Mass from an Extrinsic Point of View

We express the -th Gauss-Bonnet-Chern mass of an immersed submanifold of Euclidean space as a linear combination of two terms: the total (2)-th mean curvature and the integral, over the entire manifold, of the inner product between the (2 + 1)-th mean curvature vector and the position vector of the...

Повний опис

Збережено в:
Бібліографічні деталі
Опубліковано в: :Symmetry, Integrability and Geometry: Methods and Applications
Дата:2025
Автори: de Sousa, Alexandre, Girão, Frederico
Формат: Стаття
Мова:Англійська
Опубліковано: Інститут математики НАН України 2025
Онлайн доступ:https://nasplib.isofts.kiev.ua/handle/123456789/212873
Теги: Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
Назва журналу:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Цитувати:Mass from an Extrinsic Point of View. Alexandre de Sousa and Frederico Girão. SIGMA 21 (2025), 018, 11 pages

Репозитарії

Digital Library of Periodicals of National Academy of Sciences of Ukraine
_version_ 1860272610117419008
author de Sousa, Alexandre
Girão, Frederico
author_facet de Sousa, Alexandre
Girão, Frederico
citation_txt Mass from an Extrinsic Point of View. Alexandre de Sousa and Frederico Girão. SIGMA 21 (2025), 018, 11 pages
collection DSpace DC
container_title Symmetry, Integrability and Geometry: Methods and Applications
description We express the -th Gauss-Bonnet-Chern mass of an immersed submanifold of Euclidean space as a linear combination of two terms: the total (2)-th mean curvature and the integral, over the entire manifold, of the inner product between the (2 + 1)-th mean curvature vector and the position vector of the immersion. As a consequence, we obtain, for each , a geometric inequality that holds whenever the positive mass theorem (for the -th Gauss-Bonnet-Chern mass) holds.
first_indexed 2026-03-21T11:56:48Z
format Article
fulltext Symmetry, Integrability and Geometry: Methods and Applications SIGMA 21 (2025), 018, 11 pages Mass from an Extrinsic Point of View Alexandre DE SOUSA a and Frederico GIRÃO b a) Escola de Ensino Fundamental e Médio Santa Luzia, Fortaleza, 60110-300, Brazil E-mail: alexandre.ads@pm.me b) Departamento de Matemática, Universidade Federal do Ceará, Fortaleza, 60455-760, Brazil E-mail: fred@mat.ufc.br Received October 22, 2024, in final form March 03, 2025; Published online March 18, 2025 https://doi.org/10.3842/SIGMA.2025.018 Abstract. We express the q-th Gauss–Bonnet–Chern mass of an immersed submanifold of Euclidean space as a linear combination of two terms: the total (2q)-th mean curvature and the integral, over the entire manifold, of the inner product between the (2q + 1)-th mean curvature vector and the position vector of the immersion. As a consequence, we obtain, for each q, a geometric inequality that holds whenever the positive mass theorem (for the q-th Gauss–Bonnet–Chern mass) holds. Key words: Gauss–Bonnet–Chern mass; asymptotically Euclidean submanifolds; positive mass theorem; Hsiung–Minkowski identities 2020 Mathematics Subject Classification: 83C99; 53C40; 51M16 1 Introduction In this article, we explore the concept of mass from an extrinsic point of view. Our approach is based on the introduction of a class of immersions in Euclidean space whose members we call asymptotically Euclidean immersions (Definition 3.5) and on an integral identity (Theorem 3.6) that can be seen as a version of the classical Hsiung-Minkowski formulas [16, 19] for a class of non-compact immersed submanifolds. Through this identity, we deduce a geometric inequality (Corollary 3.7) that must be satisfied whenever the positive mass conjecture for the GBC mass (and ADM mass, in particular) is valid. We also explore the necessary conditions for vector fields to generate the asymptotic charges of interest (Proposition 2.7) and present two conjectures. In one of them, we conjecture that asymptotically Euclidean spaces admit an asymptotically Euclidean isometric immersion (Con- jecture 3.8); in the other, we conjecture that the aforementioned geometric inequality must hold whenever an asymptotically Euclidean immersion satisfies a natural hypothesis (Conjecture 3.9). If both conjectures are valid, the positive mass conjecture for the GBC mass (and ADM mass, in particular) can be directly deduced from our results. 2 Mass of asymptotically Euclidean spaces In this section, we establish the background needed for understanding our method. As far as we are aware, the results presented in Proposition 2.7 and Corollary 2.8 are novel in the literature. They are fundamental ingredients for the employment of our line of action, since they establish sufficient conditions for a vector field to generate the asymptotic charges we are interest in. We begin by remembering the concept of an asymptotically Euclidean end. Definition 2.1 (asymptotically Euclidean end). An asymptotically Euclidean end of order τ , with τ > 0, is a Riemannian manifold (En, g), n ≥ 3, for which there exists a diffeomorphism mailto:alexandre.ads@pm.me mailto:fred@mat.ufc.br https://doi.org/10.3842/SIGMA.2025.018 2 A. de Sousa and F. Girão Ψ: E → Rn \ B1(0), introducing coordinates in E, say Ψ(x) = ( x1, x2, . . . , xn ) , such that, in these coordinates, the following asymptotic condition holds: |gij − δij |+ ρ|gij,k|+ ρ2|gij,kl| = O(ρ−τ ), as ρ→ ∞, (2.1) for all i, j, k, l ∈ {1, 2, . . . , n}, where the gij ’s are the coefficients of g with respect to the coordi- nates Ψ(x), gij,k = ∂gij/∂x k, gij,kl = ∂gij/∂x k∂xl, and ρ = |Ψ(x)| denotes the distance function to the origin with respect to the Euclidean metric induced on the end. Our model mass concept is the ADM mass of an end (E, g), introduced by Arnowitt, Deser and Misner in [2]. Definition 2.2 (ADM mass). The ADM mass of an asymptotically Euclidean end (En, g) is defined by mADM(E, g) = 1 2(n− 1)ωn−1 lim ρ→∞ ∫ Sρ (gij,i − gii,j)ν j dSρ, (2.2) where ωn−1 is the volume of the unit sphere of dimension n− 1, Sρ is the Euclidean coordinate sphere of radius ρ, dSρ is the volume form induced on Sρ by the Euclidean metric, and ν is the outward pointing unit normal to Sρ (with respect to the Euclidean metric). It is known that if τ > (n − 2)/2 and the scalar curvature of (E, g) is integrable, then the limit (2.2) exists, is finite, and is a geometric invariant, that is, two coordinate systems satisfying (2.1) yield the same value for it [4, 5] (see also [30]). A complete Riemannian manifold (Mn, g), n ≥ 3, is said to be asymptotically Euclidean of order τ if there exists a compact subset K of M such that M \ K has finitely many con- nected components and, for any connected component E of M \K, it occurs that (E, g) is an asymptotically Euclidean end of order τ . If (Mn, g) is an asymptotically Euclidean Riemannian manifold of order τ > (n− 2)/2 whose scalar curvature is integrable, then its ADM mass, denoted by mADM(M, g), is defined as the sum of the ADM masses of its ends. One of the most important results in mathematical general relativity is the positive mass theorem (PMT): Theorem 2.3. If (Mn, g) is an asymptotically flat Riemannian manifold of order τ > (n−2)/2 whose scalar curvature is nonnegative and integrable, then each of its ends has nonnegative ADM mass. Moreover, if the ADM mass of at least one of its ends is zero, then (M, g) is isometric to the Euclidean space (Rn, δ). The PMT was settled by Schoen and Yau when n ≤ 7 [34, 35, 36] and when (M, g) is con- formally Euclidean [37], and by Witten when M is spin [40] (see also [33]). The general cases of the PMT were treated by Lohkamp [23, 24, 25, 26] and by Schoen and Yau [38]. Proofs for the case when (M, g) is an Euclidean graph (without the rigidity statement) were given by Lam [20, 21] for graphs of codimension one (see also [8]) and by Mirandola and Vitório [31] for graphs of arbitrary codimension. The case of Euclidean hypersurfaces (not necessarily graphs), including the rigidity statement, was treated in [18]. Note that, since Euclidean graphs (of arbi- trary codimension) and Euclidean hypersurfaces are spin, these cases also follow from Witten’s proof. In [12], a new mass (actually, a family of masses) for asymptotically Euclidean manifolds, called the Gauss–Bonnet–Chern mass, was introduced. For a positive integer q < n/2, consider the q-th Gauss–Bonnet curvature, denoted L(q), and defined by L(q) = 1 2q δ a1a2...a2q−1a2q b1b2...b2q−1b2q q∏ s=1 R b2s−1b2s a2s−1a2s = P ijkl(q) Rijkl, Mass from an Extrinsic Point of View 3 where R is the Riemann curvature tensor of (M, g) and P(q), which has the same symmetries of R (see [12, Section 3]), is given by P ijkl(q) = 1 2q δ a1a2···a2q−3a2q−2ij b1b2···b2q−3b2q−2b2q−1b2q ( q−1∏ s=1 R b2s−1b2s a2s−1a2s ) gb2q−1kgb2ql. Definition 2.4 (GBC mass). The q-th GBC mass of an asymptotically Euclidean end (En, g) is defined by mq(E, g) = c(n, q) lim ρ→∞ ∫ Sρ P ijkl(q) gjk,lνi dSρ, (2.3) where c(n, q) = (n− 2q)! 2q−1(n− 1)!ωn−1 and Sρ, dSρ, ν and ωn−1 are as in the definition of the ADM mass. Note that L(1) is just the scalar curvature and, as observed in [12], m1 coincides with the ADM mass. In the same article, it is shown that if τ > τq and L(q) is integrable, then the limit (2.3) exists, is finite, and is a geometric invariant, where here and throughout the text, τq = (n− 2q) (q + 1) . As in the q = 1 case, if (Mn, g) is an asymptotically Euclidean manifold of order τ > τq whose q-th Gauss–Bonnet curvature is integrable, then its q-th GBC mass, denoted by mq(M, g), is defined as the sum of the q-th GBC masses of its ends. The following is a version of the PMT for the GBC mass. Conjecture 2.5. Let n and q be integers such that n ≥ 3 and 0 < q < n/2. If (Mn, g) is an asymptotically Euclidean Riemannian manifold of order τ > τq whose q-th Gauss–Bonnet curvature L(q) is nonnegative and integrable, then the q-th GBC mass of each of its ends is nonnegative. Moreover, if the GBC mass of at least one of its ends is zero, then (M, g) is isometric to the Euclidean space (Rn, δ). Conjecture 2.5, without the rigidity statement, was proved for graphs of codimension one in [12]; the case of graphs of arbitrary codimension and flat normal bundle was done by Li, Wei and Xiong when q = 2 [22] and by the authors when 0 < q < n/2 [10, 11]. This conjecture, including the rigidity statement, is known to be true for conformally flat manifolds [13]. 2.1 Mass in terms of the Lovelock tensor Let (En, g), n ≥ 3, be an asymptotically Euclidean end of order τ > τ1, and let G be its Einstein tensor, that is, G = Ric− 1 2 (Sc)g, where Ric and Sc denote, respectively, the Ricci tensor and the scalar curvature of (E, g). Throughout the text, we denote by X the vector field given by X = xi ∂ ∂xi , (2.4) where Ψ(x) = ( x1, x2, . . . , xn ) is a coordinate system satisfying (2.1). 4 A. de Sousa and F. Girão Is is known that the ADM mass of (E, g) can be computed as follows (see [3, 6, 7]): mADM(E, g) = − 1 (n− 1)(n− 2)ωn−1 lim ρ→∞ ∫ Sρ G(X, νg) dS g ρ , (2.5) where ωn−1 and Sρ are as in (2.2), νg is the outward unit normal vector to Sρ with respect to the metric g, and dSgρ is the volume form induced on Sρ by g. The equivalence between formulas (2.2) and (2.5) can be shown by reducing the general case to the case of harmonic asymptotics via a density theorem (see, for example, [17]). Proofs without the use of a density theorem were given in [15, 29]. As we will see in a moment, a formula similar to (2.5) holds for the GBC mass. To state this result, we need to recall the so-called Lovelock curvature tensors. Let (Mn, g), n ≥ 3, be a Riemannian manifold and let q < n/2 be a positive integer. The q-th Lovelock curvature tensor, denoted by G(q), is defined by G(q)ij = − 1 2q+1 gikδ kb1b2...b2q−1b2q ja1a2...a2q−1a2q q∏ s=1 R a2s−1a2s b2s−1b2s . (2.6) Note that G(1) is just the Einstein tensor. Proposition 2.6. The Lovelock curvature tensor G(q) satisfies the following: (i) It is symmetric, that is, G(q)ij = G(q)ji. (2.7) (ii) It is divergent-free, that is, ∇iG(q)ij = 0. (2.8) (iii) Its trace satisfies the equation trg G(q) = −n− 2q 2 L(q). (2.9) Proof. Identities (2.7) and (2.8) follow from [28, Theorem 1] (see also [27]). Identity (2.9) is a straightforward computation. ■ Let (En, g), n ≥ 3, be an asymptotically Euclidean end of order τ > τq, where q < n/2 is a positive integer. It was shown in [39] that the q-th GBC mass of (E, g) can be computed as follows: mq(E, g) = −b(n, q) lim ρ→∞ ∫ Sρ G(q)(X, νg) dS g ρ , (2.10) where X, νg and dSg are as in (2.5) and b(n, q) = (n− 2q − 1)! 2q−1(n− 1)!ωn−1 . Note that, when q = 1, formulas (2.10) and (2.5) coincide. The following proposition establishes sufficient conditions for a vector field to generate the GBC mass. Mass from an Extrinsic Point of View 5 Proposition 2.7. Let (En, g), n ≥ 3, be an asymptotically Euclidean end of order τ > τq, where q < n/2 is a positive integer. Let Ψ(x) = (x1, x2, . . . , xn) be coordinates in E satisfying (2.1) and let X be the vector field given by (2.4). If Y is a vector field on E such that( Y i −Xi ) (x) = O ( ρ−τ+1 ) , i = 1, 2, . . . , n, as ρ→ ∞, then mq(E, g) = −b(n, q) lim ρ→∞ ∫ Sρ G(q)(Y, νg) dS g ρ . Proof. Let Ψ(x) = ( x1, x2, . . . , xn ) be coordinates on E satisfying (2.1). We can use these coordinates to compare νg, the unit normal to Sρ with respect to g, and dSgρ , the volume form induced on Sρ by g, with their Euclidean counterparts νδ and dSδρ, respectively. It holds νig − νiδ = O(ρ−τ ), as ρ→ ∞, and dSgρ = (1 + w) dSδρ for some function w : E → R satisfying w = O(ρ−τ ), as ρ→ ∞. Furthermore, the components of the Riemman curvature tensor satisfy Rlijk = O ( ρ−τ−2 ) , as ρ→ ∞. Together with (2.6), this gives G(q)ij = O ( ρ−q(τ+2) ) , as ρ→ ∞. Thus, we have∣∣∣∣∫ Sρ G(q)(Y, νg) dS g ρ − ∫ Sρ G(q)(X, νg) dS g ρ ∣∣∣∣ ≤ ∫ Sρ ∣∣G(q)ij ( Y i −Xi )∣∣(∣∣νig − νiδ ∣∣+ ∣∣νiδ∣∣)(1 + w) dSδρ ≤ C(n)ρ−[q(τ+2)+(τ−1)−(n−1)](1 + ρ−τ )2, where C(n) is a constant that depends only on the dimension. Using that τ > τq, it follows that lim ρ→∞ ∣∣∣∣∫ Sρ G(q)(Y, νg) dS g ρ − ∫ Sρ G(q)(X, νg) dS g ρ ∣∣∣∣ = 0. (2.11) The proposition follows from (2.10) and (2.11). ■ The specialization of this proposition to the case of gradient fields is a key ingredient in the development of our extrinsic approach. 6 A. de Sousa and F. Girão Corollary 2.8. Under the same hypothesis as the ones in Proposition 2.7, if there exists f ∈ C∞(E) such that ∂ ∂xi ( f − ρ2 2 ) = O ( ρ−τ+1 ) , i = 1, 2, . . . , n, (2.12) as ρ→ ∞, then mq(E, g) = −b(n, q) lim ρ→∞ ∫ Sρ G(q) ( ∇f, νg ) dSgρ . Proof. Write gij = δij + θij and take Y = ∇f . We have Y i −Xi = gij ∂f ∂xj − ∂ ∂xi ( ρ2 2 ) = δij ∂f ∂xj + θij ∂f ∂xj − ∂ ∂xi ( ρ2 2 ) = ∂ ∂xi ( f − ρ2 2 ) + θij ∂f ∂xj . (2.13) Note that (2.12) is equivalent to ∂f ∂xi = xi +O ( ρ−τ+1 ) , as ρ→ ∞. (2.14) Thus, since θij = O(ρ−τ ), as ρ→ ∞, (2.15) from (2.14) and (2.15) we find θij ∂f ∂xj = O ( ρ−τ+1 ) , as ρ→ ∞. (2.16) The corollary follows from Proposition 2.7, (2.13) and (2.16). ■ 3 Imersions in Euclidean spaces Let ψ : Mn ↬ Rd, d > n, be a smooth immersion and let δ̄ = ⟨·, ·⟩ be the canonical Euclidean metric on Rd. All quantities related to this ambient space will also be denoted with an overbar, unless otherwise indicated. Let B (which is normal-vector valued) be the second fundamental form of the immersion and let p be an integer such that 0 < p ≤ n; if p is even, then the p-th mean curvature S(p) ∈ C∞(M) is defined by S(p) = 1 p! δ b1...bp a1...ap 〈 Ba1 b1 , Ba2 b2 〉 · · · 〈 B ap−1 bp−1 , B ap bp 〉 , and, if p is odd, then the p-th mean curvature S(p) ∈ Γ ( TM⊥) is defined by S(p) = 1 p! δ b1...bp a1...ap 〈 Ba1 b1 , Ba2 b2 〉 · · · 〈 B ap−2 bp−2 , B ap−1 bp−1 〉 B ap bp , where ⟨·, ·⟩ denotes the Euclidean metric on Rd. We also set S(0) = 1 and S(n+1) = 0. Next, we turn to the so-called Newton transformations. The 0-th Newton transformation is defined as T(0) = g, Mass from an Extrinsic Point of View 7 where g = ψ∗δ̄ denotes the induced metric. If p ≥ 2 is even, then the p-th Newton transforma- tion T(p) is defined by T(p)ij = 1 p! gikδ kb1b2...bp ja1a2...ap 〈 Ba1 b1 , Ba2 b2 〉 · · · 〈 B ap−1 bp−1 , B ap bp 〉 , (3.1) and, if p is odd, then the p-th Newton transformation T(p) is defined by T(p)ij = 1 p! gikδ kb1b2...bp−1 ja1a2...ap−1 〈 Ba1 b1 , Ba2 b2 〉 · · · 〈 B ap−2 bp−2 , B ap−1 bp−1 〉 B ap bp . Note that, when p is odd, the p-th Newton transformation is normal-vector valued. Furthermore, by antisymmetry, it follows that T(p) ≡ 0 when p > n. The relation between the trace of Newton transformations and the higher-order mean curva- tures is a fact well-known in the literature (see [14, Lemma 2.2], for example). Proposition 3.1. Let p be an integer such that 0 ≤ p ≤ n. The Newton transformation T(p) satisfy trgT(p) = (n− p)S(p). (3.2) Moreover, if p is even, then T(p)ijB ij = (p+ 1)S(p+1). By a direct application of the Gauss equation, we obtain that the q-th Lovelock tensor (2.6) of the induced metric g = ψ∗δ̄ and the 2q-th Newton transformation (3.1) of an immersion contain the same information. Proposition 3.2. Let q be a positive integer such that q < n/2. It holds Gq = −(2q)! 2 T(2q). (3.3) The next lemma is an infinitesimal version of a Pohozaev–Schoen-type integral identity pre- sented in [9, Propositon 3.2]. Lemma 3.3. If K and V are, respectively, a symmetric bilinear form and a vector field on M , then the following identity holds: divg(ιVK) = ιV (divgK) + 1 2 g(K,LV g). (3.4) Throughout the text, we denote by Z̄ the vector field on Rd given by Z̄ = x̄α ∂ ∂x̄α , 1 ≤ α ≤ d, where ( x̄1, x̄2, . . . , x̄d ) is the standard coordinate system on Rd. It is known that it is a conformal Killing gradient field and that it satisfies the following identity: Z̄ = ∇ρ̄2 2 , where ρ̄ is the distance function to origin on Rd. The following proposition can be interpreted as an infinitesimal version of the integral iden- tity of Theorem 3.6. It is inspired by a infinitesimal version of the flux formula presented in [1, equation (8.4)]. 8 A. de Sousa and F. Girão Proposition 3.4. Let Y = ψ∗Z̄⊤, where Z̄⊤ denotes the tangent part of the vector field Z̄. It holds divg(ιYG(q)) = −(2q)! 2 [ (n− 2q)S2q + (2q + 1) 〈 S2q+1, Z̄ 〉] . (3.5) Proof. Since Y = ψ∗Z̄⊤, we have ψ∗Y = Z̄⊤ = Z̄ − Z̄⊥ and LY g = LY ψ∗δ̄ = ψ∗(Lψ∗Y δ̄ ) = ψ∗(L(Z̄−Z̄⊥)δ̄ ) = ψ∗(LZ̄ δ̄)− ψ∗(LZ̄⊥ δ̄ ) . (3.6) Denote by BZ̄ the symmetric bilinear form on M defined by BZ̄(V,W ) = 〈 B(V,W ), Z̄ 〉 . For any V,W ∈ Γ(TM),( LZ̄⊥ δ̄ ) (V,W ) = 〈 DV Z̄ ⊥,W 〉 + 〈 V,DW Z̄ ⊥〉 = −2 〈 B(V,W ), Z̄ 〉 = −2BZ̄(V,W ), that is, LZ̄⊥ δ̄ = −2BZ̄ . (3.7) We also have LZ̄ δ̄ = 2δ̄. (3.8) Thus, (3.6)–(3.8) yield g(G(q),LY g) = 2 trgG(q) + 2g(G(q), BZ̄). (3.9) Identity (3.5) follows from (2.8), (3.2)–(3.4) and (3.9). ■ 3.1 Asymptotically Euclidean imersions Next, we describe our main object of study: a class of smooth immersions that place certain smooth manifolds Mn into some Euclidean space Rd, where d > n, in a special way. Definition 3.5 (asymptotically Euclidean immersion). Let ψ : Mn ↬ Rd, d > n, be a smooth immersion of a smooth manifold Mn, n ≥ 3. We say that ψ is an asymptotically Euclidean immersion of order τ , for some τ > 0, if the following conditions hold: (i) the Riemannian manifold ( M,ψ∗δ̄ ) is complete and asymptotically Euclidean of order τ ; (ii) if E ⊂ M is such that (E,ψ∗δ̄) is an asymptotically Euclidean end of order τ , then, as ρ→ ∞, ∂ ∂xi ( ψ∗ρ̄2 − ρ2 ) (x) = O ( ρ−τ+1 ) , i = 1, 2, . . . , n, for any coordinate system Ψ(x) = ( x1, x2, . . . , xn ) on E satisfying (2.1). The following theorem is the main result of this article. It can be seen as a version of the Hsiung–Minkowski formulas [16, 19] to asymptotically Euclidean immersions. Mass from an Extrinsic Point of View 9 Theorem 3.6. Let n and q be integers such that n ≥ 3 and 0 < q < n/2. If ψ : Mn ↪→ ( Rd, δ̄ ) , d > n, is an asymptotically Euclidean immersion of order τ > τq such that L(q), the q-th Gauss- Bonnet curvature of the induced metric ψ∗δ̄, is integrable, then the functions S2q and 〈 S2q+1, Z̄ 〉 are integrable and the q-th Gauss–Bonnet–Chern mass of ( M,ψ∗δ̄ ) satisfies the following integral identity: mq ( M,ψ∗δ̄ ) = a(n, q) [ (n− 2q) ∫ M S2q dM + (2q + 1) ∫ M ⟨S2q+1, Z̄⟩dM ] , where dM is the Riemannian measure on ( M,ψ∗δ̄ ) and a(n, q) = (2n)!(n− 2q − 1)! 2q(n− 1)!ωn−1 . Proof. Let E be one of the ends of M . Consider the function f : M → R, with f = ψ∗ρ̄2 2 . Since ψ : Mn ↬ Rd is an asymptotically Euclidean immersion of order τ > τq, the function f satisfies (2.12), and hence, by Corollary 2.8, mq ( E,ψ∗δ̄ ) = −b(n, q) lim ρ→∞ ∫ Sρ G(q)(∇f, ν) dSρ, (3.10) where the geometric quantities in the integral are computed with respect the induced metric ψ∗δ̄. Note that ∇f = ψ∗Z̄⊤. Thus, applying equation (3.10) to all the ends of M together with the divergence theorem and identity (3.5), we find mq ( M,ψ∗δ̄ ) = a(n, q) ∫ M [ (n− 2q)S2q + (2q + 1) 〈 S2q+1, Z̄ 〉] dM. (3.11) It remains to show that both S2q and 〈 S2q+1, Z̄ 〉 are integrable, so the right hand side of (3.11) can be broken in two. The integrability of S2q follows from (2.9), (3.2), (3.3) and the integrability of L(q). Once we know that S(2q) is integrable, the integrability of 〈 S2q+1, Z̄ 〉 follows from the fact that the left hand side of (3.11) is finite. ■ An immediate consequence of this theorem is the following. Corollary 3.7. Under the same hypothesis as the ones in Theorem 3.6, mq ( M,ψ∗δ̄ ) ≥ 0 if and only if (n− 2q) ∫ M S2q dM + (2q + 1) ∫ M 〈 S2q+1, Z̄ 〉 dM ≥ 0. A famous theorem by Nash [32] states that any Riemannian manifold (M, g) can be isomet- rically immersed in some Euclidean space ( Rd, δ̄ ) . Encouraged by Nash’s theorem, we make the following conjecture. Conjecture 3.8. If (Mn, g), n ≥ 3, is an asymptotically Euclidean manifold of order τ > 0, then there exists an asymptotically Euclidean isometric immersion ψ : M → Rd of order τ (in the sense of Definition 3.5). 10 A. de Sousa and F. Girão Let n and q be integers such that n ≥ 3 and 0 < q < n/2. Suppose that Conjecture 3.8 is true for every asymptotically Euclidean manifold (Mn, g) of order τ > τq (or at least for those whose Gauss–Bonnet curvature L(q) is nonnegative and integrable). Then, by Theorem 3.6, Conjecture 2.5 would be a direct consequence of the following conjecture. Conjecture 3.9. Let n and q be integers such that n ≥ 3 and 0 < q < n/2. Let ψ : Mn → Rd, d > n, be an asymptotically Euclidean immersion of order τ > τq for which S2q is integrable and nonnegative. It holds (n− 2q) ∫ M S2q dM + (2q + 1) ∫ M 〈 S2q+1, Z̄ 〉 dM ≥ 0, with the equality holding if and only if ( M,ψ∗δ̄ ) is isometric to Euclidean space. Acknowledgements This study was partially financed by the Coordenação de Aperfeiçoamento de Pessoal de Nı́vel Superior - Brasil (CAPES) - Finance Code 001. This work was partially done while Alexandre de Sousa was a CAPES Fellow at the Mathematics Institute of Federal University of Alagoas (IM/UFAL), whose members he would like to thank for the hospitality. Frederico Girão was partially supported by CNPq, grant number 307239/2020-9. References [1] Aĺıas L.J., de Lira J.H.S., Malacarne J.M., Constant higher-order mean curvature hypersurfaces in Rieman- nian spaces, J. Inst. Math. Jussieu 5 (2006), 527–562, arXiv:math.DG/0311352. [2] Arnowitt R., Deser S., Misner C.W., Coordinate invariance and energy expressions in general relativity, Phys. Rev. 122 (1961), 997–1006. [3] Ashtekar A., Hansen R.O., A unified treatment of null and spatial infinity in general relativity. I. Universal structure, asymptotic symmetries, and conserved quantities at spatial infinity, J. Math. Phys. 19 (1978), 1542–1566. [4] Bartnik R., The mass of an asymptotically flat manifold, Comm. Pure Appl. Math. 39 (1986), 661–693. [5] Chruściel P., Boundary conditions at spatial infinity from a Hamiltonian point of view, in Topological Properties and Global Structure of Space-time (Erice, 1985), NATO Adv. Sci. Inst. Ser. B: Phys., Vol. 138, Plenum, New York, 1986, 49–59, arXiv:1312.0254. [6] Chruściel P., A remark on the positive-energy theorem, Classical Quantum Gravity 3 (1986), L115–L121. [7] Chruściel P., Corrigendum: “A remark on the positive-energy theorem”, Classical Quantum Gravity 4 (1987), 1049. [8] de Lima L.L., Girão F., The ADM mass of asymptotically flat hypersurfaces, Trans. Amer. Math. Soc. 367 (2015), 6247–6266, arXiv:1108.5474. [9] de Lima L.L., Girão F., Montalbán A., The mass in terms of Einstein and Newton, Classical Quantum Gravity 36 (2019), 075017, 11 pages, arXiv:1811.06924. [10] de Sousa A., Um breve estudo sobre a massa de Gauss–Bonnet–Chern dos gráficos euclidianos, Ph.D. Thesis, Universidade Federeal do Ceará, 2016, available at http://www.repositorio.ufc.br/handle/riufc/71518. [11] de Sousa A., Girão F., The Gauss–Bonnet–Chern mass of higher-codimension graphs, Pacific J. Math. 298 (2019), 201–216, arXiv:1509.00456. [12] Ge Y., Wang G., Wu J., A new mass for asymptotically flat manifolds, Adv. Math. 266 (2014), 84–119, arXiv:1211.3645. [13] Ge Y., Wang G., Wu J., The Gauss–Bonnet–Chern mass of conformally flat manifolds, Int. Math. Res. Not. 2014 (2014), 4855–4878, arXiv:1212.3213. [14] Grosjean J.-F., Upper bounds for the first eigenvalue of the Laplacian on compact submanifolds, Pacific J. Math. 206 (2002), 93–112. https://doi.org/10.1017/S1474748006000077 https://arxiv.org/abs/math.DG/0311352 https://doi.org/10.1103/PhysRev.122.997 https://doi.org/10.1063/1.523863 https://doi.org/10.1002/cpa.3160390505 https://doi.org/10.1007/978-1-4899-3626-4_5 https://arxiv.org/abs/1312.0254 https://dx.doi.org/10.1088/0264-9381/3/6/002 https://doi.org/10.1088/0264-9381/4/4/536 https://doi.org/10.1090/S0002-9947-2014-05902-3 https://arxiv.org/abs/1108.5474 https://doi.org/10.1088/1361-6382/ab090a https://doi.org/10.1088/1361-6382/ab090a https://arxiv.org/abs/1811.06924 http://www.repositorio.ufc.br/handle/riufc/71518 https://doi.org/10.2140/pjm.2019.298.201 https://arxiv.org/abs/1509.00456 https://doi.org/10.1016/j.aim.2014.08.006 https://arxiv.org/abs/1211.3645 https://doi.org/10.1093/imrn/rnt095 https://arxiv.org/abs/1212.3213 https://doi.org/10.2140/pjm.2002.206.93 https://doi.org/10.2140/pjm.2002.206.93 Mass from an Extrinsic Point of View 11 [15] Herzlich M., Computing asymptotic invariants with the Ricci tensor on asymptotically flat and asymptoti- cally hyperbolic manifolds, Ann. Henri Poincaré 17 (2016), 3605–3617, arXiv:1503.00508. [16] Hsiung C.-C., Some integral formulas for closed hypersurfaces in Riemannian space, Pacific J. Math. 6 (1956), 291–299. [17] Huang L.-H., On the center of mass in general relativity, in Fifth International Congress of Chinese Mathe- maticians. Part 1, 2, AMS/IP Stud. Adv. Math., Vol. 51, American Mathematical Society, Providence, RI, 2012, 575–591, arXiv:1101.0456. [18] Huang L.-H., Wu D., Hypersurfaces with nonnegative scalar curvature, J. Differential Geom. 95 (2013), 249–278, arXiv:1102.5749. [19] Kwong K.-K., An extension of Hsiung–Minkowski formulas and some applications, J. Geom. Anal. 26 (2016), 1–23, arXiv:1307.3025. [20] Lam M.-K.G., The graph cases of the Riemannian positive mass and Penrose inequalities in all dimensions, arXiv:1010.4256. [21] Lam M.-K.G., The graph cases of the Riemannian positive mass and Penrose inequalities in all dimensions, Ph.D. Thesis, Duke University, 2011, available at https://hdl.handle.net/10161/3857. [22] Li H., Wei Y., Xiong C., The Gauss–Bonnet–Chern mass for graphic manifolds, Ann. Global Anal. Geom. 45 (2014), 251–266, arXiv:1304.6561. [23] Lohkamp J., The higher dimensional positive mass theorem I, arXiv:math.DG/0608795. [24] Lohkamp J., Skin structures on minimal hypersurfaces, arXiv:1512.08249. [25] Lohkamp J., Hyperbolic geometry and potential theory on minimal hypersurfaces, arXiv:1512.08251. [26] Lohkamp J., Skin structures in scalar curvature geometry, arXiv:1512.08252. [27] Lovelock D., Divergence-free tensorial concomitants, Aequationes Math. 4 (1970), 127–138. [28] Lovelock D., The Einstein tensor and its generalizations, J. Math. Phys. 12 (1971), 498–501. [29] Miao P., Tam L.-F., Evaluation of the ADM mass and center of mass via the Ricci tensor, Proc. Amer. Math. Soc. 144 (2016), 753–761, arXiv:1408.3893. [30] Michel B., Geometric invariance of mass-like asymptotic invariants, J. Math. Phys. 52 (2011), 052504, 14 pages, arXiv:1012.3775. [31] Mirandola H., Vitório F., The positive mass theorem and Penrose inequality for graphical manifolds, Comm. Anal. Geom. 23 (2015), 273–292, arXiv:1304.3504. [32] Nash J., The imbedding problem for Riemannian manifolds, Ann. of Math. 63 (1956), 20–63. [33] Parker T., Taubes C.H., On Witten’s proof of the positive energy theorem, Comm. Math. Phys. 84 (1982), 223–238. [34] Schoen R., Yau S.-T., Complete manifolds with nonnegative scalar curvature and the positive action con- jecture in general relativity, Proc. Nat. Acad. Sci. USA 76 (1979), 1024–1025. [35] Schoen R., Yau S.-T., On the proof of the positive mass conjecture in general relativity, Comm. Math. Phys. 65 (1979), 45–76. [36] Schoen R., Yau S.-T., The energy and the linear momentum of space-times in general relativity, Comm. Math. Phys. 79 (1981), 47–51. [37] Schoen R., Yau S.-T., Conformally flat manifolds, Kleinian groups and scalar curvature, Invent. Math. 92 (1988), 47–71. [38] Schoen R., Yau S.-T., Positive scalar curvature and minimal hypersurface singularities, in Surveys in Dif- ferential Geometry 2019. Differential Geometry, Calabi–Yau Theory, and General Relativity. Part 2, Surv. Differ. Geom., Vol. 24, International Press, Boston, MA, 2022, 441–480, arXiv:1704.05490. [39] Wang G., Wu J., Chern’s magic form and the Gauss–Bonnet–Chern mass, Math. Z. 287 (2017), 843–854, arXiv:1510.03036. [40] Witten E., A new proof of the positive energy theorem, Comm. Math. Phys. 80 (1981), 381–402. https://doi.org/10.1007/s00023-016-0494-5 https://arxiv.org/abs/1503.00508 https://arxiv.org/abs/1101.0456 https://doi.org/10.4310/jdg/1376053447 https://arxiv.org/abs/1102.5749 https://doi.org/10.1007/s12220-014-9536-8 https://arxiv.org/abs/1307.3025 https://arxiv.org/abs/1010.4256 https://hdl.handle.net/10161/3857 https://doi.org/10.1007/s10455-013-9399-4 https://arxiv.org/abs/1304.6561 https://arxiv.org/abs/math.DG/0608795 https://arxiv.org/abs/1512.08249 https://arxiv.org/abs/1512.08251 https://arxiv.org/abs/1512.08252 https://doi.org/10.1007/BF01817753 https://doi.org/10.1063/1.1665613 https://doi.org/10.1090/proc12726 https://doi.org/10.1090/proc12726 https://arxiv.org/abs/1408.3893 https://doi.org/10.1063/1.3579137 https://arxiv.org/abs/1012.3775 https://doi.org/10.4310/CAG.2015.v23.n2.a2 https://doi.org/10.4310/CAG.2015.v23.n2.a2 https://arxiv.org/abs/1304.3504 https://doi.org/10.2307/1969989 https://doi.org/10.1007/BF01208569 https://doi.org/10.1073/pnas.76.3.1024 https://doi.org/10.1007/BF01940959 https://doi.org/10.1007/BF01208285 https://doi.org/10.1007/BF01208285 https://doi.org/10.1007/BF01393992 https://arxiv.org/abs/1704.05490 https://doi.org/10.1007/s00209-017-1847-6 https://arxiv.org/abs/1510.03036 https://doi.org/10.1007/BF01208277 1 Introduction 2 Mass of asymptotically Euclidean spaces 2.1 Mass in terms of the Lovelock tensor 3 Imersions in Euclidean spaces 3.1 Asymptotically Euclidean imersions References
id nasplib_isofts_kiev_ua-123456789-212873
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
issn 1815-0659
language English
last_indexed 2026-03-21T11:56:48Z
publishDate 2025
publisher Інститут математики НАН України
record_format dspace
spelling de Sousa, Alexandre
Girão, Frederico
2026-02-13T13:49:03Z
2025
Mass from an Extrinsic Point of View. Alexandre de Sousa and Frederico Girão. SIGMA 21 (2025), 018, 11 pages
1815-0659
2020 Mathematics Subject Classification: 83C99; 53C40; 51M16
arXiv:2403.06782
https://nasplib.isofts.kiev.ua/handle/123456789/212873
https://doi.org/10.3842/SIGMA.2025.018
We express the -th Gauss-Bonnet-Chern mass of an immersed submanifold of Euclidean space as a linear combination of two terms: the total (2)-th mean curvature and the integral, over the entire manifold, of the inner product between the (2 + 1)-th mean curvature vector and the position vector of the immersion. As a consequence, we obtain, for each , a geometric inequality that holds whenever the positive mass theorem (for the -th Gauss-Bonnet-Chern mass) holds.
This study was partially financed by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior- Brasil (CAPES) - Finance Code 001. This work was partially done while Alexandre de Sousa was a CAPES Fellow at the Mathematics Institute of Federal University of Alagoas (IM/UFAL), whose members he would like to thank for the hospitality. Frederico Girão was partially supported by CNPq, grant number 307239/2020-9.
en
Інститут математики НАН України
Symmetry, Integrability and Geometry: Methods and Applications
Mass from an Extrinsic Point of View
Article
published earlier
spellingShingle Mass from an Extrinsic Point of View
de Sousa, Alexandre
Girão, Frederico
title Mass from an Extrinsic Point of View
title_full Mass from an Extrinsic Point of View
title_fullStr Mass from an Extrinsic Point of View
title_full_unstemmed Mass from an Extrinsic Point of View
title_short Mass from an Extrinsic Point of View
title_sort mass from an extrinsic point of view
url https://nasplib.isofts.kiev.ua/handle/123456789/212873
work_keys_str_mv AT desousaalexandre massfromanextrinsicpointofview
AT giraofrederico massfromanextrinsicpointofview