Matched pair analysis of the Vlasov plasma (Q2230358)

Consider a Lie algebra \(\mathfrak{g}\) that admits Lie subalgebras \(\mathfrak{s},\mathfrak{n}\subset \mathfrak{g}\) so that \(\mathfrak{g}=\mathfrak{s}\oplus \mathfrak{n}\) as a vector space. Then the Lie bracket on \(\mathfrak{g}\) can be written in terms of its restriction to \(\mathfrak{s}\) and \(\mathfrak{n}\), together with compatible actions of \(\mathfrak{s}\) on \(\mathfrak{n}\) and \(\mathfrak{n}\) on \(\mathfrak{s}\). This defines a \textit{matched pair decomposition} of \(\mathfrak{g}\) in terms of \(\mathfrak{n}\) and \(\mathfrak{s}\); we write that \(\mathfrak{g}\simeq \mathfrak{s} \bowtie \mathfrak{n}\) is the double crossed sum of \(\mathfrak{s}\) and \(\mathfrak{n}\). (This is usually called a double crossed product, but the change of terminology is useful to distinguish the Lie group setting considered in Section 6.) The overall goal of this paper is to obtain such matched pair decompositions in order to rewrite the Lie-Poisson equations associated with several important Hamiltonian functionals \(\mathcal{H}:\mathfrak{g}^\ast \to \mathbb{R}\). In Section 3, the authors investigate the case where \(\mathfrak{g}=\mathcal{IQ}\) is the space of contravariant formal tensor fields on a smooth manifold \(\mathcal{Q}\), for which the Lie bracket is the Schouten concomitant. They obtain the matched pair decomposition \(\mathcal{IQ}\simeq \mathfrak{s}\bowtie \mathfrak{n}\) where \(\mathfrak{s}\) is the Lie subalgebra of tensor fields in degrees \(0\) and \(1\), while \(\mathfrak{n}\) is the Lie subalgebra of tensor fields in degrees \(\geq 2\). This allows one to decompose the dual Poisson bracket on the space of symmetric covariant tensor fields \(\mathcal{I}^\ast\mathcal{Q}\), which is the Kupershmidt-Manin bracket. In turn, it yields a natural decomposition of the Lie-Poisson equations for a Hamiltonian functional \(\mathcal{H}:\mathcal{I}^\ast\mathcal{Q}\to \mathbb{R}\). The next aim of the paper consists in adapting this matched pair decomposition to understand the Lie-Poisson dynamics of the Vlasov plasma. This goal is achieved in Section 4, where the Lie algebra \(\mathfrak{g}=\mathcal{F}_0(T^\ast \mathcal{Q})\) of functions on \(T^\ast \mathcal{Q}\) that are polynomial in the momenta is decomposed as \(\mathcal{F}_0(T^\ast \mathcal{Q}) \simeq \hat{\mathfrak{s}}\bowtie \hat{\mathfrak{n}}\); here \(\hat{\mathfrak{s}}\) is the subspace of functions which are at most linear in momenta, while \(\hat{\mathfrak{n}}\) is the subspace of functions which are at least quadratic in momenta. The case of the Vlasov equation is obtained from the Lie-Poisson equation \(\frac{df}{dt}=-\operatorname{ad}^\ast_h(f)\) for the plasma density function \(f\in\mathcal{F}_0^\ast(T^\ast \mathcal{Q})\) when \(h\in \mathcal{F}_0(T^\ast \mathcal{Q})\) is given by \(h(q,p)=\frac{p^2}{2m}+e \phi(q)\) (\(m\) is the mass, \(e\) is the charge and \(\phi\in \mathcal{F}(\mathcal{Q})\) is the potential function). The final important result of this paper is the matched pair decomposition for the Lie-Poisson dynamics of the moment-Vlasov equations presented in Section 5. The authors work with \(\mathfrak{g}=\mathfrak{X}_{\text{Ham},0}(T^\ast\mathcal{Q})\), which is the Lie algebra of Hamiltonian vector fields (for the canonical symplectic structure on \(T^\ast\mathcal{Q}\)) associated with functions of \(\mathcal{F}_0(T^\ast\mathcal{Q})\). Thus, they can use the matched pair decomposition \(\mathfrak{X}_{\text{Ham},0}(T^\ast\mathcal{Q})\simeq (\hat{\mathfrak{s}}/\mathbb{R})\bowtie \hat{\mathfrak{n}}\). This paper is nice to read. The various cumbersome constructions that are presented in the text are given in details, which is particularly helpful for any potential reader.