Branch points and non-density for finite-bending isometric immersions of hyperbolic surfaces
Gilad Derfner, Raz Kupferman, Cy Maor
Abstract
The space of W2,2-isometric immersions of a surface into R3 arises naturally in the variational theory of thin elastic sheets: it is precisely the finite-bending class, where the bending energy --- the L2-norm of the second fundamental form --- is finite. For sheets with negative Gaussian curvature, previous work has identified branch points, where "too many" asymptotic directions meet --- or, equivalently, where the index of the Gauss map is not -1 --- as a potentially important mechanism in shape selection and pattern formation. Such branch points are precluded for C2-isometric immersion. We show that this index-based notion of branch points extends to the full finite-bending class: Namely, for every W2,2 isometric immersion of a negatively-curved surface, the index of the Gauss map is well-defined at every point, and the set of branch points is discrete. We further show that the index is stable under W2,2-convergence, and thus, an isometric immersion with branch points cannot be approximated by C2-isometric immersions. Conversely, we show that every negatively-curved metric locally admits W2,2-isometric immersions (in fact, C1,1) with branch points of arbitrary order. Consequently, C2-isometric immersions are, in general, not dense among finite-bending ones, in stark contrast with the flat and positively curved cases.
Create a lesson
Related papers
Maximal symmetry rank and almost non-negative curvature in low dimensions
Samuel Bartel
Examples of Z/2-Harmonic 1-Forms
Jiahuang Chen, Siqi He
Collapsed Finite Time Singularities of the Kähler-Ricci Flow on Complex Surfaces are of Type I
Tongxin Xu, Zhenlei Zhang
Uniqueness of embedded minimal Lagrangian tori in CP2
Yong Luo, Hui Ma, Jiabin Yin
Spectral properties for critical metrics of the volume functional
Rafael Diógenes, Jaciane Gonçalves, Ernani Ribeiro
Morse resolution of mean curvature flows with cylindrical singularities
Richard H. Bamler, Felix Schulze, Lu Wang