Ideal Fluid Mechanics: Euler's Equations, Bernoulli's Equation, and Vorticity

Intuition Beginner

Hold a strip of paper just below your lower lip and blow across the top surface. The strip rises. The air you blow over the top is moving faster than the still air below, and the faster-moving air exerts less pressure. The pressure difference pushes the paper up. This is the Bernoulli effect in action, and it is the same principle that keeps airplanes aloft.

An ideal fluid (also called an inviscid fluid) is one with no internal friction -- no viscosity. Water and air have low viscosity, so in many situations they behave nearly ideally. The key property of an ideal fluid is that the internal forces are purely pressure forces: the fluid can push on itself (normal stress) but cannot exert shearing forces (tangential stress). This is why an ideal fluid at rest is described by a single number -- the pressure $p$ -- at each point, rather than the full stress tensor needed for solids 09.07.02.

Why does an airplane wing generate lift? The wing is shaped so that air flows faster over the top surface than the bottom. By Bernoulli's principle, faster flow means lower pressure. The pressure on the bottom of the wing exceeds the pressure on top, and the net upward force is lift. The standard textbook explanation (equal transit time for upper and lower flows) is misleading -- the real story involves the Kutta condition at the trailing edge and circulation. But the core physical fact is correct: the pressure difference between the upper and lower surfaces is what holds the plane up.

Now think about what happens when you pull the plug in a bathtub. The water does not just drain straight down -- it spirals. A vortex forms. In an ideal fluid, this vortex has a remarkable property: once circulation is established, it cannot be created or destroyed. This is Kelvin's circulation theorem. Real fluids (which have viscosity) can create and destroy vorticity at boundaries, but in the ideal limit, vorticity is frozen into the fluid -- it moves with the material. Tornadoes are an extreme example: a column of intensely rotating air where the vorticity has been concentrated by stretching, the same way a figure skater spins faster when pulling in their arms.

The three pillars of ideal fluid mechanics are:

1. The Euler equation -- Newton's second law for an inviscid fluid, relating acceleration to pressure gradients and body forces.
2. Bernoulli's equation -- An energy-like integral of the Euler equation along streamlines, trading pressure for speed.
3. Vorticity -- A measure of local rotation in the fluid, defined as the curl of the velocity field, whose dynamics reveal the topological structure of the flow.

Visual Beginner

Concept	Physical meaning	Key equation
Pressure difference	Faster flow = lower pressure	Bernoulli: $p+\frac{1}{2}\rho v^2+\rho gh=\text{const}$ along a streamline
Vorticity $\mathbf{\omega }$	Local spinning of the fluid	Twice the angular velocity of a tiny fluid element
Circulation $\Gamma$	Total spin around a loop	Average tangential speed times loop perimeter

Worked example Beginner

A garden hose has a nozzle that narrows from diameter 2 cm to 0.5 cm. Water flows in at speed 1 m/s. What is the speed at the nozzle exit, and how does the pressure change?

The volume flow rate is constant (water is incompressible): speed times area is the same at both ends. The inlet area is $A_1=\pi \times (0.01{)}^2=3.14\times 10^{-4}{\mathrm{m}}^2$. The exit area is $A_2=\pi \times (0.0025{)}^2=1.96\times 10^{-5}{\mathrm{m}}^2$. By continuity: $v_2=v_1\times A_1/A_2=1.0\times 16=16m/s$.

Bernoulli's equation (no height change, horizontal pipe): $p_1+\frac{1}{2}\rho v_1^2=p_2+\frac{1}{2}\rho v_2^2$. With $\rho =1000{kg/m}^3$: $p_1-p_2=\frac{1}{2}\times 1000\times (16^2-1^2)=500\times 255=127,500\mathrm{Pa}$. The pressure drops by about 1.26 atmospheres. If the inlet pressure were 2 atm, the exit pressure would be about 0.74 atm -- below atmospheric. This is how a Venturi meter measures flow speed: the pressure drop at the constriction tells you the velocity.

Check your understanding Beginner

Formal definition Intermediate+

The velocity field and the material derivative

A fluid is described by the velocity field $\mathbf{v}(\mathbf{x},t)$, which gives the velocity of the fluid element at position $\mathbf{x}$ at time $t$. This is an Eulerian description: we watch what happens at fixed points in space, not what happens to individual fluid parcels.

The material derivative (also called the convective derivative or substantial derivative) is the rate of change following a fluid element:

$$\frac{D}{Dt}=\frac{\partial }{\partial t}+\mathbf{v}\cdot \nabla .$$

The first term $\partial /\partial t$ is the local rate of change (what a stationary observer sees). The second term $\mathbf{v}\cdot \nabla$ is the convective rate of change: even in a steady flow (where $\partial /\partial t=0$ everywhere), a fluid element accelerates as it moves into regions of different velocity. This convective term is what makes fluid dynamics nonlinear, and it is the source of most of the mathematical difficulty of the subject.

The continuity equation

Conservation of mass for an infinitesimal fluid element gives:

$$\frac{\partial \rho }{\partial t}+\nabla \cdot (\rho \mathbf{v})=0.$$

For an incompressible fluid (constant density $\rho$), this reduces to:

$$\boxed{\nabla \cdot \mathbf{v}=0.}$$

The velocity field is solenoidal (divergence-free). This is an excellent approximation for water and for low-speed air flows (Mach number much less than 1). Incompressibility is not a property of the fluid itself but of the flow regime -- air is effectively incompressible at speeds below about 100 m/s.

The Euler equation

Newton's second law ($\mathbf{F}=m\mathbf{a}$) applied to an infinitesimal fluid element of an ideal (inviscid) fluid gives the Euler equation:

$$\boxed{\rho \frac{D\mathbf{v}}{Dt}=-\nabla p+\rho \mathbf{g},}$$

or equivalently, expanding the material derivative:

$$\rho \left(\frac{\partial \mathbf{v}}{\partial t}+(\mathbf{v}\cdot \nabla )\mathbf{v}\right)=-\nabla p+\rho \mathbf{g},$$

where $p(\mathbf{x},t)$ is the pressure field and $\mathbf{g}$ is the gravitational body force per unit mass. The left side is mass times acceleration (per unit volume). The right side is the sum of the pressure force ($-\nabla p$: fluid accelerates from high pressure toward low) and the gravitational force. There is no viscous force because the fluid is ideal.

The Euler equation was derived by Leonhard Euler in 1757 and is the fundamental equation of inviscid fluid dynamics. Together with the continuity equation and appropriate boundary and initial conditions, it determines the evolution of the velocity and pressure fields.

Bernoulli's equation

For steady, incompressible, inviscid flow along a streamline, the Euler equation integrates to Bernoulli's equation:

$$\boxed{p+\frac{1}{2}\rho v^2+\rho gz=\text{const (along a streamline)}.}$$

Here $z$ is the vertical height and $v=|\mathbf{v}|$ is the speed. The three terms are:

• $p$ -- the static pressure (thermodynamic pressure).
• $\frac{1}{2}\rho v^2$ -- the dynamic pressure (kinetic energy per unit volume).
• $\rho gz$ -- the hydrostatic pressure (gravitational potential energy per unit volume).

Bernoulli's equation states that the total mechanical energy per unit volume is constant along a streamline. Faster flow means lower static pressure, and vice versa. This is the basis for the Venturi meter, the Pitot tube (airspeed indicator), the carburetor, and the lift on an airfoil.

If the flow is also irrotational ($\nabla \times \mathbf{v}=0$), the Bernoulli constant is the same on all streamlines, and Bernoulli's equation holds throughout the flow field.

Vorticity

The vorticity is defined as the curl of the velocity field:

$$\boxed{\mathbf{\omega }=\nabla \times \mathbf{v}.}$$

Vorticity measures the local rotation of the fluid. A fluid element at a point where $\mathbf{\omega }\ne 0$ is spinning -- it has angular velocity $\mathbf{\omega }/2$. If you place a tiny cross-shaped paddlewheel in the flow, it will rotate if and only if the vorticity at that point is nonzero. In flow that is irrotational ($\mathbf{\omega }=0$ everywhere), the paddlewheel translates but never turns.

For two-dimensional flow with velocity $(u(x,y,t),v(x,y,t),0)$, the vorticity has only a $z$-component: ${\omega }_z=\partial v/\partial x-\partial u/\partial y$.

Kelvin's circulation theorem

The circulation around a closed curve $C$ is defined as:

$$\Gamma ={\oint }_C\mathbf{v}\cdot d\mathbf{l}.$$

By Stokes' theorem, $\Gamma ={\int }_S\mathbf{\omega }\cdot \hat{n}dS$ where $S$ is any surface spanning $C$.

Kelvin's circulation theorem (1869): For an ideal, barotropic fluid (pressure is a function of density only) under conservative body forces, the circulation around any closed material curve $C(t)$ (a curve that moves with the fluid) is conserved:

$$\frac{D\Gamma }{Dt}=0.$$

This is a powerful topological constraint on the flow. It implies that vorticity cannot be created or destroyed in the interior of an ideal fluid -- it can only be transported, stretched, and tilted. Vorticity enters or leaves the fluid only at boundaries (where viscosity, ignored in the ideal model, becomes important).

Potential flow

A flow is irrotational if $\mathbf{\omega }=0$ everywhere. Then the velocity can be written as the gradient of a velocity potential $\varphi$:

$$\mathbf{v}=\nabla \varphi .$$

Incompressibility $\nabla \cdot \mathbf{v}=0$ then gives Laplace's equation:

$${\nabla }^2\varphi =0.$$

This linear equation is the foundation of potential flow theory. Solutions can be superposed (the sum of two solutions is also a solution), which makes analytic progress possible. Complex-variable methods (conformal mapping) provide exact solutions for two-dimensional flows around airfoils, cylinders, and other shapes. The pressure is recovered from Bernoulli's equation.

Key derivation Intermediate+

Derivation of Bernoulli's equation from the Euler equation

Start with the steady Euler equation for incompressible flow: $\rho (\mathbf{v}\cdot \nabla )\mathbf{v}=-\nabla p+\rho \mathbf{g}$. Use the vector identity $(\mathbf{v}\cdot \nabla )\mathbf{v}=\nabla (v^2/2)-\mathbf{v}\times \mathbf{\omega }$. The gravitational force is $\rho \mathbf{g}=-\rho g\nabla z$. The Euler equation becomes:

$$\rho \nabla (v^2/2)-\rho \mathbf{v}\times \mathbf{\omega }=-\nabla p-\rho g\nabla z.$$

Taking the dot product with $\mathbf{v}$ along a streamline: $\mathbf{v}\cdot (\mathbf{v}\times \mathbf{\omega })=0$ (the cross product is perpendicular to $\mathbf{v}$). The remaining terms give:

$$\mathbf{v}\cdot \nabla \left(\frac{1}{2}\rho v^2+p+\rho gz\right)=0.$$

The directional derivative of $p+\frac{1}{2}\rho v^2+\rho gz$ along the streamline direction $\mathbf{v}$ is zero -- hence this quantity is constant along each streamline.

The vorticity equation

Taking the curl of the Euler equation (for incompressible, barotropic flow with conservative body forces) yields the vorticity equation (Helmholtz equation):

$$\frac{\partial \mathbf{\omega }}{\partial t}+(\mathbf{v}\cdot \nabla )\mathbf{\omega }=(\mathbf{\omega }\cdot \nabla )\mathbf{v}.$$

In material derivative form:

$$\boxed{\frac{D\mathbf{\omega }}{Dt}=(\mathbf{\omega }\cdot \nabla )\mathbf{v}.}$$

The right side represents vortex stretching and tilting: vorticity aligned with a stretching flow is amplified (a vortex tube being stretched gets more intense, like the ice-skater pulling in their arms). If the flow is two-dimensional, $(\mathbf{\omega }\cdot \nabla )\mathbf{v}=0$ and vorticity is simply convected: $D\omega /Dt=0$.

Worked example: flow past a circular cylinder

For a uniform stream $U$ flowing past a cylinder of radius $a$ in two dimensions, the complex potential is $w(z)=U(z+a^2/z)$ where $z=x+iy$. The velocity potential is $\varphi =U(r+a^2/r)\cos \theta$ in polar coordinates $(r,\theta )$. The velocity components are:

$$v_r=U\left(1-\frac{a^2}{r^2}\right)\cos \theta ,v_{\theta }=-U\left(1+\frac{a^2}{r^2}\right)\sin \theta .$$

On the cylinder surface ($r=a$): $v_r=0$ (no-penetration condition) and $v_{\theta }=-2U\sin \theta$. The maximum speed $2U$ occurs at $\theta =\pi /2$ (top and bottom of the cylinder) -- twice the free-stream speed.

The pressure on the cylinder surface from Bernoulli: $p+\frac{1}{2}\rho v_{\theta }^2=p_{\infty }+\frac{1}{2}\rho U^2$. So $p=p_{\infty }+\frac{1}{2}\rho U^2(1-4{\sin }^2\theta )$. The pressure is symmetric front-to-back and top-to-bottom, so the net force on the cylinder is zero (d'Alembert's paradox): a body in steady potential flow experiences no drag. This contradicts experience and is resolved by viscosity and flow separation 09.07.04.

Adding circulation $\Gamma$ around the cylinder modifies the complex potential to $w(z)=U(z+a^2/z)+(i\Gamma /2\pi )\ln z$. The additional velocity $v_{\theta }=\Gamma /(2\pi r)$ breaks the top-bottom symmetry and generates a lift force $L=\rho U\Gamma$ per unit span (the Kutta-Joukowski theorem). This is the theoretical basis for airfoil lift: the circulation around the wing, established by the Kutta condition at the trailing edge, generates lift proportional to the product of free-stream speed and circulation.

Worked example: Torricelli's law

A large open tank of water with a small hole at depth $h$ below the free surface. Apply Bernoulli between a point on the surface (pressure $p_{\mathrm{atm}}$, speed $\approx 0$, height $h$) and a point at the hole (pressure $p_{\mathrm{atm}}$, speed $v$, height $0$):

$$p_{\mathrm{atm}}+0+\rho gh=p_{\mathrm{atm}}+\frac{1}{2}\rho v^2+0.$$

The exit speed is $v=\sqrt{2gh}$ -- Torricelli's law (1643). This equals the speed of a body falling freely from height $h$, a result that would later be recognised as a consequence of energy conservation. The volume flow rate is $Q=C_{d}A\sqrt{2gh}$ where $A$ is the hole area and $C_d\approx 0.6$ is a discharge coefficient that accounts for the vena contracta (the stream contracts after exiting the hole).

Bridge. The Euler equation and the vorticity equation form a tightly coupled system: the Euler equation governs the momentum, Bernoulli's equation is its energy integral, and the vorticity equation is its curl. The foundational reason the separation into irrotational and rotational parts works is that the vorticity equation has the form of a transport equation -- vorticity is carried by the flow and stretched by the velocity gradient -- which is a consequence of the material derivative structure. This is dual to the Helmholtz decomposition of the displacement field in elastic waves 09.07.02, where the irrotational part carries the P-wave and the solenoidal part carries the S-wave. In ideal fluids, the irrotational part (potential flow) satisfies Laplace's equation and the rotational part (vorticity) satisfies a nonlinear transport equation. The load-bearing structure connecting them is Kelvin's circulation theorem, which constrains the topology of vortex lines and guarantees that vorticity is frozen into the fluid in the ideal limit.

Exercises Intermediate+

Lean formalization Intermediate+

Mathlib does not contain the Euler equation for ideal fluids, Bernoulli's equation, the vorticity equation, or any aspect of fluid dynamics. The closest infrastructure is the vector calculus library (Mathlib.Analysis.Calculus.Curl, Mathlib.Analysis.Calculus.Divergence) which defines the divergence and curl on EuclideanSpace, and the PDE framework for Laplace's equation. What is missing: the material derivative $D/Dt=\partial /\partial t+\mathbf{v}\cdot \nabla$ as an operator along a time-dependent flow; the Euler equation as an evolution equation on the space of divergence-free vector fields; the incompressibility constraint $\nabla \cdot \mathbf{v}=0$ as a type-class or predicate; Bernoulli's equation as a line integral of the Euler equation; Kelvin's circulation theorem and its proof; the stream function and its relation to vorticity via Laplace's equation; and the Clebsch representation. Formalising even the statement that irrotational velocity fields derive from a potential requires the Poincare lemma, which is available in Mathlib for simply-connected domains but would need extension to the multiply-connected case relevant for circulation around obstacles. The Hamiltonian structure on the Lie algebra of volume-preserving diffeomorphisms requires infinite-dimensional Lie theory that does not exist in Mathlib. This unit ships without a Lean module.

Advanced results Master

The Clebsch representation

Every sufficiently smooth velocity field on ${\mathbb{R}}^3$ can be written in the Clebsch representation:

$$\mathbf{v}=\nabla \varphi +\alpha \nabla \beta ,$$

where $\varphi$, $\alpha$, $\beta$ are scalar functions. The vorticity is $\mathbf{\omega }=\nabla \alpha \times \nabla \beta$. This representation reduces the three components of $\mathbf{v}$ to three scalar potentials and makes the vorticity structure explicit: vortex lines lie on the intersection of surfaces of constant $\alpha$ and constant $\beta$.

The Clebsch representation is not unique (gauge freedom: $\alpha$ and $\beta$ can be reparametrised). Moreover, it cannot represent all divergence-free fields with a single pair $(\alpha ,\beta )$ in general -- a second pair may be needed for flows with linked vortex tubes. The mathematical statement is that the space of divergence-free vector fields is generated by two pairs of Clebsch potentials, and one pair suffices for flows whose vortex topology is sufficiently simple (all vortex lines are closed or go to infinity without linking). The Clebsch representation connects to the Hamiltonian formulation through the canonical variables $(\alpha ,\beta )$, which play the role of coordinates and momenta.

Hamiltonian structure of the ideal fluid: Lie-Poisson on SDiff

The Euler equation for an ideal incompressible fluid has a remarkable Hamiltonian structure discovered by Arnold (1966). The configuration space of the fluid is the group $\mathrm{SDiff}(M)$ of volume-preserving diffeomorphisms of the flow domain $M$ (a Riemannian manifold). The kinetic energy

$$H[\mathbf{v}]=\frac{1}{2}{\int }_M\rho |\mathbf{v}{|}^{2}dV$$

defines a right-invariant Riemannian metric on $\mathrm{SDiff}(M)$. The Euler equation is the geodesic equation on $\mathrm{SDiff}(M)$ with respect to this metric.

The Lie algebra of $\mathrm{SDiff}(M)$ is the space of divergence-free vector fields $\mathfrak{sdiff}(M)=\{\mathbf{v}:\nabla \cdot \mathbf{v}=0\}$ with the Lie bracket $[\mathbf{u},\mathbf{v}]=(\mathbf{u}\cdot \nabla )\mathbf{v}-(\mathbf{v}\cdot \nabla )\mathbf{u}$ (the commutator of vector fields). The dual space consists of 1-forms $\mathbf{m}$ (momentum densities) modulo gradients. The Euler equation is the Lie-Poisson equation on $\mathfrak{sdiff}(M{)}^{\ast }$:

$$\frac{\partial \mathbf{m}}{\partial t}={\mathrm{ad}}_{\mathbf{v}}^{\ast }\mathbf{m},$$

where ${\mathrm{ad}}^{\ast }$ is the coadjoint action. For the kinetic energy Hamiltonian, $\mathbf{m}=\rho \mathbf{v}$ (after projecting out the gradient part to enforce incompressibility).

This Lie-Poisson structure has two Casimir functions:

• The total vorticity $\int \mathbf{\omega }dV$ (conserved by the divergence-free constraint).
• The helicity $\mathcal{H}={\int }_M\mathbf{v}\cdot \mathbf{\omega }dV$ (a topological invariant measuring the knottedness and linkedness of vortex lines -- the Hopf invariant of the velocity field).

The helicity was identified as a Casimir by Moffatt (1969) and provides a lower bound on the kinetic energy via the Arnold inequality: $H\ge C|\mathcal{H}{|}^{2/3}$, relating the dynamics to the topology of vortex lines.

Arnold's geometric interpretation

Arnold's interpretation of the Euler equation as a geodesic on $\mathrm{SDiff}(M)$ has deep consequences:

1. Stability of stationary flows. A steady Euler flow (an equilibrium of the Lie-Poisson system) corresponds to a critical point of $H$ restricted to a coadjoint orbit. The second variation determines nonlinear stability via the Arnold method (energy-Casimir stability): if the constrained Hessian is definite, the flow is Lyapunov-stable. This gives rigorous stability criteria for shear flows, vortex patches, and Couette flow.

2. Curvature of SDiff. The sectional curvature of $\mathrm{SDiff}(M)$ can be negative, and exponentially negative curvature implies that nearby fluid configurations diverge exponentially -- a mechanism for the unpredictability of fluid motion. In two dimensions, the curvature can have either sign, but in three dimensions, the curvature is predominantly negative for generic velocity fields, suggesting that three-dimensional ideal flow is generically Lagrangian chaotic (nearby fluid elements separate exponentially fast).

3. Integrability. The Euler equation is completely integrable (in the Arnold-Liouville sense) only on low-dimensional Lie algebras. For the full $\mathfrak{sdiff}(M)$, the system has infinitely many degrees of freedom and is expected to be non-integrable. The point-vortex dynamics on the plane (a finite-dimensional truncation) is integrable for $N\le 3$ vortices and chaotic for $N\ge 4$.

Vortex dynamics

The Helmholtz vortex theorems (1858) follow from Kelvin's circulation theorem:

1. Vortex lines move with the fluid. A vortex line at time $t=0$ evolves into a vortex line at time $t$ -- the vorticity is "frozen" into the fluid.
2. Vortex tubes have constant strength. The circulation $\Gamma$ around a vortex tube is the same at every cross-section and does not change in time.
3. Vortex tubes cannot end in the fluid interior. They either close on themselves (vortex rings), end at boundaries, or extend to infinity.

Vortex stretching is the mechanism by which vorticity intensifies in three-dimensional flows. When a vortex tube is stretched (its cross-section shrinks), conservation of circulation ($\Gamma =\omega \cdot A=\text{const}$) demands that the vorticity $\omega$ increases. This is the primary mechanism for the energy cascade in three-dimensional turbulence: large-scale vortices are stretched into smaller, more intense vortices, transferring kinetic energy from large scales to small scales. In two-dimensional flow, vortex stretching is absent (vortex lines are perpendicular to the flow plane and cannot be stretched), so vorticity is merely advected and the energy cascade is inverse -- energy flows from small scales to large scales. This is why two-dimensional turbulence produces large coherent vortices (like Jupiter's Great Red Spot) while three-dimensional turbulence produces a fine-grained cascade.

Point-vortex dynamics. For two-dimensional flow, the vorticity field can be approximated by a finite collection of point vortices $\omega (x,y)={\sum }_i{\Gamma }_i\delta (\mathbf{x}-{\mathbf{x}}_i)$. Each vortex induces a velocity on all others: the velocity at ${\mathbf{x}}_j$ due to vortex $i$ is ${\mathbf{v}}_{ij}=({\Gamma }_i/2\pi )\hat{z}\times ({\mathbf{x}}_j-{\mathbf{x}}_i)/|{\mathbf{x}}_j-{\mathbf{x}}_i{|}^2$. The equations of motion are a Hamiltonian system with $N$ degrees of freedom. For $N=2$, the vortices orbit their centre of vorticity. For $N=3$, the system is integrable (the centre of vorticity, the angular impulse, and the Hamiltonian are three independent integrals). For $N\ge 4$, the system is in general chaotic (the phase space is 4-dimensional, and three integrals are not enough for integrability).

Weak solutions and the Euler equation

The Euler equation is a nonlinear PDE whose solutions may develop singularities in finite time. For two-dimensional flows, global regularity is known: the vorticity is transported by the flow and cannot develop singularities because the maximum of $|\mathbf{\omega }|$ is non-increasing (a consequence of $D\omega /Dt=0$ in 2D). For three-dimensional flows, the question of whether smooth initial data can develop a finite-time singularity is one of the open Millennium Prize Problems (Clay Mathematics Institute, $1M prize).

The mathematical framework for dealing with rough or discontinuous solutions is the theory of weak solutions (cross-ref 09.02.16 pending). A weak solution of the Euler equation satisfies the integral form of the momentum balance: for every smooth test function $\mathbf{\varphi }$ with compact support,

$${\int }_0^T{\int }_{\Omega }\left[\mathbf{v}\cdot \frac{\partial \mathbf{\varphi }}{\partial t}+(\mathbf{v}\otimes \mathbf{v}):\nabla \mathbf{\varphi }+p(\nabla \cdot \mathbf{\varphi })\right]dVdt=0.$$

In 2D, weak solutions exist globally (Yudovich, 1963) and are unique in the class $L^{\infty }(\mathbf{\omega })$. In 3D, existence of weak solutions is unknown for the Euler equation. For the viscous Navier-Stokes equation 09.07.04, Leray (1934) proved existence of weak solutions in 3D, but their uniqueness and regularity remain open.

Synthesis

The Euler equation, Bernoulli's equation, and the vorticity equation form a tightly coupled triad that governs ideal fluid flow. The Euler equation is the momentum balance; Bernoulli's equation is its energy integral along streamlines; the vorticity equation is its curl, governing the rotational structure of the flow. Kelvin's circulation theorem provides the topological constraint that links the Euler equation to the geometry of the flow domain. Arnold's geometric interpretation elevates this from a PDE system to a Hamiltonian system on an infinite-dimensional Lie group, revealing that the Euler equation is a geodesic on the group of volume-preserving diffeomorphisms, with stability determined by curvature and conserved quantities determined by the topology of vortex lines. The practical payoff is that potential flow (irrotational solutions of Laplace's equation) provides an extensive toolbox for engineering applications, while the vorticity formulation reveals the deeper dynamical structure: vortex stretching drives the energy cascade in 3D turbulence, vortex conservation underpins two-dimensional flow dynamics, and the Hamiltonian structure connects fluid mechanics to the broader framework of symplectic geometry and Lie-Poisson systems.

Connections Master

• Continuum mechanics field theory 09.07.01 provides the general Lagrangian and Eulerian descriptions of continuous media from which the Euler equation is derived as the specialisation to inviscid fluids. The stress tensor for an ideal fluid is ${\sigma }_{ij}=-p{\delta }_{ij}$, the isotropic special case of the general Cauchy stress tensor.

• Field Lagrangians 09.02.06 give the Lagrangian density for the ideal fluid from which the Euler equation and Bernoulli's equation are derived by the Euler-Lagrange equations. The fluid Lagrangian has the form $\mathcal{L}=\frac{1}{2}\rho |\mathbf{v}{|}^2-\rho e(\rho )-\rho \Phi$ where $e$ is the internal energy and $\Phi$ the gravitational potential.

• Elastic waves 09.07.02 are governed by the Navier equation, which reduces to the Euler equation when the shear modulus $\mu =0$. Only the longitudinal (acoustic) mode survives; transverse waves vanish because the fluid cannot sustain shear stress.

• Viscous flow 09.07.04 adds the viscous stress ${\tau }_{ij}=\eta (\partial v_i/\partial x_j+\partial v_j/\partial x_i)$ to the Euler equation, giving the Navier-Stokes equations. Viscosity allows vorticity to diffuse and be created at boundaries, breaking Kelvin's circulation theorem.

• Weak solutions 09.02.16 pending provide the mathematical framework for defining solutions of the Euler equation that may develop discontinuities (shocks in compressible flow) or singularities, connecting to the open problems of regularity and uniqueness in three dimensions.

• Differential geometry provides the Lie group $\mathrm{SDiff}(M)$, the Lie algebra of divergence-free vector fields, and the coadjoint orbits on which the Euler equation evolves. The helicity Casimir is a topological invariant related to the Hopf invariant of the velocity field.

Historical & philosophical context Master

Daniel Bernoulli published Hydrodynamica in 1738, establishing the principle that bears his name: the sum of pressure and kinetic energy density is constant along a streamline in steady flow. Bernoulli's work was motivated by the problem of blood pressure and the design of water supply systems. The book also introduced the kinetic theory of gases, making it a landmark in both fluid mechanics and thermodynamics.

Leonhard Euler derived the equations of motion for an ideal fluid in Principes generaux du mouvement des fluides (1757), separating the acceleration into local and convective parts and writing the first partial differential equation of continuum mechanics. Euler's equation was the first field equation in physics -- a PDE governing a continuous field rather than the trajectory of a discrete particle. Euler also introduced the concept of internal pressure and showed that the pressure in an ideal fluid is a scalar field (no shear stresses).

Hermann von Helmholtz published Uber Integrale der hydrodynamischen Gleichungen, welche den Wirbelbewegungen entsprechen (1858), establishing the vortex theorems that bear his name: vortex lines move with the fluid, vortex strength is conserved, and vortex tubes cannot end in the fluid interior. Helmholtz's work revealed that vorticity has a topological character -- it is frozen into the fluid and its conservation is tied to the connectivity of vortex lines.

William Thomson (Lord Kelvin) proved the circulation theorem in 1869, showing that circulation around a material loop is conserved in ideal flow. Kelvin's theorem is the integral form of Helmholtz's results and provides the most general statement of vorticity conservation. Kelvin was motivated by the vortex-atom theory of matter, which held that atoms were knotted vortex rings in the ether -- a hypothesis that, while incorrect, drove the development of knot theory (by Tait) and the topological study of fluid flow.

Vladimir Arnold discovered the Hamiltonian structure of the Euler equation in 1966 (Sur la geometrie differentielle des groupes de Lie de dimension infinie et ses applications a l'hydrodynamique des fluides parfaits), showing that the Euler equation is a geodesic on the group of volume-preserving diffeomorphisms. Arnold's insight unified ideal fluid mechanics with symplectic geometry and Lie group theory, opening the field of geometric hydrodynamics. Arnold and Khesin's Topological Methods in Hydrodynamics (1998) is the definitive treatment.

A philosophical point: the Euler equation is a statement of Newton's second law applied to a continuum, but its Hamiltonian structure on $\mathrm{SDiff}(M)$ is an emergent property -- there is no microscopic Hamiltonian for individual fluid parcels that directly produces this Lie-Poisson bracket. The bracket arises from the kinematics of the fluid (the fact that the configuration space is a group of diffeomorphisms) combined with the kinetic energy metric. This is another instance of the general pattern that the symplectic structure of field theories is determined by the geometry of the configuration space, not by the microscopic dynamics. The same pattern appears in magnetohydrodynamics (where the magnetic field plays the role of vorticity), in the rigid body (where the configuration space is SO(3)), and in plasma physics (where the Vlasov equation has a Lie-Poisson structure on the group of canonical transformations of phase space).

Bibliography Master

• Euler, L., "Principes generaux du mouvement des fluides," Memoires de l'Academie des Sciences de Berlin 11 (1757), 274-315.

• Bernoulli, D., Hydrodynamica, sive de viribus et motibus fluidorum commentarii (Joh. Reinhold Dulsecker, 1738).

• Helmholtz, H., "Uber Integrale der hydrodynamischen Gleichungen, welche den Wirbelbewegungen entsprechen," J. Reine Angew. Math. 55 (1858), 25-55.

• Kelvin, Lord (W. Thomson), "On vortex motion," Trans. Roy. Soc. Edinburgh 25 (1869), 217-260.

• Landau, L. D. & Lifshitz, E. M., Fluid Mechanics, 2nd ed. (Course of Theoretical Physics Vol. 6, Pergamon, 1987), §1-10.

• Taylor, J. R., Classical Mechanics (University Science Books, 2005), Ch. 14.

• Arnold, V. I., "Sur la geometrie differentielle des groupes de Lie de dimension infinie et ses applications a l'hydrodynamique des fluides parfaits," Ann. Inst. Fourier 16 (1966), 319-361.

• Arnold, V. I. & Khesin, B. A., Topological Methods in Hydrodynamics (Springer, 1998).

• Batchelor, G. K., An Introduction to Fluid Dynamics (Cambridge University Press, 2000).

• Moffatt, H. K., "The degree of knottedness of tangled vortex lines," J. Fluid Mech. 35 (1969), 117-129.

• Yudovich, V. I., "Non-stationary flow of an ideal incompressible liquid," USSR Comput. Math. & Math. Phys. 3 (1963), 1407-1456.

• Saffman, P. G., Vortex Dynamics (Cambridge University Press, 1992).

• Majda, A. J. & Bertozzi, A. L., Vorticity and Incompressible Flow (Cambridge University Press, 2002).