Curvature at the infinity of asymptotically flat Einstein manifold

Bing Wang, Hao Yin · 2025

Read Curvature at the infinity of asymptotically flat Einstein manifold by Bing Wang, Hao Yin (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-2508-16288.txt.

Introduction

Bing Wang and Hao Yin build coordinates at the far end of an asymptotically flat Ricci-flat manifold in 2025, provided the curvature stays finite in a suitable averaged sense. With those coordinates in place, they can define a Weyl tensor at infinity for such spaces. They also extend a version of renormalized volume, following earlier work, to ALE Ricci-flat manifolds and orbifolds of any dimension. The results give cleaner tools for describing the geometry of these spaces far from their cores.

Catalog record

CREATORBing Wang, Hao Yin
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE92 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-16288

Browse the full library · About the Tower of Babel

Full text

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

Curvature at the infinity of asymptotically flat Einstein manifold Bing Wang, Hao Yin Abstract . In this paper, we construct coordinates at the infinity of an asymptotically flat end of a Ricci-flat manifold ( M m , g ) (M^{m},g) as long as the L m / 2 L^{m/2} norm of the curvature is finite in this end. As applications, we can define a Weyl tensor at the infinity of M M and a version of renormalized volume following Biquard and Hein to ALE Ricci-flat manifolds/orbifolds of any dimension. 1. Introduction Suppose that (M,g) is an m -dimensional Riemannian manifold and \Omega is an end of M satisfying (1) (M,g) is Ricci-flat on \Omega ; (2) Vol(B(o,r)\cap\Omega)\geq Cr^{m} , for some fixed o\in M and C>0 and any r>0 ; (3) \int_{\Omega}\left|Rm\right|^{m/2}dV_{g}<\infty . It was proved in a famous paper of Bando, Kasue and Nakajima [ BKN89 ] that this end is ALE of order \tau=m-1 in the sense that there exists a compact subset K\subset M , R>0 , a finite subgroup \Gamma\subset O(m) acting freely on \mathbb{R}^{m}\setminus B(0,R) , a diffeomorphism \mathcal{L} from \Omega\setminus K to (\mathbb{R}^{m}\setminus B(0,R))/\Gamma such that if \tilde{g}:=(\mathcal{L}^{-1}\circ{\rm proj})^{*}g where {\rm proj} is the projection from \mathbb{R}^{m}\setminus B(0,R) to (\mathbb{R}^{m}\setminus B(0,R))/\Gamma , then \tilde{g}_{ij}(z)=\delta_{ij}+O(\left|z\right|^{-\tau}) and for any positive integer k (1.1) \partial_{k}\tilde{g}_{ij}(z)=O(\left|z\right|^{-\tau-k}). The estimates on the derivatives follow from the Ricci-flat condition and the fact that the coordinate obtained in [ BKN89 ] is a harmonic coordinate system. Moreover, in the case m=4 , or (M,g) is Kähler, this \tau can be improved to be m (see Theorem 1.5 of [ BKN89 ] ). Later, it was observed by Cheeger and Tian [ CT94 ] that as long as \Gamma is not trivial, \tau can be taken to be m in general. Kröncke and Szabó recently [ KS24 ] pointed out a gap in [ CT94 ] and fixed it as a corollary of their main theorems. It is worth noting that both [ CT94 ] and [ KS24 ] discuss the more general setting of asymptotically conical end. While \tau=m is optimal, various authors continue to improve our understanding towards the regularity of metric at the infinity. For example, Chen [ Che20 ] proved a full expansion of \tilde{g}_{ij} in the coordinate constructed in [ BKN89 ] . Recently, more precise expansion for harmonic functions and harmonic one-forms were obtained by Chen and Yan [ CY25 ] . For m=4 , Biquard and Hein conducted a detailed analysis of the leading term h_{0} in the expansion of \tilde{g}_{ij} . In particular, in Section 2 of [ BH23 ] , it was proved that h_{0} is the sum of harmonic gauge terms and Kronheimer terms. While the former could be removed by a further coordinate change, the latter was shown to be exactly the leading term of the metric expansion for the family of ALE Kähler Ricci-flat four manifold due to Kronheimer. Although not explicitly mentioned, the computation in Proposition 2.6 in [ BH23 ] actually established an isomorphism between the space of Kronheimer terms and the space of Weyl tensors on \mathbb{R}^{4} (See also Appendix A of [ Ozu24 ] ). 1.1. The main results In this paper, we revisit the problem of establishing optimal decay rates and characterizing the leading terms in general dimensions. We need the following definition. Definition 1.1 . For a function f defined on \mathbb{R}^{m}\setminus B(0,R) , it is said to be in O(\left|x\right|^{-\tau}) if there exist constants C(k) such that \left|\partial^{(k)}f\right|(x)\leq C(k)\left|x\right|^{-\tau-k}\qquad\text{for}\quad k=0,1,2,\cdots on \mathbb{R}^{m}\setminus B(0,R) . Our first result is Theorem 1.2 . Let (M,g) be m -dimensional Riemannian manifold and \Omega is an end satisfying (1) (M,g) is Ricci-flat on \Omega ; (2) Vol(B(o,r)\cap\Omega)\geq Cr^{m} , for some fixed o\in M and C>0 and any r>0 ; (3) \int_{\Omega}\left|Rm\right|^{m/2}dV_{g}<\infty . Then there is a compact set K in M such that there is a diffeomorphism \Psi from the universal cover of \Omega\setminus K to \mathbb{R}^{m}\setminus B that gives a coordinate system satisfying (I) it is Bianchi, in the sense that if g_{e} is the flat metric given by the coordinates, then B_{g_{e}}g:=\delta_{g_{e}}g-\frac{1}{2}d({\rm Tr}_{g_{e}}g)=0. Moreover, this is equivalent to say that \Psi is a harmonic map from g_{e} to g . (II) the metric tensor has an expansion (1.2) g_{ij}=\delta_{ij}+W_{iklj}\frac{x_{k}x_{l}}{\left|x\right|^{{m}+2}}+O(\left|x\right|^{-m-1}). Here W_{ikjl} is a Weyl tensor. Remark 1.3 . Let \Gamma be the fundamental group of \Omega\setminus K . In previously mentioned results such as [ CT94 ] , [ KS24 ] , it is necessary to assume \Gamma\neq\left\{1\right\} . This is not needed for Theorem 1.2 . The key point is that instead of focusing on one coordinate, we use an intertwined process of finding new coordinates and improving the regularity of the metric. Indeed, in the rest of the paper, we always assume that \Omega\setminus K is simply connected. In case \Gamma\neq\left\{1\right\} , it is easy to pass to the universal cover and work with \Gamma -equivariant spaces in the proofs. Remark 1.4 . There is a harmonic coordinate version of the above theorem, with (I) replaced by: (I’) the components of \Psi are harmonic functions. This could be proved similarly. Instead of repeating, we will indicate at various places how the proofs can be modified for this purpose. Given Theorem 1.2 , it is natural to ask if the Weyl tensor W is independent of the choice of coordinate. Or equivalently, is it an intrinsic geometric property of an end of manifold satisfying the assumptions of Theorem 1.2 ? Our next theorem provides a positive answer. Theorem 1.5 . Let (M,g) and \Omega be as in Theorem 1.2 . Assume that there are two coordinate systems (x_{k}) and (y_{k}) defined on (the universal cover of) \Omega\setminus K such that in each of these coordinate systems, we have Weyl-type tensor W and \tilde{W} respectively satisfying g=(\delta_{ij}+W_{iabj}\frac{x_{a}x_{b}}{\left|x\right|^{m+2}}+O(\left|x\right|^{-(m+1)}))dx_{i}\otimes dx_{j} and g=(\delta_{ij}+\tilde{W}_{iabj}\frac{y_{a}y_{b}}{\left|y\right|^{m+2}}+O(\left|y\right|^{-(m+1)}))dy_{i}\otimes dy_{j}. Then there exists an A\in O(m) such that W_{ijkl}=\tilde{W}_{abcd}A_{ai}A_{bj}A_{ck}A_{dl}. Remark 1.6 . In Definition 1.1 , we have assumed a bound for derivatives of any order. This is not necessary for the validity of Theorem 1.5 . We have chosen to keep the exposition simple. The coordinate in Theorem 1.2 in spite of possessing good properties is not unique (up to rotation). Hence, it is natural to ask the following question (with the exponential map and the normal coordinate in mind): is it possible to define a better coordinate that is unique (up to rotation) and whose metric expansion contains more geometric information? We then move on to discuss the renormalized volume. On noncompact manifolds, the total volume is often infinity. However, under reasonable assumptions on the geometry of the manifolds, it is often useful to compare this infinity (more precisely the growth rate) with some standard example and the difference is usually called the renormalized volume. It was first defined for asymptotically hyperbolic Einstein metrics by Graham [ Gra00 ] . Later, Anderson [ And01 ] found an explicit formula for the renormalized volume in this case. Brendle and Chodosh [ BC14 ] defined a renormalized volume in a different setting for asymptotically hyperbolic metrics. For the asymptotically Euclidean (or flat) case, the renormalized volume was defined by Biquard and Hein [ BH23

Read the complete text