Inverse mean curvature flow with outer obstacle
Kai Xu
TL;DR
This work develops a rigorous variational framework for the inverse mean curvature flow (IMCF) in bounded domains under an outer obstacle boundary condition, where flowing hypersurfaces stick tangentially to the boundary. It constructs and analyzes weak IMCF solutions using elliptic regularization, calibration, and energy methods, establishing existence, uniqueness (via maximality), and $C^{1,\alpha}$ regularity up to the obstacle. A key novelty is a robust outer-obstacle formulation that unifies sublevel-set and Dirichlet-energy perspectives and yields well-posed initial-value problems; this is complemented by Liouville-type results on the half-space and parabolic estimates near smooth obstacles, which underpin the blow-up analysis and regularity theory. The results provide a canonical way to obtain nontrivial maximal weak solutions in bounded domains and relate to the global behavior of IMCF through isoperimetric considerations, with potential implications for Hawking mass monotonicity and scalar curvature problems. Overall, the paper advances the analytic toolbox for weak IMCF with boundary constraints and clarifies the interaction between obstacle geometry, calibration, and parabolic regularity in the evolution.
Abstract
We develop a new boundary condition for the weak inverse mean curvature flow, which gives canonical and non-trivial solutions in bounded domains. Roughly speaking, the boundary of the domain serves as an outer obstacle, and the evolving hypersurfaces are assumed to stick tangentially to the boundary upon contact. In smooth bounded domains, we prove an existence and uniqueness theorem for weak solutions, and establish $C^{1,α}$ regularity of the level sets up to the obstacle. The proof combines various techniques, including elliptic regularization, blow-up analysis, and certain parabolic estimates. As an analytic application, we address the well-posedness problem for the usual weak inverse mean curvature flow, showing that the initial value problem always admits a unique maximal (or innermost) weak solution.
