Difference between revisions of "Gauss-Bonnet theorem"
Ulf Rehmann (talk | contribs) m (moved Gauss–Bonnet theorem to Gauss-Bonnet theorem: ascii title) |
(TeX) |
||
Line 1: | Line 1: | ||
− | The total curvature | + | {{TEX|done}} |
+ | The total curvature $\omega$ of a two-dimensional compact Riemannian manifold $V^2$, closed or with boundary, and the rotation $\tau$ of its smooth boundary $\partial V^2$ are connected with the Euler characteristic $\chi$ of $V^2$ by the relation | ||
− | + | $$\omega+\tau=2\pi\chi.$$ | |
Here | Here | ||
− | + | $$\omega=\int\limits_{V^2}KdS,$$ | |
− | where | + | where $K$ is the Gaussian curvature and $dS$ is the area element; |
− | + | $$\tau=\int\limits_{\partial V^2}k_gdl,$$ | |
− | where | + | where $k_g$ is the geodesic curvature and $dl$ is the line element of the boundary. The Gauss–Bonnet theorem is also valid for a manifold with a piecewise-smooth boundary; in that case |
− | + | $$\tau=\int k_gdl+\sum_i(\pi-\alpha_i),$$ | |
− | where | + | where $\pi-\alpha_i$ is the rotation of the boundary at a vertex. The theorem is valid, in particular, on regular surfaces in $E^3$. The Gauss–Bonnet theorem was known to C.F. Gauss [[#References|[1]]]; it was published by O. Bonnet [[#References|[2]]] in a special form (for surfaces homeomorphic to a disc). |
− | For a non-compact complete manifold | + | For a non-compact complete manifold $V^2$ without boundary, an analogue of the Gauss–Bonnet theorem is the Cohn-Vossen inequality [[#References|[3]]]: |
− | + | $$\int\limits_{V^2}KdS\leq2\pi\chi.$$ | |
The Gauss–Bonnet theorem and the given inequality are also valid for convex surfaces and two-dimensional manifolds of bounded curvature. | The Gauss–Bonnet theorem and the given inequality are also valid for convex surfaces and two-dimensional manifolds of bounded curvature. | ||
− | The Gauss–Bonnet theorem can be generalized to even-dimensional compact Riemannian manifolds | + | The Gauss–Bonnet theorem can be generalized to even-dimensional compact Riemannian manifolds $V^{2p}$, closed or with boundary: |
− | + | $$\int\limits_{V^{2p}}\Omega dS+\int\limits_{\partial V^{2p}}\phi dl=\frac{(2\pi)^p}{(2p-1)!}\chi,$$ | |
− | where | + | where $dS$, $dl$ denote the volume elements in $V^{2p}$ and $\partial V^{2p}$, while $\Omega$ is some polynomial in the components of the curvature tensor of $V^{2p}$, and $\phi$ is some polynomial in the components of the curvature tensor and the coefficients of the second fundamental form of $\partial V^{2p}$ [[#References|[4]]]. The Gauss–Bonnet theorem has also been generalized to Riemannian polyhedra [[#References|[5]]]. Other generalizations of the theorem are connected with integral representations of characteristic classes by parameters of the Riemannian metric [[#References|[4]]], [[#References|[6]]], [[#References|[7]]]. |
====References==== | ====References==== |
Latest revision as of 09:38, 22 August 2014
The total curvature $\omega$ of a two-dimensional compact Riemannian manifold $V^2$, closed or with boundary, and the rotation $\tau$ of its smooth boundary $\partial V^2$ are connected with the Euler characteristic $\chi$ of $V^2$ by the relation
$$\omega+\tau=2\pi\chi.$$
Here
$$\omega=\int\limits_{V^2}KdS,$$
where $K$ is the Gaussian curvature and $dS$ is the area element;
$$\tau=\int\limits_{\partial V^2}k_gdl,$$
where $k_g$ is the geodesic curvature and $dl$ is the line element of the boundary. The Gauss–Bonnet theorem is also valid for a manifold with a piecewise-smooth boundary; in that case
$$\tau=\int k_gdl+\sum_i(\pi-\alpha_i),$$
where $\pi-\alpha_i$ is the rotation of the boundary at a vertex. The theorem is valid, in particular, on regular surfaces in $E^3$. The Gauss–Bonnet theorem was known to C.F. Gauss [1]; it was published by O. Bonnet [2] in a special form (for surfaces homeomorphic to a disc).
For a non-compact complete manifold $V^2$ without boundary, an analogue of the Gauss–Bonnet theorem is the Cohn-Vossen inequality [3]:
$$\int\limits_{V^2}KdS\leq2\pi\chi.$$
The Gauss–Bonnet theorem and the given inequality are also valid for convex surfaces and two-dimensional manifolds of bounded curvature.
The Gauss–Bonnet theorem can be generalized to even-dimensional compact Riemannian manifolds $V^{2p}$, closed or with boundary:
$$\int\limits_{V^{2p}}\Omega dS+\int\limits_{\partial V^{2p}}\phi dl=\frac{(2\pi)^p}{(2p-1)!}\chi,$$
where $dS$, $dl$ denote the volume elements in $V^{2p}$ and $\partial V^{2p}$, while $\Omega$ is some polynomial in the components of the curvature tensor of $V^{2p}$, and $\phi$ is some polynomial in the components of the curvature tensor and the coefficients of the second fundamental form of $\partial V^{2p}$ [4]. The Gauss–Bonnet theorem has also been generalized to Riemannian polyhedra [5]. Other generalizations of the theorem are connected with integral representations of characteristic classes by parameters of the Riemannian metric [4], [6], [7].
References
[1] | C.F. Gauss, , Werke , 8 , K. Gesellschaft Wissenschaft. Göttingen (1900) |
[2] | O. Bonnet, J. École Polytechnique , 19 (1848) pp. 1–146 |
[3] | S.E. Cohn-Vossen, "Some problems of differential geometry in the large" , Moscow (1959) (In Russian) |
[4] | V.A. Sharafutdinov, "Relative Euler class and the Gauss–Bonnet theorem" Siberian Math. J. , 14 : 6 (1973) pp. 930–940 Sibirsk Mat. Zh. , 14 : 6 pp. 1321–1635 |
[5] | C.B. Allendörfer, A. Weil, "The Gauss–Bonnet theorem for Riemannian polyhedra" Trans. Amer. Math. Soc. , 53 (1943) pp. 101–129 |
[6] | J. Eells, "A generalization of the Gauss–Bonnet theorem" Trans. Amer. Math. Soc. , 92 (1959) pp. 142–153 |
[7] | L.S. Pontryagin, "On a connection between homologies and homotopies" Izv. Akad. Nauk SSSR Ser. Mat. , 13 (1949) pp. 193–200 (In Russian) |
Comments
References
[a1] | M. Spivak, "A comprehensive introduction to differential geometry" , 5 , Publish or Perish (1975) pp. 1–5 |
Gauss-Bonnet theorem. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Gauss-Bonnet_theorem&oldid=33081