Topology of symplectic torus actions with symplectic orbits (Q2430520)

The paper presents an account of recent results on the classification of compact manifolds with symplectic toral actions whose principal orbits are symplectic. These actions are never Hamiltonian. Although the paper is essentially based on a paper by the second author [\textit{A. Pelayo}, ``Symplectic actions of 2-tori on 4-manifolds'', Mem. Am. Math. Soc. 959, i--vii, 1--81 (2010; Zbl 1189.53081)], it presents some new results, the most notable one being the following: Theorem 1.1. If \(M\) is a compact, connected symplectic manifold on which a torus \(T\) acts effectively with symplectic orbits, then the orbit space \(M/T\) is a very good orbifold with Betti number \(b_1(M/T) = b_1(M)- \dim T.\) The theorem can be considered a symplectic version of the now classical result of \textit{F. C. Kirwan} on the Betti numbers of symplectic quotients in the case of Hamiltonian actions (cf. [Cohomology of quotients in symplectic and algebraic geometry. Princeton, New Jersey: Princeton University Press (1984; Zbl 0553.14020)]). The paper is divided into five sections. The first two are very short and of purely introductory character. The third is devoted to models for symplectic torus actions with symplectic principal orbits and reviews Chapters 2--7 of the publication of the second author mentioned above. There is one new result on principal toral bundles with flat infinitesimal connections (Theorem 3.3). The subject of Section 4 is the topology of the orbit space in which more detailed results on local models for actions permit to draw interesting conclusions on the topology of the orbit space. Moreover, the authors prove properties from which the main result is an easy corollary. The authors pay particular attention to the case when the orbit space is 2-dimensional, i.e., an orbisurface (Section 4.1). This part is the longest, very technical and most detailed one. The final section is concerned with the classification of the codimension-two case in which much more can be said; e.g., Theorem 5.1 gives a characterization of isomorphic models.