Rigidity and nonexistence of complete spacelike hypersurfaces in the steady state space

Weiller F.C. Barboza 1 , Henrique F. de Lima 1,∗ and Marco Antonio L. Velásquez 1 · 2025

Read Rigidity and nonexistence of complete spacelike hypersurfaces in the steady state space by Weiller F.C. Barboza 1 , Henrique F. de Lima 1,∗ and Marco Antonio L. Velásquez 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-11614.txt.

Introduction

Barboza and co-authors prove strict results about curved sheets in a model cosmos. The setting is the steady state space, a region of de Sitter space used in cosmology. Inside it sit spacelike hypersurfaces, sheets that cut across time. The team asks what shapes such sheets can take when they stretch on without edge.

Under set limits on how the sheets bend, the only answer is a flat sheet, a spacelike hyperplane. Other shapes are ruled out in the cases they treat. The proof leans on a broad form of a classic maximum principle for curved spaces. Such rigidity results help map which geometries a steady state cosmos can host.

Catalog record

CREATORWeiller F.C. Barboza 1 , Henrique F. de Lima 1,∗ and Marco Antonio L. Velásquez 1
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE33 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-11614

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Rigidity and nonexistence of complete spacelike hypersurfaces in the steady state space Thanks: ∗ Corresponding author. Weiller F.C. Barboza 1 , Henrique F. de Lima 1,∗ and Marco Antonio L. Velásquez 1 Abstract . We study complete spacelike hypersurfaces immersed in an open region of the de Sitter space 𝕊 1 n + 1 \mathbb{S}^{n+1}_{1} which is known as the steady state space ℋ n + 1 \mathcal{H}^{n+1} . In this setting, under suitable constraints on the behavior of the higher order mean curvatures of these hypersurfaces, we prove that they must be spacelike hyperplanes of ℋ n + 1 \mathcal{H}^{n+1} . Nonexistence results concerning these spacelike hypersurfaces are also given. Our approach is based on a suitable extension of the Omori-Yau’s generalized maximum principle due to Alías, Impera and Rigoli in [ 5 ] . 2010 Mathematics Subject Classification Primary 53C42; Secondary 53A10, 53C20, 53C50. 1. Introduction Let \mathbb{L}^{n+2} denote the (n+2) -dimensional Lorentz-Minkowski space ( n\geq 2 ), that is, the real vector space \mathbb{R}^{n+2} endowed with the Lorentz metric defined by \left\langle v,w\right\rangle={\displaystyle\sum\limits_{i=1}^{n+1}}v_{i}w_{i}-v_{n+2}w_{n+2}, for all v,w\in\mathbb{R}^{n+2} . We define the (n+1) -dimensional de Sitter space \mathbb{S}_{1}^{n+1} as the following hyperquadric of \mathbb{L}^{n+2} \mathbb{S}_{1}^{n+1}=\left\{p\in\mathbb{L}^{n+2}:\left\langle p,p\right\rangle=1\right\}. The induced metric from \left\langle,\right\rangle makes \mathbb{S}_{1}^{n+1} a Lorentz manifold with constant sectional curvature one. Moreover, for all p\in\mathbb{S}_{1}^{n+1} , we have T_{p}\left(\mathbb{S}_{1}^{n+1}\right)=\left\{v\in\mathbb{L}^{n+2}:\left\langle v,p\right\rangle=0\right\}. Let a\in\mathbb{L}^{n+2}\setminus\{0\} be a past-pointing null vector, that is, \left\langle a,a\right\rangle=0 and \left\langle a,e_{n+2}\right\rangle>0 , where e_{n+2}=(0,\dots,0,1) . Then, the open region of the de Sitter space \mathbb{S}_{1}^{n+1} , given by \mathcal{H}^{n+1}=\left\{p\in\mathbb{S}_{1}^{n+1}:\left\langle p,a\right\rangle>0\right\} is the so-called steady state space . The importance of considering \mathcal{H}^{n+1} comes from the fact that, in Cosmology, \mathcal{H}^{4} is the steady state model of the universe proposed by Bondi-Gold [ 9 ] and Hoyle [ 14 ] , when looking for a model of the universe which looks the same not only at all points and in all directions (that is, spatially isotropic and homogeneous), but also at all times. For more details, we recommend for the readers to see Section 5.2 of [ 13 ] or Section 14.8 of [ 18 ] . From a mathematical point of view, the interest in the study of spacelike hypersurfaces immersed in a Lorentzian space is motivated by their nice Bernstein-type properties. In this direction, several authors have approached the problem of to characterizing spacelike hyperplanes of \mathcal{H}^{n+1} , which are totally umbilical spacelike hypersurfaces isometric to the Euclidean space \mathbb{R}^{n} and give a complete foliation of \mathcal{H}^{n+1} . We refer to readers, for instance, the works [ 1 , 7 , 10 , 11 , 16 ] . Proceeding with this picture, our purpose in this article is to apply a suitable extension of the generalized maximum principle of Omori [ 17 ] and Yau [ 19 ] due to Alías, Impera and Rigoli in [ 5 ] (see Lemma 5 ) in order to establish rigidity results for complete spacelike hypersurfaces of \mathcal{H}^{n+1} . For this, we will assume certain appropriate constraints on the behavior of the higher order mean curvatures. In this setting, We also obtain some nonexistence results concerning these spacelike hypersurfaces. This paper is organized in the following manner: initially, in Section 2 , we recall some basic facts concerning the higher order mean curvatures, the Newton transformations and the linearized operators naturally attached to a spacelike hypersurface immersed in \mathcal{H}^{n+1} . In the last part of this same section, we also quote some key lemmas that will be essential to prove our main results. Afterwards, in Section 3 we establish our rigidity and nonexistence results for complete spacelike hypersurfaces of \mathcal{H}^{n+1} (see Theorems 1 , 2 and 3 , and Corollaries 1 , 2 and 3 ). Finally, in Section 4 we obtain further rigidity results considering that the height and angle functions of the spacelike hypersurfaces are linearly related (see Theorems 4 and 5 , and Corollary 4 ). 2. Preliminaries and key lemmas Along this article, we will deal with connected spacelike hypersurfaces \psi:\Sigma\rightarrow\mathcal{H}^{n+1} with a future-pointing normal unitary timelike vector field N . Let us denote by A:\mathfrak{X}(\Sigma)\rightarrow\mathfrak{X}(\Sigma) the Weingarten endomorphism of \Sigma^{n} with respect to N . We recall that at each point p\in\Sigma^{n}, the Weingarten operator A restricts to a self-adjoint linear map A_{p}:T_{p}\Sigma\rightarrow T_{p}\Sigma. For 0\leq r\leq n, let S_{r}(p) denote the r -th elementary symmetric function on the eigenvalues of A_{p}. Hence, one gets n smooths functions S_{r}:\Sigma\rightarrow\mathbb{R}, such that \det(tI-A)=\sum_{k=0}^{n}(-1)^{k}S_{k}t^{n-k}, where S_{0}=1 by convention. If p\in\Sigma^{n} and \{e_{k}\} is a basis of T_{p}\Sigma formed by eigenvectors of A_{p} , with corresponding eigenvalues \{\lambda_{k}\} , one immediately sees that S_{r}=\sigma_{r}(\lambda_{1},\ldots,\lambda_{n}), where \sigma_{r}\in\mathbb{R}[X_{1},\ldots,X_{n}] is the r -th elementary symmetric polynomial on the indeterminates X_{1},\ldots,X_{n} . In the sequel, with this setting, we define the r -th mean curvature H_{r} of \Sigma^{n} , 0\leq r\leq n , by (2.1) {n\choose r}H_{r}=(-1)^{r}S_{r}=\sigma_{r}(-\lambda_{1},\ldots,-\lambda_{n}). We observe that H_{0}=1 , while H_{1}=-\frac{1}{n}S_{1} is the usual mean curvature H of \Sigma^{n} . For 0\leq r\leq n , one defines the r -th Newton transformation P_{r} on \Sigma^{n} by setting P_{0}=I (the identity operator) and, for 1\leq r\leq n , via the recurrence relation (2.2) P_{r}={n\choose r}H_{r}I+AP_{r-1}. With a trivial induction, from ( 2.2 ) we verify that (2.3) P_{r}={n\choose r}H_{r}I+{n\choose r-1}H_{r-1}A+{n\choose r-2}H_{r-2}A^{2}+\cdots+A^{r}, so that Cayley-Hamilton theorem gives P_{n}=0 . Moreover, since P_{r} is a polynomial in A for every r , it is also self-adjoint and commutes with A . Therefore, all bases of T_{p}\Sigma diagonalizing A at p\in\Sigma^{n} also diagonalize all of the P_{r} at p . So, let \{e_{1},\ldots,e_{n}\} be an orthonormal frame on T_{p}\Sigma which diagonalizes A_{p} , A_{p}(e_{i})=\lambda_{i}(p)e_{i} , then from ( 2.3 ) we have that (2.4) (P_{r})_{p}e_{i}=(-1)^{r}\sum_{i_{1}<\ldots<i_{r},i_{j}\neq i}\lambda_{i_{1}}(p)\ldots\lambda_{i_{r}}(p)e_{i}. Moreover, it is not difficult to check that P_{r}e_{i}=(-1)^{r}S_{r}(A_{i})e_{i} and, consequently, we obtain the following lemma (see Lemma 2.1 of [ 8 ] ). Lemma 1 . With the above notations, the following formulas hold: (a) S_{r}(A_{i})=S_{r}-\lambda_{i}S_{r-1}(A_{i}) ; (b) {\rm tr}(P_{r})=(-1)^{r}\displaystyle\sum_{i=1}^{n}S_{r}(A_{i})=(-1)^{r}(n-r)S_{r}=c_{r}H_{r} ; (c) {\rm tr}(AP_{r})=(-1)^{r}\displaystyle\sum_{i=1}^{n}\lambda_{i}S_{r}(A_{i})=(-1)^{r}(r+1)S_{r+1}=-c_{r}H_{r+1} ; (d) {\rm tr}(A^{2}P_{r})=(-1)^{r}\displaystyle\sum_{i=1}^{n}\lambda_{i}^{2}S_{r}(A_{i})=\displaystyle{{n\choose r+1}}(nHH_{r+1}-(n-r-1)H_{r+2}) , where c_{r}=(n-r)\displaystyle{n\choose r} . Associated to each Newton transformation P_{r} one has the second-order linear differential operator L_{r} , defined by (2.5) L_{r}f={\rm tr}(P_{r}\nabla^{2}f), where \nabla^{2}f:\mathfrak{X}(\Sigma)\rightarrow\mathfrak{X}(\Sigma) denotes the self-adjoint linear operator metrically equivalent to the Hessian of f , which is given by \langle\nabla^{2}f(X),Y\rangle=\langle\nabla_{X}\nabla f,Y\rangle, for all X,Y\in\mathfrak{X}(\Sigma) . In particular, for r=0, we get the well known Laplacian operator L_{0}=\Delta , which is always elliptic. The next lemma gives a geometric condition which guarantees the ellipticity of L_{1} (cf. Lemma 3.2 of [ 3 ] ). Lemma 2 . Let \psi:\Sigma^{n}\rightarrow\mathcal{H}^{n+1} be a spacelike hypersurface in \mathcal{H}^{n+1} . If H_{2}>0 on \Sigma^{n} , then L_{1} is elliptic or, equivalently, P_{1} is positive definite (for an appropriate choice of orientation N ). When r\geq 2 , the following lemma establishes sufficient conditions to guarantee the ellipticity of L_{r} (cf. Lemma 3.3 of [ 3 ] ). Lemma 3 . Let \psi:\Sigma^{n}\rightarrow\mathcal{H}^{n+1} be a spacelike hypersurface in \mathcal{H}^{n+1} . If there exists an elliptic point of \Sigma^{n} , with respect to an appropriate choice of orientation N , and H_{r+1}>0 on \Sigma^{n} , for 2\leq r\leq n-1 , then for all 1\leq k\leq r the operator L_{k} is elliptic or, equivalently, P_{k} is positive definite (for

Read the complete text