In classical physics, we treat the flow of a fluid and the shape of a surface as distinct problems. One is governed by the conservation of momentum and mass (the Euler equations), while the other is governed by the rules of how a manifold bends and stretches in space (the Gauss-Codazzi equations). However, a growing body of research demonstrates that these two fields are mathematically dual: the "wild" oscillations of a turbulent fluid are the physical manifestation of the "wrinkled" geometry of a non-smooth surface.
The Gauss-Codazzi and Euler Equivalence
The core of this research direction is a formal mapping between the variables of fluid mechanics and the fundamental forms of differential geometry. In From the Nash--Kuiper Theorem of Isometric Embeddings to the Euler Equations for Steady Fluid Motions, Marshall Slemrod demonstrates that the second fundamental form of an isometrically embedded surface—a tensor describing how the surface curves—can be identified with the density, velocity, and pressure of a steady compressible fluid.
This is not merely a superficial similarity. As Amit Acharya, Gui‐Qiang Chen, Marshall Slemrod, and Dehua Wang show in Fluids, Elasticity, Geometry, and the Existence of Wrinkled Solutions, the equations governing these two systems are structurally identical. By mapping physical variables into geometric components, fluid dynamics can be analyzed through the lens of surface embeddings. For example, a surface with constant mean curvature (like a sphere) serves as a geometric analogue for an incompressible fluid flow.
Wrinkles as Wild Solutions
One of the most striking implications of this duality involves the Nash-Kuiper theorem, which describes how surfaces can be "wrinkled" via non-smooth isometric embeddings. In fluid dynamics, "wild" solutions—irregular, non-unique velocity fields—have long been a mathematical challenge.
The research suggests these are two sides of the same coin. The high-frequency geometric wrinkles found in low-regularity embeddings correspond exactly to high-frequency oscillations in a fluid's velocity and density. This allows researchers to apply techniques from geometry to solve fluid problems and vice versa. For instance, Some recent developments on isometric immersions via compensated compactness and gauge transforms explains how techniques for handling shock waves in fluid dynamics (like vanishing viscosity) can be used to reconstruct "rough" geometric shapes.
The Role of Curvature in Boundary Layers
When fluids move through curved domains, the geometry of the boundary itself dictates the physics. Standard models often fail here because they assume flat or simplified boundaries. Gui‐Qiang Chen and Zhongmin Qian argue in The Inviscid Limit of the Navier-Stokes Equations with Kinematic and Navier Boundary Conditions that simplified slip conditions are mathematically incompatible with the behavior of an ideal fluid on a curved manifold. To maintain mathematical consistency, boundary conditions must explicitly account for the surface's shape operator and second fundamental form.
This geometric sensitivity is also present in transport equations. Jianfeng Lu, Qin Li, and Weiran Sun show in Validity and Regularization of Classical Half-Space Equations that grazing angles in curved geometries create singularities that traditional models struggle to resolve. Similarly, Mohamed Ghattassi and Nader Masmoudi demonstrate that in radiative heat transfer, the curvature of the domain prevents solutions from converging unless the coordinate systems are aligned with the intrinsic geometry of the boundary.
Geometry as a Computational Tool
Beyond theoretical duality, this perspective offers new ways to simulate complex systems. In Fluid structure interaction with curved space lattice Boltzmann, Miller Mendoza Jimenez and colleagues propose treating deforming boundaries not as moving objects, but as changes in the spatial metric.
By borrowing the mathematics of general relativity, they represent fluid-structure interactions as local inertial forces within a fixed grid. Instead of re-meshing a grid every time a boundary moves, the solver simply updates the "curvature" of the space the fluid lives in. This transforms a difficult topological problem (tracking a moving interface) into a geometric one (calculating metric-induced forces), providing a more stable and efficient path for simulating how fluids and structures interact.