A boundary element regularised Stokeslet method applied to cilia and flagella-driven flow

A boundary element implementation of the regularised Stokeslet method of Cortez is applied to cilia and flagella-driven flows in biology. Previously-published approaches implicitly combine the force discretisation and the numerical quadrature used to evaluate boundary integrals. By contrast, a boundary element method can be implemented by discretising the force using basis functions, and calculating integrals using accurate numerical or analytic integration. This substantially weakens the coupling of the mesh size for the force and the regularisation parameter, and greatly reduces the number of degrees of freedom required. When modelling a cilium or flagellum as a one-dimensional filament, the regularisation parameter can be considered a proxy for the body radius, as opposed to being a parameter used to minimise numerical errors. Modelling a patch of cilia, it is found that: (1) For a fixed number of cilia, reducing cilia spacing reduces transport. (2) For fixed patch dimension, increasing cilia number increases the transport, up to a plateau at $9\times 9$ cilia. Modelling a choanoflagellate cell it is found that the presence of a lorica structure significantly affects transport and flow outside the lorica, but does not significantly alter the force experienced by the flagellum.

💡 Research Summary

The paper presents a novel implementation of the regularised Stokeslet method within a boundary‑element framework, targeting flows driven by cilia and flagella in biological systems. Traditional regularised Stokeslet approaches intertwine the discretisation of surface forces with the quadrature used to evaluate the associated boundary integrals. Consequently, the regularisation parameter (ε) and the mesh spacing (h) become tightly coupled, forcing ε to be chosen primarily for numerical stability rather than representing a physical length scale.

In contrast, the authors separate force discretisation from integral evaluation. Forces are expanded in simple basis functions (constant or linear) over each boundary element, while the integrals are computed with high‑order numerical quadrature or analytically where possible. This decoupling weakens the ε‑h relationship: ε can now be interpreted as a proxy for the filament radius, and h can be refined independently to achieve the desired accuracy. The result is a dramatic reduction in the number of degrees of freedom, because a single basis function per element replaces the per‑node force vectors required in earlier schemes.

The method is first validated on a one‑dimensional filament model of a single cilium/flagellum. By setting ε equal to the physical radius of the filament, the authors demonstrate that the flow field converges to the expected Stokes‑flow solution and that the regularisation no longer serves merely as a numerical artifact.

Next, the authors explore collective effects in a patch of cilia. Two key observations emerge: (1) For a fixed number of cilia, decreasing the spacing between them reduces net material transport. The reduction is attributed to stronger hydrodynamic interference, which diminishes the effective pumping speed of each cilium. (2) When the overall patch size is held constant, increasing the number of cilia initially raises transport, but the benefit saturates at a 9 × 9 array. Beyond this size, additional cilia contribute little because the flow field becomes saturated and further cilia are largely hydrodynamically screened. These findings provide quantitative guidance for the design of artificial ciliated surfaces and for interpreting the functional limits of natural ciliary carpets.

Finally, the method is applied to a choanoflagellate cell model that includes a lorica—a porous, cage‑like extracellular structure surrounding the flagellum. Simulations reveal that the lorica markedly reshapes the flow outside the cage, reducing the far‑field velocity magnitude, yet it does not significantly alter the force experienced by the flagellum itself. This suggests that the lorica functions primarily as a flow‑modulating scaffold, protecting the organism or enhancing nutrient capture without imposing a large mechanical load on the flagellum.

Overall, the paper delivers a robust, computationally efficient boundary‑element regularised Stokeslet technique that decouples physical and numerical parameters, reduces computational cost, and yields new insights into cilia‑ and flagellum‑driven microflows. The approach is readily extensible to more complex geometries, time‑dependent beating patterns, and could become a standard tool for researchers studying low‑Reynolds‑number locomotion, microfluidic device design, and the biomechanics of microorganisms.