Symplectic rigidity of fibers in cotangent bundles of open Riemann surfaces (Q6169566)

Following Weinstein's creed that every symplectic is Lagrangian, the study of Lagrangian submanifolds of a symplectic manifold \(M\) is of great interest. In the context where \(M\) is some cotangent bundle \(T^*N\), the so-called nearby Lagrangian conjecture states that every exact closed Lagrangian submanifold of \(T^*N\), where \(N\) is closed and connected, is Hamiltonian isotopic to the zero-section. This statement has only been proven in a few very specific case: \(S^1\) [folklore], \(S^2\) [\textit{R. Hind}, Geom. Funct. Anal. 14, No. 2, 303--318 (2004; Zbl 1066.53129)], \(\mathbb{R}P^2\) [\textit{R. Hind} et al., J. Symplectic Geom. 14, No. 1, 203--226 (2016; Zbl 1355.57026)], and \(\mathbb{T}^2\) [\textit{G. Dimitroglou Rizell} et al., Geom. Funct. Anal. 26, No. 5, 1297--1358 (2016; Zbl 1358.53089)]. However, a variant of the question may be asked when \(N\) is noncompact. In this case, the Lagrangian submanifolds \(L\) considered must agree with the zero-section outside some compact set. Likewise, Hamiltonian isotopies are required to be compactly supported. This variant was proven for the case \(Q=\mathbb{R}^2\) in [\textit{Y. Eliashberg} and \textit{L. Polterovich}, Ann. Math. (2) 144, No. 1, 61--76 (1996; Zbl 0872.57030)], but has also remained mostly open. The authors of the present paper explore a rigidity problem in the same vein to the noncompact nearby Lagrangian conjecture, although strictly perpendicular to it. More precisely, they prove that if \(N=\Sigma\) is an open Riemann surface and \(L\subseteq T^*\Sigma\) is a Lagrangian submanifold diffeomorphic to \(\mathbb{R}^2\) and which corresponds to a fiber \(F_x\) of \(T^*\Sigma\) outside some compact set, then \(L\) is Hamiltonian isotopic to \(F_x\) (through a compactly supported Hamiltonian isotopy). Note that this result is equivalent to the nearby Lagrangian conjecture when \(\Sigma=\mathbb{R}^2\), so that it follows from the previously cited paper of Eliashberg and Polterovich in that case. This result also implies some rigidity for closed Riemann surfaces. Indeed, it follows as a corollary that if \(\Sigma\) is any Riemann surface and \(L\) is as above and additionally disjoint from another fiber \(F_y\), \(y\neq x\), then \(L\) is Hamiltonian isotopic to \(F_x\) (through a Hamiltonian isotopy compactly supported in \(T^*\Sigma-F_y\)). Being disjoint from a fiber is a necessary assumption here, since the Dehn twist of a fiber \(F_x\) of \(T^*S^2\) respects all other assumptions, but is known to not be Hamiltonian isotopic to any fiber. This corollary is remarkable in that it represents the first rigidity result for exact Lagrangian submanifolds in \(T^*\Sigma\) when \(\Sigma\) is of genus more than 1, compare the above list of cases where the nearby Lagrangian conjecture is proven. The paper itself is divided into two main sections, following the introduction: Section~2, where the proper setup of the paper is explained in details, and Section~3, where the main theorem is proven. The proof of the main theorem is morally inspired by the approach of \textit{Y. Eliashberg} and \textit{L. Polterovich} [Ann. Math. (2) 144, No. 1, 61--76 (1996; Zbl 0872.57030)], but uses punctured holomorphic curves instead of holomorphic curves with boundary in totally real submanifolds. It is the additional plasticity brought by the punctured holomorphic curves which allows the authors to go from \(\mathbb{R}^2\) to a general open Riemann surfaces. More precisely, the proof goes as follows. \begin{itemize} \item[1.] In Section~3.2, the authors construct a convenient model for \(\Sigma\) along with a special Riemannian metric \(g_\eta\) on it. \item[2.] In Section~3.3, the authors introduce a hypersurface ``with holes'' called \(W'\), which contains \(L\) and is diffeomorphic to \(\mathbb{R}^3\) with two solid cylinders removed. They also construct from \(g_\eta\) an almost complex structure \(J\) defined on most of \(T^*\Sigma\). \item[3.] In Section~3.4, the authors ``fill'' the holes of \(W'\) to get a hypersurface \(Q\) diffeomorphic to \(\mathbb{R}^3\). This is the technical heart of the proof and relies on techniques coming from symplectic field theory, applied to the special \(J\) that they constructed. \item[4.] In Section~3.5, the authors conclude the proof by deforming \(Q\) to a better suited hypersurface \(\widetilde{Q}\). It is from \(\widetilde{Q}\) that they construct the desired Hamiltonian isotopy from \(L\) to a cotangent fiber. \end{itemize}