6.1. Differential Geometry#
References
Math basics: todo check and complete the chapters about tensor calculus and differential geometry
Basics
Let \(\mathbf{X}(q^k)\) a parametrization of points in a \(n\)-dimensional space, \(k = 1:n\).
Natural basis, \(\mathbf{b}_k := \frac{\partial \mathbf{X}}{\partial q^k}\). Vectors of the natural basis can be used to write vectors and tensor fields as a linear combination of them, or in their components w.r.t. that basis
Metric tensor, The covariant compoenents are \(\mathbf{b}_k \cdot \mathbf{b}_l =: g_{kl}\). Some properties: \(g_{k}^{\ l} = \delta_{k}^{l}\), \(g_{ab} g^{bc} = \delta_{a}^{c}\).
Contravariant basis, \(\{ \mathbf{b}^k \}\), s.t. \(\mathbf{b}_k \cdot \mathbf{b}^l = \delta_k^l\). It’s easy to prove that \(\mathbf{b}_k = g_{kl} \mathbf{b}^l\) (just scalar product with \(\mathbf{b}^j\)…), and the law for raising or lower indices holds, \(A^{ij} = g^{ik} A_{k}^{\ j}\),…
Derivatives of the basis vectors. The components of the derivative \(\frac{\partial \mathbf{b}_i}{\partial q^k} = \Gamma_{ik}^l\mathbf{b}_l\) are defined as the Christoffel symbols of the second type. As these are second-order derivatives, for Schwartz theorem about mixed partial derivatives, the symmetry \(\Gamma_{ij}^k = \Gamma_{ji}^k\) immediately follows. The derivative of the vectors of the contravariant basis follows from
and thus \(\partial_k \mathbf{b}^m = - \Gamma_{kl}^{m} \mathbf{b}^l\).
Derivatives of the metric tensor. Using the definition of the covariant components of the metric tensor and the derivatives of the vectors of the natural basis, it’s easy to prove
Using the property \(g_{ab} g^{bc} = \delta_{a^c}\), from its derivatives \(\partial_d\),
it follows that \(\partial_d g^{ec} = - g^{ea} \partial_d g_{ab} g^{bc}\).
Relations between the Christoffel symbols and the derivatives of the metric tensor.
so that, summing the first 2 equations and subracting the last one,
6.1.1. Gradient, covariant derivative and directional derivative#
Gradient of a scalar field
Gradient of a vector field
Proof
Gradient of a 2-nd order tensor field
Proof
…
6.1.2. Curvature Tensor#
todo. Meaning of this definition
…
Definition through the action on an arbitrary vector field, whose components are
First term
\(\ \nabla_{\sigma \eta} v_{\nu} - \nabla_{\eta \sigma} v_{\nu} \ \)
Exploiting symmetry properties of Christoffel symbols,
todo!!! change indices to match the definition of the curvature tensor
Thus, it follows that
6.1.2.1. Ricci’s tensor#
todo Meaning
Ricci’s tensor is defined as the contraction of the first and third indices of the curvature tensor with the metric tensor,
6.1.2.2. Curvature scalar#
todo Meaning
6.1.2.3. Properties#
\(R_{abcd} = - R_{abdc}\)
Proof
By direct inspection of the expression of the components of the curvature tensor,
and the symmetry of the Christoffel symbols, \(\Gamma_{ab}^c = \Gamma_{ba}^c\).
\(R_{abcd} = - R_{bacd}\)
\(R_{abcd} + R_{acdb} + R_{adbc} = 0\)
\(R_{abcd} = R_{cdab}\)
\(R_{abcd;e} + R_{abde;c} + R_{abec;d} = 0\) (Second Bianchi identity)
Proof
todo