---
title: 'Euler–Hamilton Equation: Canonical Symmetry'
url: https://www.emergentmind.com/topics/euler-hamilton-equation
type: topic
---

# Euler–Hamilton Equation: Canonical Symmetry

The Euler–Hamilton equation refers collectively to canonical equations of motion that emerge from Hamiltonian formulations in both particle mechanics and field theory, with emphasis on systems where variables can be classified into generalized coordinates and momenta, and where symmetry between these variables is key. In modern continuum mechanics and mathematical physics, Euler–Hamilton equations appear as the defining PDEs that govern reversible (Hamiltonian) evolution of field variables. They manifest through symmetric, variationally derived structures in both Lagrangian (material-reference) and Eulerian (spatial) frames, and generalize to Lie-Poisson and semidirect-product formulations for complex media [1907.03396][1212.1955].

## 1. Canonical Hamiltonian Formalism and Symmetric Euler–Hamilton Equations

The canonical Hamiltonian formalism begins by associating to every generalized coordinate $q_s$ a conjugate momentum $p_s$. For many systems, notably with second-order Lagrangians, the standard Hamilton equations are:
\[
\dot{q}_s = \frac{\partial H}{\partial p_s},\qquad \dot{p}_s = -\frac{\partial H}{\partial q_s}
\]
For field theories and systems with first-order Lagrangians, Guo Liang and Qi Guo [1212.1955] showed that the canonical equations generalize to a "beautifully symmetric" form. For a system split into $v$ independent coordinates $q_\lambda$ and $p$ independent momenta $p_\eta$ (with $v+p=N$), and with remaining variables constrained by functions $f_a$ and $g_\alpha$, the Euler–Hamilton equations for the Hamiltonian density $h$ are:
\[
\frac{\delta h}{\delta q_\lambda} = \sum_{s=1}^N \left[\,\dot{p}_s \frac{\partial g_s}{\partial q_\lambda} - \dot{q}_s \frac{\partial f_s}{\partial q_\lambda}\right] + \sum_{a=v+1}^{N}\partial_x\left(\frac{\partial h}{\partial q_{a,x}}\right)\frac{\partial f_a}{\partial q_\lambda}
\]
\[
\frac{\delta h}{\delta p_\eta} = \sum_{s=1}^N \left[\,\dot{q}_s \frac{\partial g_s}{\partial p_\eta} - \dot{p}_s\frac{\partial f_s}{\partial p_\eta}\right] + \sum_{a=v+1}^N \partial_x\left(\frac{\partial h}{\partial q_{a,x}}\right)\frac{\partial f_a}{\partial p_\eta}
\]
This symmetric form applies to both first- and second-order systems, automatically reducing to the standard canonical equations when all variables are independent and constraints $f_a,g_\alpha$ vanish.

## 2. Lagrangian and Eulerian Frames in Continuum Mechanics

In continuum mechanics, evolution equations can be formulated in either the Lagrangian frame (fixed to "continuum particles") or the Eulerian frame (fixed in space). In the Lagrangian frame, labels $X \in \Omega_0 \subset \mathbb R^3$ denote material points whose current positions are $x(t, X)$. Conjugate momentum densities $M(t, X)$ are defined per reference volume, leading to observables $F[x, M]$.

The Lagrangian Poisson bracket between two observables $F, G$ is:
\[
\{F, G\}^{(L)} = \int_{\Omega_0} dX\, \left( \frac{\delta F}{\delta x^i(X)} \frac{\delta G}{\delta M_i(X)} - \frac{\delta G}{\delta x^i(X)}\frac{\delta F}{\delta M_i(X)} \right)
\]
This bracket is antisymmetric, obeys the Leibniz rule, and satisfies the Jacobi identity.

Transformation to the Eulerian frame is accomplished by pushforward through the inverse map $X = X(x)$, producing Eulerian fields—density $\rho(x)$, momentum $m_i(x)$, entropy $s(x)$, and deformation gradient $F^i{}_I(x) = \partial x^i/\partial X^I |_{X = X(x)}$. The induced Eulerian Poisson bracket encodes the evolution structure of the system in spatial coordinates [1907.03396].

## 3. Eulerian Poisson Brackets and Hamiltonian Structure

In the Eulerian frame, the system is governed by a Poisson bracket for functionals $\mathcal F[\rho, m, s, F], \mathcal G[\rho, m, s, F]$:
\[
\{F, G\}^{(\mathrm{Eu})} = \{F, G\}^{(\mathrm{FM})} + \int_\Omega dx\, F^j{}_I \Big( \partial_j \frac{\delta F}{\delta m_i} \frac{\delta G}{\delta F^i{}_I} - \partial_j \frac{\delta G}{\delta m_i} \frac{\delta F}{\delta F^i{}_I}\Big)
\]
\[
-\int_\Omega dx\, \partial_k F^i{}_I \Big( \frac{\delta F}{\delta F^i{}_I} \frac{\delta G}{\delta m_k} - \frac{\delta G}{\delta F^i{}_I} \frac{\delta F}{\delta m_k}\Big)
\]
where $\{F, G\}^{(\mathrm{FM})}$ is the standard compressible-fluid bracket on $(\rho, m, s)$.

The Hamiltonian functional (total energy)
\[
H[\rho, m, s, F] = \int_\Omega dx\, \left( \frac{1}{2}\frac{m_i m_i}{\rho} + \varepsilon(\rho, s, F)\right)
\]
generates ideal reversible dynamics, where $\varepsilon$ is the internal/elastic energy density.

## 4. Euler–Hamilton Equations of Motion in Field Theory

Given a Poisson bracket structure and Hamiltonian, the Euler–Hamilton equations prescribe the time evolution of primary field variables (state variables). For continuum mechanics in the Eulerian frame, they are:
\[
\begin{cases}
\partial_t \rho = -\partial_i(\rho H_{m_i}) \\
\partial_t m_i = -\partial_j(m_i H_{m_j}) - \rho \partial_i H_\rho - m_j \partial_i H_{m_j} - s \partial_i H_s - F^j{}_I \partial_i H_{F^j{}_I} + \partial_j(F^j{}_I H_{F^i{}_I}) \\
\partial_t s = -\partial_i(s H_{m_i}) \\
\partial_t F^i{}_I = -H_{m_k} \partial_k F^i{}_I + F^j{}_I \partial_j H_{m_i}
\end{cases}
\]
Here $H_{m_i} = \delta H/\delta m_i$ (the Eulerian velocity), $H_\rho$, $H_s$, $H_{F^i{}_I}$ are the functional derivatives with respect to their arguments. The equations enforce conservation of mass, momentum, and entropy, and govern advection of the deformation gradient.

A direct translation to evolution equations for the distortion matrix $A = F^{-1}$ is immediate when $A$ is the preferred variable.

## 5. Lie–Poisson and Semidirect Product Structure

The generalized Poisson structure for fluid–elasticity systems is an example of a Lie–Poisson bracket on the dual of a semidirect-product Lie algebra:
\[
\mathfrak{g} = \mathrm{Vect}(\Omega) \ltimes \left[ C^\infty(\Omega) \oplus C^\infty(\Omega) \oplus \Gamma(T^1_1 \Omega)\right]
\]
with $\mathrm{Vect}(\Omega)$ the Lie algebra of vector fields (fluid velocities), and the densities $(\rho,s)$ and tensor $F^i{}_I$ forming a linear representation. The semidirect-product form expresses how the momentum density $m_i$ couples to advected fields via the coadjoint action.

Schematically, if functionals $F,G$ are considered over field variables, the Poisson bracket reads:
\[
\{F, G\} = \langle m, [F_m, G_m] \rangle + \langle \rho, \pounds_{F_m} G_\rho - \pounds_{G_m} F_\rho \rangle + \langle F\text{-tensor}, \pounds_{F_m} G_F - \pounds_{G_m} F_F\rangle
\]
where $F_m = \delta F/\delta m$, and $\pounds_{F_m}$ is the Lie derivative along the field $F_m$. This formulation underpins the geometric origin of the bracket and elucidates its conservation properties [1907.03396].

## 6. Jacobi Identity, Hyperbolicity, and Conservation Laws

The Jacobi identity for the Poisson bracket is essential: it ensures consistent composition of time evolutions (Hamiltonian flows), closure of conserved vector fields under commutation, and gauge-theoretic self-consistency of the Poisson bivector. For hydrodynamic-type brackets,
\[
\{q^\alpha(x), q^\beta(y)\} = g^{\alpha\beta}(q) \partial_x \delta(x-y) + b^{\alpha\beta}_\gamma(q)\, \partial_x q^\gamma \delta(x-y)
\]
where $g^{\alpha\beta}(q)$ defines a pseudo-Riemannian metric and $b$ encodes the connection, the Jacobi identity follows from reduction from the canonical (Lagrangian) bracket.

If the Hamiltonian density $e(q)$ is convex, the associated evolution PDEs are symmetric hyperbolic, implying well-posedness.

Conservation laws—mass, momentum, angular momentum, and Galilean booster—are built in as Casimirs or constants of motion, each generating an infinitesimal symmetry via the Poisson bracket. Any functional $G[q]$ with $\{G, H\} = 0$ creates an infinitesimal invariant flow, establishing the direct link between conservation and gauge invariance [1907.03396].

## 7. Implications and Illustrative Examples

The symmetric form of the Euler–Hamilton equations clarifies the treatment of constrained and unconstrained Hamiltonian systems. In the absence of constraints, canonical variables are treated independently, and the formalism reduces to the classical Goldstein–type Hamiltonian system.

For first-order systems (whose Lagrangians are linear in velocities), the resulting phase-space is effectively halved and only $N$ canonical equations are obtained, compared to the $2N$ of second-order systems [1212.1955]. The nonlinear Schrödinger equation (NLSE) exemplifies such structure, where both the field and its conjugate momentum can be consistently captured by the symmetric Euler–Hamilton framework, enforcing consistency between the evolution equations for $\psi$ and $\psi^*$.

A plausible implication is that this symmetric formulation enables uniform canonical treatments of diverse physical systems—ranging from classical fields with constraints to quantum field evolutions—under a single generalized Hamiltonian paradigm. This provides both powerful analytic tools and a robust geometrical foundation for the study of reversible continuum systems.

---

**References**:  
[1907.03396]: On Hamiltonian continuum mechanics  
[1212.1955]: Canonical equations of Hamilton with beautiful symmetry

Source: https://www.emergentmind.com/topics/euler-hamilton-equation