Rigidity and non-rigidity results on the sphere (Q871739)

As a consequence of the positive mass theorems, the following rigidity theorem for the unit ball in \(\mathbb{R}^n\) has been established: Let \((M^n,g)\) be an \(n\)-dimensional compact Riemannian manifold with boundary and scalar curvature satisfying \(R\geq 0\). Assume that the boundary is isometric to the standard sphere \(\mathbb{S}^{n-1}\) and that it has mean curvature \(n-1\). Then \((M^n,g)\) is isometric to the unit ball in \(\mathbb{R}^n\) (it is assumed that \((M^n,g)\) is spin if \(n>7\)). Similar results can be given in the hyperbolic space. In this paper the authors give some rigidity and non-rigidity results on the standard unit sphere \(\mathbb{S}^n\). In this context, the following rigidity conjecture was posed by \textit{M. Min-Oo} [Scalar curvature rigidity of the hemisphere, unpublished work (1995)]: Let \((M^n,g)\) be an \(n\)-dimensional compact Riemannian manifold with boundary and scalar curvature satisfying \(R\geq n(n-1)\). Assume that the boundary is isometric to the standard sphere \(S^{n-1}\) and that it is totally geodesic. Then \((M^n,g)\) is isometric to the hemisphere \(\mathbb{S}^n_+\). In the first section of the paper, the authors show (theorem 2.1) that it is possible to conformally deform the standard sphere metric without decreasing the scalar curvature and with the deformation supported in any given open geodesic ball of radius \(r>\frac{\pi}{2}\); therefore the corresponding rigidity for geodesic balls of radius \(r>\frac{\pi}{2}\) does not hold even for conformal deformations. They also construct a rotationally symmetric metric on \(\mathbb{S}^ n\) such that its sectional curvature is at least \(1\) and strictly greater than 1 somewhere, and near the north pole and south pole the metric is identical to the standard one. However, they verify the conjecture for the hemisphere among conformal deformations of the standard metric, even if the boundary is not totally geodesic. They also confirm the conjecture for Einstein metrics. Theorem 2.1 can also be seen in the light of a result due to \textit{J. Corvino} [Commun. Math. Phys. 214, No.~1, 137--189 (2000; Zbl 1031.53064)] which says that for any non-static metric one can always find compact deformations with the scalar curvature going in either direction. On the other hand, Euclidean and hyperbolic metrics are static and one cannot have any compact deformations of them without decreasing the scalar curvature somewhere. Now, the standard metric on the sphere is also static, but theorem 2.1 proves that it is possible to deform it without decreasing the scalar curvature on any geodesic ball of radius \(r>\frac{\pi}{2}\)