Splittings and \(Cr\)-structures for manifolds with nonpositive sectional curvature (Q5944879): Difference between revisions

Throughout the paper, \(M\) is a compact Riemannian manifold of nonpositive curvature and \(A\) -- an abelian structure on \(M\). Roughly speaking, such an \(A\) consists of a set of deck transformations of the universal covering \(\tilde{M}\) of \(M\) and of a, parametrized by these transformations, family of closed subsets of \(\tilde{M}\) which cover \(\tilde{M}\) and satisfy (among the other, more technical conditions) the following: Any two deck transformations of \(A\) commute whenever the underlying sets overlap. The authors show that if \(M\) admits an abelian structure, then (1) \(M\) admits a \(Cr\)-structure in the sense of \textit{S. Buyalo} [{Collapsing manifolds of non-positive curvature}, Leningr. Math. J. {1}, 1135--1155 and 1371--1399 (1990; Zbl 0735.53033 and Zbl 0735.53034)], (2) \(M\) has vanishing minimal volume. Moreover, they prove that any \(M\) with nonpositive curvature \(K\), \(-1\leq K\leq 0\), and injectivity radii bounded from above by a universal constant (which depends on \(n = \dim M\) only) admits an abelian structure. They obtain also a number of interesting corollaries which follow from comparison of the above results. Reviewer's remark: Note that the notion of \(Cr\)-structure does not coincide with that of CR-structure which exists in literature too.