A remark on unique ergodicity and the contact type condition (Q896043): Difference between revisions

It is well known that almost all regular energy level sets of the Hamiltonian systems on \(\mathbb R^{2n}\) have periodic orbits, but there are examples of level sets without periodic orbits. A natural question is how far from having periodic orbits a regular level set of a Hamiltonian \(H\) can be: for example to be uniquely ergodic (the flow is uniquely ergodic if it admits only one ergodic measure). It is proved that for a broad class of exact symplectic manifolds including \(\mathbb R^{2n}\), the Hamiltonian flow on a regular compact energy level of an autonomous Hamiltonian cannot be uniquely ergodic. The statement is a consequence of the Weinstein conjecture and the observation that a Hamiltonian structure with non-vanishing self-linking number must have contact type, by the contact type criterion given in [\textit{D. McDuff}, Ann. Inst. Fourier 37, No. 1, 107--133 (1987; Zbl 0572.58010)]. As an example, it is proved that certain types of exact twisted geodesic flows on closed surfaces cannot be uniquely ergodic.