Marked length spectrum rigidity for Anosov surfaces (Q7013928)

Let \(\Sigma\) be a smooth, closed, connected, oriented surface of genus at least 2. A Riemannian metric \(g\) on \(\Sigma\) is called {Anosov} if its geodesic flow on the unit tangent bundle \(S\Sigma\) is Anosov. While every negatively curved metric is Anosov, the class of Anosov metrics is larger. The paper under review addresses the {``marked length spectrum conjecture''}, also known as the Burns-Katok conjecture in the setting of negatively curved metrics, for Anosov metrics.\N\NHere is the main theorem.\N\NTheorem. If \(g_1\) and \(g_2\) are two Anosov metrics on \(\Sigma\) with the same marked length spectrum, then there exists a diffeomorphism \(\phi: \Sigma \to \Sigma\), isotopic to the identity, such that \(\phi^{*}g_1 = g_2\).\N\NIn combination with earlier results, the authors derive several interesting corollaries. For instance, the main theorem implies that the set of self-conjugacies of an Anosov geodesic flow on a surface is given (after taking the quotient by the flow) by the finite group of isometries of the metric. This generalizes the result of \textit{C. B. Croke} [Comment. Math. Helv. 65, No. 1, 150--169 (1990; Zbl 0704.53035)].\N\NThe proof strategy is as follows. Given two Anosov metrics \(g_1, g_2\) with the same marked length spectrum, each metric \(g_i\) determines a class of conformal structures \([J_{g_i}]\), hence a point in the Teichmüller space of \(\Sigma\). The key step is to show that \(g_1\) and \(g_2\) yield the same point in the Teichmüller space of \(\Sigma\) (see {Proposition 3.7}). Once this is established, a result of \textit{A. Katok} [Ergodic Theory Dyn. Syst. 8, 139--152 (1988; Zbl 0668.58042)] yields the main theorem.\N\NTo reach this conclusion, the authors employ sophisticated ideas, such as {fiberwise holomorphic invariant distributions}. These are used to show that the marked length spectrum determines the period matrix of the underlying Riemann surface. By the Torelli theorem, this in turn determines the point in the moduli space of \(\Sigma\) (see {Propositions 3.1} and {Proposition 3.6}). This constitutes the most technically challenging part of the paper. To strengthen the result from the moduli space to the Teichmüller space, the authors prove a further interesting theorem in Teichmüller theory: {any two distinct points in Teichmüller space lie in different mapping class group orbits after passing to a suitable finite cover}.\N\NThis is a very interesting and well-written paper. The exposition is particularly clear in the first section, which provides an excellent introduction to the marked length spectrum conjecture. The paper will be of broad interest to researchers working on topics related to the marked length spectrum and rigidity of Anosov flows.