A rigidity theorem in Alexandrov spaces with lower curvature bound (Q425134)

From MaRDI portal
scientific article
Language Label Description Also known as
English
A rigidity theorem in Alexandrov spaces with lower curvature bound
scientific article

    Statements

    A rigidity theorem in Alexandrov spaces with lower curvature bound (English)
    0 references
    0 references
    7 June 2012
    0 references
    Comparison theorems are indispensable in the investigation of the geometry of non-smooth metric spaces with curvature bounds. Indeed, the very definition of a space \(X\) with curvature bounded from below by \(\kappa\) has the form of an inequality of distances between points in \(X\) and the corresponding points in a model space \(M_{\kappa}^2\), see [\textit{Yu. Burago, M. Gromov} and \textit{G. Perel'man}, Russ. Math. Surv. 47, No. 2, 1--58 (1992; Zbl 0802.53018)] and [\textit{J. Cheeger} and \textit{D. G. Ebin}, Comparison theorems in Riemannian geometry. North-Holland Mathematical Library. Vol. 9. Amsterdam-Oxford: North-Holland Publishing Company; New York: American Elsevier Publishing Company, Inc. VIII (1975; Zbl 0309.53035)]. To define a lower curvature bound for non-geodesic spaces, one can use comparison angles \(\widetilde \angle_\kappa(x;y,z)\): Precisely, a metric space \(X\) has curvature \(\geq \kappa\) if the inequality \[ \widetilde \angle_\kappa(x;y,z)+\widetilde \angle_\kappa(x;z,w)+\widetilde \angle_\kappa(x;w,y)\leq 2\pi \tag{1} \] holds for any quadruple of distinct points of admissible size. (The size requirement is relevant only if \(\kappa>0\) and means that the maximum of perimeters of all three point subsets of the quadrangle is less than \(2\pi/\sqrt{\kappa}\).) If, in addition, \(X\) is a complete length space, it is called an Alexandrov space of curvature \(\geq\kappa\). It is natural to consider the equality case in (1). One expects such a quadruple of points to exhibit some rigidity, and the following result (Theorem C in the paper) shows this is indeed the case. Suppose that equality holds in (1) and \(x\) does not lie on any geodesic through two of the points \(y,z,w\). Then \(y,z,w\) span the unique closed triangular region \(D\) which contains \(x\) and is isometric to the triangular region in \(M_\kappa^2\) with vertices \(\tilde y,\tilde z,\tilde w\). Theorem C is in fact a very special case of the more general Theorems A and B that involve probability measures on \(X\). It is important to note that the space \(X\) is not assumed to be locally compact; thus, it can be infinite-dimensional. A corollary of Theorem C extends the rigidity part of Theorem 2.4 in [\textit{K. Grove} and \textit{S. Markvorsen}, J. Am. Math. Soc. 8, No. 1, 1--28 (1995; Zbl 0829.53033)] to infinite-dimensional spaces.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    lower curvature bound
    0 references
    Alexandrov space
    0 references
    rigidity
    0 references
    isometric embedding
    0 references
    probability measures
    0 references
    0 references
    0 references