Dynamics of nonlinear Klein-Gordon equations in low regularity on \(\mathbb{S}^2\) (Q6075020)

The authors consider the nonlinear Klein-Gordon model on \(S^{2}\), with a general polynomial interaction of a degree \(p \geq 3\), and a ``coupling constant'' being an arbitrary function \(g(x)\). The quantities under investigation are the \textit{super-actions} \(\mathcal{E}_{l}(\Phi(t),\dot{\Phi}(t))\), i.e., certain well-behaved conserved quantities corresponding to the free theory (parametrized by the degree \(l\) of the spherical harmonics), and their (almost) conservation in the full nonlinear theory. For short time-scales the super-actions, as the authors explain in the text, are almost preserved. The main purpose of the paper is to investigate the long-time behaviour and the preservation of the conserved quantities for such time scales. The paper is a continuation of a series of previous papers, and, in particular, the one where such investigation had been carried out for Zoll manifolds (i.e., manifolds for which the geodesic flow is periodic), under certain regularity conditions. The main result of the paper is an improved estimate for super-actions \(\mathcal{E}_{l}(\Phi(t),\dot{\Phi}(t))\) without imposing the regularity condition, in particular, for smaller \(l\). The proof, presented in detail, is technically quite involved as it is based on the infinite-dimensional version of the Birkhoff normal form (proved by the first two authors in [J. Éc. Polytech., Math. 9, 681--745 (2022; Zbl 1490.35392)]), which is substantially more complex than the finite-dimensional Birkhoff's theorem (e.g., new concepts such as ``tame modulus property'' must be introduced). The latter is an essential method in the study of finite-dimensional Hamiltonian systems, and is a statement on the existence of a real analytic symplectic transformation which reduces the Hamiltonian, in a neighborhood of an elliptic stationary point, to a normal form consisting of small perturbations of an integrable system. There are also many interesting comments and open problems discussed by the authors.