Locally constrained curvature flows and geometric inequalities in hyperbolic space (Q2122069): Difference between revisions

The locally constrained curvature flow of hypersurfaces in hyperbolic space (introduced by \textit{S. Brendle} et al. [``An inverse curvature type hypersurface flow in \({\mathbb{H}}^{n+1}\)'', Preprint]) was introduced as follows. The hyperbolic space \(\mathbf H^{n+1}\), is regarded as the warped product \(\mathbf H^{n+1}=[0,\infty)\times \mathbf {S}^{n}\) equipped with the metric \(\bar {g}=dr^2+\lambda(r)^2 g_{\mathbf S^n}\), where \(\lambda(r)=\sinh(r)\). \newline A hypersurface \(M\subset \mathbf {H}^{n+1}\) is called star-shaped if its support function \(u=\langle \lambda(r)\partial_{r},\nu\rangle\) is positive for every point on \(M\). They defined the locally constrained inverse curvature type flow \(X: M\times [0,T)\rightarrow \mathbf {H}^{n+1}\) satisfying \[\frac{\partial}{\partial t}X(x,t)=\left(\lambda'(r)\frac{E_{m-1}(\kappa)}{E_{m}(\kappa)}-u\right)\nu(x,t),m=1,\dots,n,X(\cdot,0)=X_0(\cdot),\] where \(\kappa=(\kappa_1,\dots,\kappa_n)\) are the principal curvatures of \(M_t=X(M,t)\). A smooth bounded domain \(\Omega\) in \(\mathbf {H}^{n+1}\) is called \(h\)-convex if the principal curvatures of the boundary \(\partial \Omega\) satisfy \(\kappa_{i}\geq 1\) for all \(i=1,\dots,n\). They prove that if the initial manifold \(M_0\) is a smooth, closed, and \(h\)-convex, then the above flow has a unique smooth solution for all time \(t\in[0,\infty)\) and it converges smoothly and exponentially to a geodesic sphere of radius \(r_\infty\) determined by the \(k\)th quermassintegral \(W_k\). \textit{B. Andrews} et al. [J. Eur. Math. Soc. (JEMS) 23, No. 7, 2467--2509 (2021; Zbl 1482.53116)] introduced a family of new quermassintegrals \(\tilde{W}_k\) and obtained a variational formula using the shifted principal curvatures. The authors of this paper introduce a new locally constrained curvature flow by replacing the curvature \(\kappa\) by the shifted principal curvatures. Then they provide a similar result for this case along with many applications, for example, the higher-order result of \textit{S. Brendle} et al. [Commun. Pure Appl. Math. 69, No. 1, 124--144 (2016; Zbl 1331.53078)]. Also they provide a different proof of an inequality by \textit{B. Andrews} and \textit{Y. Wei} [Geom. Funct. Anal. 28, No. 5, 1183--1208 (2018; Zbl 1401.53047)] using this new quermassintegrals.