Uniqueness of the measure of maximal entropy for the standard map (Q2030899)

The standard map is a paradigmatic example of a complicated dynamical system generated by a simple formula. Considering \(\mathbb{T}^2\) with coordinates \((x,y)\in[0,1)^2\), for each \(k\in\mathbb{R}\), the standard map \( f_k:\mathbb{T}^2\to \mathbb{T}^2\) is defined as \[ f_k :(x,y)\mapsto (2x-y+k\sin(2\pi x),x). \] It was introduced by \textit{B. V Chirikov} [Phys. Rep. 52, No. 5, 263--379 (1979; doi:10.1016/0370-1573(79)90023-1)] and is related with many physical problems. The map \(f_k\) preserves the Lebesgue measure and for large values of \(k\) has positive topological entropy. It is still unknown if the metric entropy of the Lebesgue measure is positive. The author gives some account on the progress on this quite important problem in the beginning of his introduction. The goal of the paper is to describe some fundamental ergodic properties of \(f_k\) and its perturbations. It is known (due to a theorem of \textit{S. E. Newhouse} [Ann. Math. (2) 129, No. 2, 215--235 (1989; Zbl 0676.58039)]) that \(f_k\) has at least one measure of maximal entropy. The main result of the paper shows that for \(k\) large enough, for every sufficiently small \(C^2\) perturbation of \(f_k\), if there exists a measure of maximal entropy then this measure is unique. In particular, \(C^{\infty}\) maps \(C^2\) close to \(f_k\) have a unique measure of maximal entropy. In the \(C^{\infty}\) case, using a recent result of \textit{D. Burguet} [J. Eur. Math. Soc. (JEMS) 22, No. 2, 413--454 (2020; Zbl 1437.37052)] it is shown that the hyperbolic periodic points with a lower bound on the Lyapunov exponent equidistribute towards the unique measure of maximal entropy. A quite interesting question which remains open, and to which this paper also contributes is: how ``large'' is this measure? Theorem C of the present paper gives the following answer: for every \(\varepsilon>0\), there exists \(k_0\) such that for any \(k\) larger than \(k_0\) if \(g\) is \(C^2\) close to \(f_k\) and if \(\mu\) is its measure of maximal entropy then the Hausdorff dimension of \(\mu\) is larger than \(2-\varepsilon\). Estimations for the size of the support of \(\mu\) are also given. The proof of the main result relies on techniques previous developed by the author in his proof of the stable ergodicity of \textit{P. Berger} and \textit{P. D. Carrasco}'s example [Commun. Math. Phys. 329, No. 1, 239--262 (2014; Zbl 1350.37032)]), which is a partially hyperbolic skew-product derived from the standard map. Using the work of \textit{J. Buzzi} et al. [Measures of maximal entropy for surface diffeomorphisms, Preprint, \url{arXiv:1811.02240}], the author reduces the proof to a careful study of the geometry of stable and unstable Pesin manifolds of hyperbolic measures. The paper is very well written, including a detailed description of the strategy of the proof.