Mean curvature flow with bounded Gauss image (Q545504)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Mean curvature flow with bounded Gauss image
scientific article

    Statements

    Mean curvature flow with bounded Gauss image (English)
    0 references
    22 June 2011
    0 references
    The author studies the mean curvature flow of a space-like submanifold \(M^{m}\) with codimension \(n\) in a pseudo-Riemannian manifold \(\mathbb{R}^{m+n}_{n}\), where \(\mathbb{R}^{m+n}_{n}\) denotes the Euclidean space \(\mathbb{R}^{m+n}\) together with the bilinear form given by \(\langle x,y \rangle = \sum_{i=1}^{m} x_{i} y_{i} - \sum_{\alpha=m+1}^{m+n} x_{\alpha} y_{\alpha}\). Let \(G^{n}_{m,n}\) denote the collection of all space-like \(m\)-dimensional subspaces of \(\mathbb{R}^{m+n}\). The group \(G = SO(m,n)\) acts transitively on \(G^{n}_{m,n}\) and the stability group of the space-like m-plane \(P = \mathrm{span}\{e_{1}, \dots , e_{m}\}\) is the maximal compact subgroup \(K = SO(m) \times SO(n)\). The space \(G^{n}_{m,n} = G/K\) is therefore a complete, simply connected Riemannian symmetric space of nonpositive sectional curvature. In particular there is a unique geodesic joining any two points of \(G^{n}_{m,n}\) and metric balls in \(G^{n}_{m,n}\) are convex. The Gauss map \(\gamma : M^{m} \rightarrow G^{n}_{m,n}\) assigns to each point \(m\) of \(M^{m}\) the parallel translate of \(T_{m}M\) to the origin of \(\mathbb{R}^{m+n}_{n}\). Let \(\{M_{t} \}\) denote the evolution of \(M\) by the mean curvature flow. The manifolds \(\{M_{t}\}\) remain space-like for all t for which the flow is defined. If \(\gamma(t)\) denotes the Gauss map of \(M_{t}\), then \(\gamma(t)\) satisfies the equation \(\frac{d\gamma}{dt} = \tau(\gamma(t))\), where \(\tau(\gamma(t))\) denotes the tension field of \(\gamma(t)\). The main result is the following Theorem~6.1. Let \(F : M^{m} \rightarrow \mathbb{R}^{m+n}\) be a complete, \(m\)-dimensional space-like submanifold with bounded Gauss image in \(G^{n}_{m,n}\) and bounded curvature. The mean curvature and its covariant derivatives are uniformly bounded. Then the mean curvature flow equations have a long time solution. A key step in the proof is to show that if the curvature of \(M^{m}\) is uniformly bounded, then for short time the Gauss images of the submanifolds \(\{M_{t} \}\) remain in any bounded metric ball of \(G^{n}_{m,n}\) that contains the Gauss image of \(M^{m} = M_{0}\). This step follows by applying a maximum principle that results from the bounded curvature hypothesis. The hypothesis that the Gauss image of \(M^{m}\) lies in a bounded metric ball implies that the mean curvature equations are uniformly parabolic.
    0 references
    mean curvature flow
    0 references
    evolution equation
    0 references
    complete space-like submanifolds in pseudo-Euclidean space
    0 references
    bounded Gauss image
    0 references
    confinable property
    0 references
    long time existence
    0 references
    0 references
    0 references
    0 references
    0 references

    Identifiers