Table of Contents
Fetching ...

On the Optimal Regularity Implied by the Assumptions of Geometry I: Connections on Tangent Bundles

Moritz Reintjes, Blake Temple

Abstract

We resolve the problem of optimal regularity and Uhlenbeck compactness for affine connections in General Relativity and Mathematical Physics. First, we prove that any affine connection $Γ$, with components $Γ\in L^{2p}$ and components of its Riemann curvature ${\rm Riem}(Γ)$ in $L^p$, in some coordinate system, can be smoothed by coordinate transformation to optimal regularity, $Γ\in W^{1,p}$ (one derivative smoother than the curvature), $p> \max\{n/2,2\}$, dimension $n\geq 2$. For Lorentzian metrics in General Relativity this implies that shock wave solutions of the Einstein-Euler equations are non-singular -- geodesic curves, locally inertial coordinates and the Newtonian limit, all exist in a classical sense, and the Einstein equations hold in the strong sense. The proof is based on an $L^p$ existence theory for the Regularity Transformation (RT) equations, a system of elliptic partial differential equations (introduced by the authors) which determine the Jacobians of the regularizing coordinate transformations. Secondly, this existence theory gives the first extension of Uhlenbeck compactness from Riemannian metrics, to general affine connections bounded in $L^\infty$, with curvature in $L^{p}$, $p>n$, including semi-Riemannian metrics, and Lorentzian metric connections of relativistic Physics. We interpret this as a "geometric" improvement of the generalized Div-Curl Lemma. Our theory shows that Uhlenbeck compactness and optimal regularity are pure logical consequences of the rule which defines how connections transform from one coordinate system to another -- what one could take to be the "starting assumption of geometry".

On the Optimal Regularity Implied by the Assumptions of Geometry I: Connections on Tangent Bundles

Abstract

We resolve the problem of optimal regularity and Uhlenbeck compactness for affine connections in General Relativity and Mathematical Physics. First, we prove that any affine connection , with components and components of its Riemann curvature in , in some coordinate system, can be smoothed by coordinate transformation to optimal regularity, (one derivative smoother than the curvature), , dimension . For Lorentzian metrics in General Relativity this implies that shock wave solutions of the Einstein-Euler equations are non-singular -- geodesic curves, locally inertial coordinates and the Newtonian limit, all exist in a classical sense, and the Einstein equations hold in the strong sense. The proof is based on an existence theory for the Regularity Transformation (RT) equations, a system of elliptic partial differential equations (introduced by the authors) which determine the Jacobians of the regularizing coordinate transformations. Secondly, this existence theory gives the first extension of Uhlenbeck compactness from Riemannian metrics, to general affine connections bounded in , with curvature in , , including semi-Riemannian metrics, and Lorentzian metric connections of relativistic Physics. We interpret this as a "geometric" improvement of the generalized Div-Curl Lemma. Our theory shows that Uhlenbeck compactness and optimal regularity are pure logical consequences of the rule which defines how connections transform from one coordinate system to another -- what one could take to be the "starting assumption of geometry".

Paper Structure

This paper contains 45 sections, 50 theorems, 442 equations.

Key Result

Theorem 3.1

Assume $\Gamma \in L^{2p}(\Omega)$ and $d\Gamma \in L^{p}(\Omega)$ in $x$-coordinates, for some $p>\max\{ n/2, 2\}$,The assumption $p>2$ takes effect only in dimensions $n=2$ and $n= 3$, and is only required by our use of the Sobolev embedding Theorem in Section Sec_gauge_lowreg.$n\geq 2$, and let $ Then for any point $q\in \Omega$ there exists a neighborhood $\Omega' \subset \Omega$ of $q$ and a

Theorems & Definitions (88)

  • Theorem 3.1
  • Theorem 3.2
  • Corollary 4.1
  • proof
  • Corollary 4.2
  • Corollary 4.3
  • proof
  • Corollary 4.4
  • Theorem 6.1
  • Corollary 6.2
  • ...and 78 more