Convexity of Hypersurfaces in Spherical Spaces

A spherical set is called convex if for every pair of its points there is at least one minimal geodesic segment that joins these points and lies in the set. We prove that for n >= 3 a complete locally-convex (topological) immersion of a connected (n-1)-manifold into the n-sphere is a surjection onto the boundary of a convex set.

💡 Research Summary

The paper investigates convexity of hypersurfaces embedded in spherical spaces. After introducing a precise definition of convexity on the sphere—namely, a set is convex if for any two points there exists at least one minimal geodesic segment joining them that stays entirely inside the set—the author focuses on complete, locally convex topological immersions of connected (n‑1)-dimensional manifolds into the n‑sphere Sⁿ, with n ≥ 3. Local convexity means that around each point the immersion lies on one side of a supporting great‑sphere, equivalently that the tangent space is contained in an exterior hemisphere. Completeness is understood with respect to the intrinsic geodesic metric, guaranteeing that the immersed manifold is geodesically complete.

The main theorem states that under these hypotheses the image of the immersion coincides with the boundary of a convex subset K of Sⁿ; in symbols, f(M)=∂K. In other words, a globally defined convex body in the sphere is forced upon any complete locally convex hypersurface when the ambient dimension is three or higher. The result is a spherical analogue of the classical Hadamard‑Stoker theorem for Euclidean space and extends Alexandrov’s global convexity results to the curved setting of Sⁿ.

The proof proceeds in two major stages. First, the author shows that local convexity yields a family of supporting hemispheres {Hₚ}ₚ∈M, each containing the image of a small neighborhood of p and touching the immersion at f(p). By exploiting the completeness of M and the fact that n ≥ 3, the author proves that the intersection of all these hemispheres is non‑empty. This intersection defines a closed convex set K in Sⁿ. The second stage establishes that the boundary of K is exactly the image of M. The argument uses the fact that any point of ∂K can be approached by points of f(M) along minimal geodesics, and completeness rules out missing boundary points. Consequently, f(M) is surjective onto ∂K.

The paper also explains why the dimension restriction n ≥ 3 is essential. In the two‑dimensional case (S²), a locally convex closed curve need not be the boundary of a convex region; simple counter‑examples such as a small latitude circle demonstrate that the theorem fails when the codimension is one.

Throughout, the author carefully relates the new spherical result to earlier work. The Euclidean Hadamard‑Stoker theorem requires global convexity to be inferred from local convexity and completeness; however, the spherical setting introduces additional complications because each pair of points may have two distinct minimal geodesics. By combining tools from spherical geometry (great‑sphere support, geodesic distance) with algebraic topology (connectedness, covering arguments), the paper overcomes these obstacles.

Potential applications are discussed. The theorem provides a theoretical foundation for problems in spherical optimization, geometric modeling on the globe, and the study of convex bodies in spaces of constant positive curvature. For instance, in computer graphics or geodesic dome design, ensuring that a locally convex surface automatically yields a globally convex shape simplifies verification procedures. The author suggests further research directions, such as weakening the completeness assumption, exploring non‑compact immersions, or extending the result to other constant curvature manifolds.

In summary, the work delivers a rigorous and elegant characterization of convex hypersurfaces in spherical spaces, showing that for dimensions three and higher, local convexity together with completeness forces the immersion to be the entire boundary of a spherical convex body. This bridges a gap between Euclidean convexity theory and its spherical counterpart, and opens new avenues for both theoretical investigation and practical application.