Maximal tori in the symplectomorphism groups of Hirzebruch surfaces. (Q1565811): Difference between revisions

An effective Hamiltonian action of \(T^k\) on a symplectic manifold \(M\) gives a \(T^k\) embedded in \(\text{Ham}(M,\omega)\), the group of Hamiltonian diffeomorphisms of \(M\). A torus in \(\text{Ham}(M,\omega)\) is maximal if it isn't contained in a torus of strictly larger dimension. Two tori \(T_1\) and \(T_2\) are conjugate in \(\text{Ham}(M,\omega)\) if there is some \(g \in \text{Ham}(M,\omega)\) with \(g T_1 g^{-1} = T_2\). The main result of this paper is the determination of the number of conjugacy classes of maximal tori in \(\text{Ham}(S^2 \times S^2, \omega_{a,b})\) and in \(\text{Ham}(\widetilde{\mathbb CP^2}, \widetilde\omega_{l,e})\). Here, \(\widetilde{\mathbb CP^2}\) is \(\mathbb CP^2\) blown up at a point and the subscripts on the symplectic forms denote certain specific constructions of them. For instance, for \(\text{Ham}(S^2 \times S^2, \omega_{a,b})\), the number of conjugacy classes is shown to be \([a/b]\). A key point in the proof is that a Delzant polygon (associated to a toric action --- that is, an action of a torus of half the dimension of the manifold) with four edges is a Hirzebruch trapezoid. Then, since the number of edges of a Delzant polygon is \(2 + b_2(M)\), the spaces under consideration have four-edge polygons. Combinatorics then reduces the problem to the Hirzebruch case.