor covered in floating rectangular rafts (the ice floes).

• The Old Model: Previous studies pretended these rafts were glued together with no space between them. They could only move up and down (like a piston in an engine). This is called the "mass-loading" model. It's simple, but is it realistic?
• The New Model: In reality, there are small gaps of water between the ice chunks. The authors wanted to see if allowing the ice to slide (surge) and tilt (pitch) through those gaps changes the wave behavior significantly.

2. The Physics: The "Three-Step" Dance

When a wave hits a floating ice chunk, it doesn't just do one thing. It triggers a three-part dance:

1. Heave: The ice bobs up and down.
2. Surge: The ice slides forward and backward.
3. Pitch: The ice tilts like a seesaw.

The water in the gaps between the ice chunks acts like a resonant organ pipe. If the wave frequency matches the size of the gap, the water sloshes violently, creating a "resonance" that changes how the wave travels.

3. The Method: The "Infinite Hall of Mirrors"

To solve this, the authors used a mathematical trick called Bloch-Floquet theory.

• The Analogy: Imagine you are in a hallway with mirrors on both sides. If you look in the mirror, you see yourself, and then you see the reflection of the reflection, stretching on forever.
• The Math: Instead of simulating a million ice chunks, they simulated just one chunk and its immediate neighbors, then told the math to "repeat this pattern forever." This allowed them to calculate exactly how the wave speed (wavenumber) changes based on the wave's frequency.

4. The Results: What They Found

The Good News (The "Simple" Answer)

For low-frequency waves (long, slow swells), the old, simple model (glued ice, only bobbing) is actually very accurate.

• The Takeaway: If you just want to know how fast the wave is moving through the ice, you don't need to worry about the gaps or the tilting. The "glued ice" model works great.

The Bad News (The "Surprising" Answer)

While the speed of the wave is predicted correctly by the simple model, the movement of the ice itself is completely different.

• The Twist: In the real world (with gaps), the ice isn't just bobbing up and down. It is performing a circular motion.
• The Analogy: Imagine a child on a swing. The simple model says the child is just moving up and down. The real model shows the child is actually moving in a circle (up, forward, down, back). The timing of the wave is the same, but the dance is totally different. The ice is bobbing and sliding at the same time, out of sync with each other.

The "Pitch" Surprise

For square-shaped ice chunks, there is a second, hidden "lane" of wave travel.

• The Analogy: Imagine a highway with two lanes. The simple model only sees the fast lane (bobbing). The new model reveals a slow lane where the ice chunks are mostly tilting (pitching) back and forth.
• The Catch: This "tilting lane" only exists if the ice chunks are roughly square. If the ice chunks are long and thin (like a long raft), this tilting lane disappears, and the ice just bobs.

5. Why Does This Matter?

This research is crucial for understanding broken sea ice in the Arctic and Antarctic.

• Climate Models: Scientists use these models to predict how much energy waves lose as they hit the ice. If the ice absorbs energy differently because it's tilting and sliding (rather than just bobbing), our predictions for how fast ice melts or how far waves penetrate the ice pack could be wrong.
• The Verdict: The simple model is a great shortcut for calculating wave speed, but if you want to know exactly how the ice is moving or how it interacts with the water in the gaps, you need the complex model.

Summary

The paper proves that while the "glued ice" model is a fantastic shortcut for predicting how fast waves travel through broken ice, it misses the chaotic dance the ice actually performs. The ice doesn't just bob; it slides and tilts in a circular motion, and for square ice chunks, there's an extra "tilting wave" mode that the simple model completely ignores.

In short: The map (the wave speed) is accurate, but the terrain (the ice movement) is much more complex than we thought.

Technical Summary

1. Problem Definition and Motivation

The paper investigates the two-dimensional propagation of small-amplitude waves through an infinite periodic array of freely floating rectangular ice floes. This study serves as a critical extension to previous work by the authors (Dafydd & Porter, 2024, 2026) which modeled broken ice as a continuous, zero-gap mass-loading system where floes were constrained to move only in heave (vertical motion).

Key Questions:

• Does the simplified "zero-gap, heave-only" model accurately represent the physics of a broken ice cover where small fluid gaps exist between floes?
• How do the additional rigid-body degrees of freedom—surge (horizontal translation) and pitch (rotation)—couple with heave via the fluid to alter wave dispersion?
• What is the impact of fluid resonance within the narrow gaps between floes on wave propagation, particularly at low frequencies?

The physical motivation is the modeling of wave attenuation in Marginal Ice Zones (MIZ), where discrepancies exist between theoretical predictions (based on mass-loading) and field measurements regarding frequency-dependent attenuation.

2. Methodology

Mathematical Formulation

• Governing Equations: The problem is modeled using inviscid, linearized potential theory for an incompressible fluid of infinite depth. The fluid velocity is derived from a velocity potential $\phi$ satisfying Laplace's equation.
• Boundary Conditions:
  • Free Surface: Linearized kinematic and dynamic conditions.
  • Floe Motion: The floes are free to move in heave ($\zeta$), surge ($\xi$), and pitch ($\theta$). Kinematic conditions are applied to the floe base and sidewalls.
  • Periodicity: Bloch-Floquet theory is employed to model the infinite array. The solution in a fundamental cell ($0 < x < L$) is linked to neighboring cells via a dimensionless Bloch wavenumber $kL$.
• Dynamic Equations: The motion of the floe is governed by equations of motion for the three rigid-body modes. These equations couple the modes through hydrodynamic forces and moments generated by the fluid.

Solution Strategy

• Decomposition: The velocity potential is decomposed into three components corresponding to forced heave, surge, and pitch motions.
• Integral Equations: The problem is reduced to solving integral equations for the vertical fluid velocity across the gap between floes. The kernel of these equations is derived from matching separation solutions in the fluid domains above and below the floe draft.
• Dispersion Relation: The dispersion relationship (linking frequency $\omega$ and wavenumber $k$) is determined by the vanishing of the determinant of a $3 \times 3$ matrix. This matrix encodes the hydrodynamic coupling between the three modes (heave, surge, pitch) and includes terms for solid inertia, added fluid inertia, and hydrostatic restoring forces.
• Numerical Solution: Galerkin's method is used to approximate the integral equations. The basis functions are chosen to capture the singular behavior of the flow near the sharp corners of the floes (proportional to $x^{-1/3}$).

Asymptotic Analysis

A significant portion of the methodology focuses on deriving explicit analytical approximations for the case where the fluid gap ($\ell$) is small compared to the floe draft ($d$), defined by the small parameter $\epsilon = \ell/d \ll 1$.

• The authors derive asymptotic expansions for the hydrodynamic force coefficients ($F^{(a,b)}$) up to $O(\epsilon)$.
• Special attention is paid to terms involving $\cot(kL/2)$, which become significant near $kL \approx 0$ and $kL \approx 2\pi$ (long wavelengths and standing wave limits), preventing the breakdown of the approximation near resonance.

3. Key Contributions

1. Generalized Rigid-Body Model: The paper presents a comprehensive spectral problem for freely floating rectangular floes that fully couples heave, surge, and pitch. This is distinct from previous models that often neglected surge or assumed zero draft.
2. Explicit Small-Gap Approximations: The authors derived simple, explicit formulas for the dispersion relation in the limit of small gaps. These approximations are shown to be highly accurate across a wide range of frequencies and gap sizes, offering a computationally efficient alternative to full numerical simulations.
3. Discovery of Low-Frequency Pitch Modes: The study reveals the existence of a distinct low-frequency wave propagation branch dominated by pitch motion. This mode was not captured in previous heave-only models and exists even at frequencies well below the fluid resonance in the gaps.
4. Circular Floe Motion: Through modal amplitude analysis, the authors demonstrate that the dominant low-frequency wave mode is not purely heave. Instead, it represents a circular motion of the floes, resulting from a roughly equal combination of heave and surge with a quarter-period phase shift.
5. Validation of Mass-Loading Model (with caveats): The study confirms that for low frequencies, the dispersion relation of the complex free-floating system is well-approximated by the simpler mass-loading (heave-only, zero-gap) model. However, it clarifies that while the dispersion is similar, the underlying kinematics (actual floe motion) are fundamentally different.

4. Results

• Dispersion Curves: The paper presents dispersion curves ($kL$ vs. $K\hat{\rho}d$) for various aspect ratios ($L/d$) and gap sizes ($\epsilon$).
  • Heave-Constrained: Matches the mass-loading model at low frequencies.
  • Surge-Constrained: Shows solutions primarily in the range $1 \le K\hat{\rho}d \le 4$, with symmetry breaking near $kL=0, 2\pi$ due to gap resonance effects.
  • Pitch-Constrained: Reveals a low-frequency branch extending from $(0,0)$ to $(2\pi, 0)$. This branch is highly sensitive to the aspect ratio; as $L/d$ increases, the pitch branch shifts to higher frequencies and becomes less prominent.
• Coupled Modes:
  • Heave-Surge: The coupled dispersion diagram largely superimposes the individual modes, with saddle-point bifurcations where curves intersect.
  • Heave-Pitch & Surge-Pitch: Coupling suppresses the heave resonance asymptote ($K\hat{\rho}d=1$) found in the constrained case. In surge-pitch coupling, one component branch disappears from the combined diagram.
• Aspect Ratio Effects: As the floes become longer (higher $L/d$), the influence of the pitch mode diminishes (scaling with $(d/L)^3$), and the system behavior becomes dominated by the heave mode.
• Resonance: Near the fluid resonance frequency in the narrow gaps ($K\hat{\rho}d \approx 1$), the asymptotic approximations break down, and full numerical solutions are required to capture strong modal interactions.

5. Significance and Implications

• Ice Modeling: The findings suggest that while the simple mass-loading model is sufficient for predicting the speed (dispersion) of low-frequency waves in broken ice, it fails to capture the true kinematics of the ice floes. The presence of surge and pitch, and the resulting circular motion, could significantly impact energy dissipation mechanisms and wave attenuation rates.
• Multi-Modal Propagation: The identification of a low-frequency pitch-dominated branch implies that wave energy in broken ice may propagate via multiple distinct modes simultaneously. This complexity challenges the assumptions of single-mode scattering theories used in current ensemble-averaged models for the Marginal Ice Zone.
• Computational Efficiency: The derived explicit asymptotic formulas provide a powerful tool for rapid estimation of dispersion relations in large-scale ice models without the computational cost of solving full integral equations, provided the gaps are small and frequencies are not near resonance.
• Future Directions: The paper highlights the need to extend these models to finite water depths and three-dimensional geometries (including roll motion) to fully understand the modal structure of wave-ice interactions.

In conclusion, Dafydd and Porter demonstrate that introducing small gaps and allowing full rigid-body motion reveals a rich and complex wave propagation landscape. While the simplified heave-only model remains a robust predictor of wave speed, the physical reality of ice floe motion is far more dynamic, involving coupled circular motions and additional low-frequency pitch modes that could alter our understanding of wave attenuation in polar regions.