Traveling waves in a continuum model of schooling swimmers

This paper presents a continuum model of schooling swimmers interacting through temporally nonlocal hydrodynamic forces, demonstrating that such interactions can destabilize uniform schools into stable, coarsening traveling waves of dense sub-schools.

The Gist

Imagine a school of fish swimming in a perfect line, like a train of cars on a single track. For a long time, scientists have wondered: How do they stay organized? Is it because they are looking at each other and reacting, or is there a hidden physical force at play?

This paper suggests that the answer lies in the water itself. It proposes that fish don't just swim through empty space; they swim through a "ghostly" history of the water left behind by the fish in front of them.

Here is the story of the paper, broken down into simple concepts:

1. The "Ghost Wake" Analogy

When a fish flaps its tail, it doesn't just push water away; it creates a swirling vortex, like a whirlpool that lingers for a moment before fading away.

• The Old Way: Most models assume fish react instantly to their neighbors, like people in a crowd bumping into each other.
• The New Idea: This paper says fish are more like drivers on a highway who can "feel" the turbulence left by the car ahead, even if that car passed a few seconds ago. The water has a memory. The fish in the back are swimming through the lingering "ghosts" of the wakes created by the fish in front.

2. The "Memory Lane" Equation

The researchers built a mathematical model to describe this. Imagine every fish leaves a trail of invisible, wiggly ribbons in the water that slowly fade away.

• If a fish swims into a downward ripple from a neighbor's wake, it gets pushed forward (thrust).
• If it swims into an upward ripple, it gets slowed down (drag).
• Because the ribbons take time to fade, the fish are reacting to the past positions of their neighbors, not just where they are right now. This creates a system of "time-delayed" interactions.

3. The "Traffic Jam" Surprise

The team asked: "What happens if you have a huge, perfectly uniform school of fish, all swimming at the same speed with equal spacing?"

• The Result: They found that this perfect order is actually unstable. It's like a line of cars on a highway that is perfectly spaced out; eventually, tiny bumps in the road (or small changes in speed) cause the line to break apart.
• The Breakdown: The uniform school spontaneously shatters into clumps. You get dense groups of fish (let's call them "sub-schools") separated by empty gaps.

4. The "Traveling Wave" Phenomenon

Here is the most fascinating part: These clumps don't just sit there. They move!

• Imagine a wave of traffic congestion moving backward through a line of cars. In this fish school, the dense clumps of fish form a traveling wave.
• The "sub-schools" (the crowded parts) and the "gaps" (the empty parts) travel together as a single, moving pattern.
• The paper shows that these waves can be stable. You can have a school that looks like a uniform line, or a school that looks like a moving train of dense clusters, and both can exist under the same conditions. It's like having a highway where you can drive in a smooth line or in a rhythmic pattern of stop-and-go waves, and both are stable.

5. The "Finite School" Experiment

The researchers also tested what happens if the school isn't infinite (like a long line) but is a finite group, like a real school of fish with a front and a back.

• The Front: The front of the school slowly spreads out into the empty water ahead, like a fan opening up.
• The Back: The back of the school develops a sharp drop-off, like a cliff where the fish suddenly stop.
• The Middle: Inside this expanding school, the traveling waves of dense clusters form and grow. Over time, these waves merge and simplify (a process called "coarsening"), leaving fewer, larger clumps of fish moving together.

The Big Picture

The main takeaway is that hydrodynamics (the physics of water flow) alone is enough to create complex, organized patterns in fish schools. You don't necessarily need the fish to be "smart" or following complex social rules. The simple fact that they leave a lingering wake that affects the fish behind them is enough to turn a uniform line of swimmers into a dynamic, wave-like structure.

It's as if the water itself is conducting an orchestra, turning a group of individual swimmers into a synchronized, wave-moving collective.

Technical Summary

Problem Statement
The physical and sensory mechanisms driving organized collective behavior in schooling fish remain elusive, particularly regarding how hydrodynamic flows generated by individual swimmers influence large-scale collective patterns. While prior research has established that hydrodynamic interactions mediate schooling behavior (e.g., energy savings in schools) and has modeled these interactions using discrete particle models or spatially nonlocal sensory models, there is a lack of continuum descriptions for swimmers interacting via high-Reynolds number flows. A critical challenge is that these interactions are temporally nonlocal: vortical wake structures generated by a swimmer persist over timescales comparable to the interaction time between fish, unlike the instantaneous interactions found in low-Reynolds number (Stokes) flow systems like bacteria. Existing models often fail to capture this fluid-mediated memory, which is essential for understanding the emergence of complex formations in large schools.

Methodology
The authors develop a continuum theory for a school of swimmers in an inline formation, modeling the swimmers as flapping wings.

1. Discrete Model: The system begins with a set of nonlinear delay-differential equations (DDEs) describing $N$ swimmers. Each swimmer's velocity is determined by a self-propulsion speed and hydrodynamic forces arising from the vortical wakes of upstream swimmers. The interactions are non-reciprocal (downstream swimmers are affected by upstream wakes, but not vice versa) and temporally nonlocal, governed by time delays dependent on the swimmer's position and the decay of the wake.
2. Coarse-Graining: To derive a continuum description, the authors introduce new field variables, $C(x,t)$ and $S(x,t)$, which represent the aggregated hydrodynamic influence of the wakes. This transformation converts the system of DDEs with state-dependent time delays into a system of ordinary differential equations (ODEs) for the fields.
3. Continuum Limit: By taking the limit as $N \to \infty$, the authors derive a system of three coupled nonlinear partial differential equations (PDEs) governing the swimmer density $\varrho$, and the fields $C$ and $S$. A diffusive term is added to regularize the equations and account for potential random variations in flapping parameters.
4. Analysis: The authors perform a linear stability analysis of the uniform-density steady state to identify instability regimes. They then conduct numerical simulations using a pseudospectral method in both periodic domains and finite domains with compactly supported initial data to observe the nonlinear evolution of these instabilities.

Key Contributions and Results

• Continuum PDE Derivation: The paper successfully derives a continuum PDE model for schooling swimmers that explicitly accounts for temporally nonlocal hydrodynamic interactions through fluid-mediated memory, a feature previously difficult to incorporate analytically in high-Reynolds number regimes.
• Linear Instability: Linear stability analysis reveals that a uniform-density school is unstable to perturbations within a specific range of swimmer densities ($\varrho_0 \in (\varrho_-, \varrho_+)$), provided the dimensionless flapping frequency $\alpha$ exceeds a critical threshold. The instability is characterized by a complex eigenvalue, indicating that the uniform state destabilizes into dispersive traveling waves.
• Traveling Wave Solutions: Numerical simulations in periodic domains demonstrate that the uniform school evolves into stable traveling waves consisting of densely populated "sub-schools" separated by sparse regions. The study identifies families of stable traveling wave solutions corresponding to different numbers of density oscillations ($n$).
  • The wave speed decreases as the mean density increases and as the number of oscillations increases.
  • Multiple stable traveling wave modes can coexist for the same set of kinematic parameters (density, frequency, diffusivity), suggesting a form of multistability.
  • The most unstable wavenumber predicted by linear theory aligns with the wavenumbers of the stable traveling waves observed in simulations.
• Finite School Dynamics: Simulations of a "finite school" (initially compact support) show that the interior destabilizes into propagating waves. The overall envelope of the school evolves according to a quasistatic model, characterized by a rarefaction fan expanding upstream and a terminating shock downstream. The internal waves coarsen and become trapped within this contracting envelope.

Significance
The paper claims that temporally nonlocal hydrodynamic interactions alone are sufficient to generate rich collective behavior, specifically traveling wave instabilities, in schools of swimmers. This provides a physical mechanism for the formation of "sub-schools" and density waves without invoking complex behavioral rules or sensory feedback. The results offer a theoretical framework that bridges the gap between discrete hydrodynamic models and continuum theories, providing testable predictions regarding the relationship between swimmer density, flapping frequency, and the speed and structure of collective waves. The authors suggest that these findings could help rationalize the complex dynamics observed in fish schools, such as shimmering waves, and provide a foundation for future studies incorporating behavioral mechanisms or three-dimensional motion.