# List of formulas in Riemannian geometry

This is a list of formulas encountered in Riemannian geometry.

## Christoffel symbols, covariant derivative

In a smooth coordinate chart, the Christoffel symbols of the first kind are given by

${\displaystyle \Gamma _{kij}={\frac {1}{2}}\left({\frac {\partial }{\partial x^{j}}}g_{ki}+{\frac {\partial }{\partial x^{i}}}g_{kj}-{\frac {\partial }{\partial x^{k}}}g_{ij}\right)={\frac {1}{2}}\left(g_{ki,j}+g_{kj,i}-g_{ij,k}\right)\,,}$

and the Christoffel symbols of the second kind by

{\displaystyle {\begin{aligned}\Gamma ^{m}{}_{ij}&=g^{mk}\Gamma _{kij}\\&={\frac {1}{2}}\,g^{mk}\left({\frac {\partial }{\partial x^{j}}}g_{ki}+{\frac {\partial }{\partial x^{i}}}g_{kj}-{\frac {\partial }{\partial x^{k}}}g_{ij}\right)={\frac {1}{2}}\,g^{mk}\left(g_{ki,j}+g_{kj,i}-g_{ij,k}\right)\,.\end{aligned}}}

Here ${\displaystyle g^{ij}}$  is the inverse matrix to the metric tensor ${\displaystyle g_{ij}}$ . In other words,

${\displaystyle \delta ^{i}{}_{j}=g^{ik}g_{kj}}$

and thus

${\displaystyle n=\delta ^{i}{}_{i}=g^{i}{}_{i}=g^{ij}g_{ij}}$

is the dimension of the manifold.

Christoffel symbols satisfy the symmetry relations

${\displaystyle \Gamma _{kij}=\Gamma _{kji}}$  or, respectively, ${\displaystyle \Gamma ^{i}{}_{jk}=\Gamma ^{i}{}_{kj}}$ ,

the second of which is equivalent to the torsion-freeness of the Levi-Civita connection.

The contracting relations on the Christoffel symbols are given by

${\displaystyle \Gamma ^{i}{}_{ki}={\frac {1}{2}}g^{im}{\frac {\partial g_{im}}{\partial x^{k}}}={\frac {1}{2g}}{\frac {\partial g}{\partial x^{k}}}={\frac {\partial \log {\sqrt {|g|}}}{\partial x^{k}}}}$

and

${\displaystyle g^{k\ell }\Gamma ^{i}{}_{k\ell }={\frac {-1}{\sqrt {|g|}}}\;{\frac {\partial \left({\sqrt {|g|}}\,g^{ik}\right)}{\partial x^{k}}}}$

where |g| is the absolute value of the determinant of the metric tensor ${\displaystyle g_{ik}}$ . These are useful when dealing with divergences and Laplacians (see below).

The covariant derivative of a vector field with components ${\displaystyle v^{i}}$  is given by:

${\displaystyle v^{i}{}_{;j}=(\nabla _{j}v)^{i}={\frac {\partial v^{i}}{\partial x^{j}}}+\Gamma ^{i}{}_{jk}v^{k}}$

and similarly the covariant derivative of a ${\displaystyle (0,1)}$ -tensor field with components ${\displaystyle v_{i}}$  is given by:

${\displaystyle v_{i;j}=(\nabla _{j}v)_{i}={\frac {\partial v_{i}}{\partial x^{j}}}-\Gamma ^{k}{}_{ij}v_{k}}$

For a ${\displaystyle (2,0)}$ -tensor field with components ${\displaystyle v^{ij}}$  this becomes

${\displaystyle v^{ij}{}_{;k}=\nabla _{k}v^{ij}={\frac {\partial v^{ij}}{\partial x^{k}}}+\Gamma ^{i}{}_{k\ell }v^{\ell j}+\Gamma ^{j}{}_{k\ell }v^{i\ell }}$

and likewise for tensors with more indices.

The covariant derivative of a function (scalar) ${\displaystyle \phi }$  is just its usual differential:

${\displaystyle \nabla _{i}\phi =\phi _{;i}=\phi _{,i}={\frac {\partial \phi }{\partial x^{i}}}}$

Because the Levi-Civita connection is metric-compatible, the covariant derivatives of metrics vanish,

${\displaystyle (\nabla _{k}g)_{ij}=0,\quad (\nabla _{k}g)^{ij}=0}$

as well as the covariant derivatives of the metric's determinant (and volume element)

${\displaystyle \nabla _{k}{\sqrt {|g|}}=0}$

The geodesic ${\displaystyle X(t)}$  starting at the origin with initial speed ${\displaystyle v^{i}}$  has Taylor expansion in the chart:

${\displaystyle X(t)^{i}=tv^{i}-{\frac {t^{2}}{2}}\Gamma ^{i}{}_{jk}v^{j}v^{k}+O(t^{3})}$

## Curvature tensors

### Riemann curvature tensor

If one defines the curvature operator as ${\displaystyle R(U,V)W=\nabla _{U}\nabla _{V}W-\nabla _{V}\nabla _{U}W-\nabla _{[U,V]}W}$  and the coordinate components of the ${\displaystyle (1,3)}$ -Riemann curvature tensor by ${\displaystyle (R(U,V)W)^{\ell }=R^{\ell }{}_{ijk}W^{i}U^{j}V^{k}}$ , then these components are given by:

${\displaystyle R^{\ell }{}_{ijk}={\frac {\partial }{\partial x^{j}}}\Gamma ^{\ell }{}_{ik}-{\frac {\partial }{\partial x^{k}}}\Gamma ^{\ell }{}_{ij}+\Gamma ^{\ell }{}_{js}\Gamma _{ik}^{s}-\Gamma ^{\ell }{}_{ks}\Gamma ^{s}{}_{ij}}$

Lowering indices with ${\displaystyle R_{\ell ijk}=g_{\ell s}R^{s}{}_{ijk}}$  one gets

${\displaystyle R_{ik\ell m}={\frac {1}{2}}\left({\frac {\partial ^{2}g_{im}}{\partial x^{k}\partial x^{\ell }}}+{\frac {\partial ^{2}g_{k\ell }}{\partial x^{i}\partial x^{m}}}-{\frac {\partial ^{2}g_{i\ell }}{\partial x^{k}\partial x^{m}}}-{\frac {\partial ^{2}g_{km}}{\partial x^{i}\partial x^{\ell }}}\right)+g_{np}\left(\Gamma ^{n}{}_{k\ell }\Gamma ^{p}{}_{im}-\Gamma ^{n}{}_{km}\Gamma ^{p}{}_{i\ell }\right).}$

The symmetries of the tensor are

${\displaystyle R_{\mathbf {ik} \ell m}=R_{\ell m\mathbf {ik} }}$  and ${\displaystyle R_{\mathbf {ik} \ell m}=-R_{\mathbf {ki} \ell m}=-R_{\mathbf {ik} m\ell }.}$

That is, it is symmetric in the exchange of the first and last pair of indices, and antisymmetric in the flipping of a pair.

The cyclic permutation sum (sometimes called first Bianchi identity) is

${\displaystyle R_{ik\ell m}+R_{imk\ell }+R_{i\ell mk}=0.}$

The (second) Bianchi identity is

${\displaystyle \nabla _{m}R^{n}{}_{ik\ell }+\nabla _{\ell }R^{n}{}_{imk}+\nabla _{k}R^{n}{}_{i\ell m}=0,}$

that is,

${\displaystyle R^{n}{}_{ik\ell ;m}+R^{n}{}_{imk;\ell }+R^{n}{}_{i\ell m;k}=0}$

which amounts to a cyclic permutation sum of the last three indices, leaving the first two fixed.

### Ricci and scalar curvatures

Ricci and scalar curvatures are contractions of the Riemann tensor. They simplify the Riemann tensor, but contain less information.

The Ricci curvature tensor is essentially the unique (up to sign) nontrivial way of contracting the Riemann tensor:

${\displaystyle R_{ij}=R^{\ell }{}_{i\ell j}=g^{\ell m}R_{i\ell jm}=g^{\ell m}R_{\ell imj}={\frac {\partial \Gamma ^{\ell }{}_{ij}}{\partial x^{\ell }}}-{\frac {\partial \Gamma ^{\ell }{}_{i\ell }}{\partial x^{j}}}+\Gamma ^{m}{}_{ij}\Gamma ^{\ell }{}_{\ell m}-\Gamma ^{m}{}_{i\ell }\Gamma ^{\ell }{}_{jm}.}$

Due to the symmetries of the Riemann tensor, contracting on the 4th instead of the 3rd index yields the same tensor, but with the sign reversed - see sign convention (contracting on the 1st lower index results in an array of zeros). The Ricci tensor ${\displaystyle R_{ij}}$  is symmetric.

By the contracting relations on the Christoffel symbols, we have

${\displaystyle R_{ik}={\frac {\partial \Gamma ^{\ell }{}_{ik}}{\partial x^{\ell }}}-\Gamma ^{m}{}_{i\ell }\Gamma ^{\ell }{}_{km}-\nabla _{k}\left({\frac {\partial }{\partial x^{i}}}\left(\log {\sqrt {|g|}}\right)\right).}$

The scalar curvature is the trace of the Ricci curvature,

${\displaystyle R=g^{ij}R_{ij}=g^{ij}g^{\ell m}R_{i\ell jm}}$ .

The "gradient" of the scalar curvature follows from the Bianchi identity (proof):

${\displaystyle \nabla _{\ell }R^{\ell }{}_{m}={1 \over 2}\nabla _{m}R,}$

that is,

${\displaystyle R^{\ell }{}_{m;\ell }={1 \over 2}R_{;m}.}$

### Einstein tensor

The Einstein tensor Gab is defined in terms of the Ricci tensor Rab and the Ricci scalar R,

${\displaystyle G^{ab}=R^{ab}-{1 \over 2}g^{ab}R}$

where g is the metric tensor.

The Einstein tensor is symmetric, with a vanishing divergence (proof) which is due to the Bianchi identity:

${\displaystyle \nabla _{a}G^{ab}=G^{ab}{}_{;a}=0.}$

### Weyl tensor

The Weyl tensor is given by

${\displaystyle C_{ik\ell m}=R_{ik\ell m}+{\frac {1}{n-2}}\left(R_{im}g_{k\ell }-R_{i\ell }g_{km}+R_{k\ell }g_{im}-R_{km}g_{i\ell }\right)+{\frac {1}{(n-1)(n-2)}}R\left(g_{i\ell }g_{km}-g_{im}g_{k\ell }\right),}$

where ${\displaystyle n}$  denotes the dimension of the Riemannian manifold.

The Weyl tensor satisfies the first (algebraic) Bianchi identity:

${\displaystyle C_{ijkl}+C_{kijl}+C_{jkil}=0.}$

The Weyl tensor is a symmetric product of alternating 2-forms,

${\displaystyle C_{ijkl}=-C_{jikl}\qquad C_{ijkl}=C_{klij},}$

just like the Riemann tensor. Moreover, taking the trace over any two indices gives zero,

${\displaystyle C^{i}{}_{jki}=0}$

The Weyl tensor vanishes (${\displaystyle C=0}$ ) if and only if a manifold ${\displaystyle M}$  of dimension ${\displaystyle n\geq 4}$  is locally conformally flat. In other words, ${\displaystyle M}$  can be covered by coordinate systems in which the metric ${\displaystyle ds^{2}}$  satisfies

${\displaystyle ds^{2}=f^{2}\left(dx_{1}^{2}+dx_{2}^{2}+\ldots dx_{n}^{2}\right)}$

This is essentially because ${\displaystyle C^{i}{}_{jkl}}$  is invariant under conformal changes.

The gradient of a function ${\displaystyle \phi }$  is obtained by raising the index of the differential ${\displaystyle \partial _{i}\phi dx^{i}}$ , whose components are given by:

${\displaystyle \nabla ^{i}\phi =\phi ^{;i}=g^{ik}\phi _{;k}=g^{ik}\phi _{,k}=g^{ik}\partial _{k}\phi =g^{ik}{\frac {\partial \phi }{\partial x^{k}}}}$

The divergence of a vector field with components ${\displaystyle V^{m}}$  is

${\displaystyle \nabla _{m}V^{m}={\frac {\partial V^{m}}{\partial x^{m}}}+V^{k}{\frac {\partial \log {\sqrt {|g|}}}{\partial x^{k}}}={\frac {1}{\sqrt {|g|}}}{\frac {\partial (V^{m}{\sqrt {|g|}})}{\partial x^{m}}}.}$

The Laplace–Beltrami operator acting on a function ${\displaystyle f}$  is given by the divergence of the gradient:

{\displaystyle {\begin{aligned}\Delta f&=\nabla _{i}\nabla ^{i}f={\frac {1}{\sqrt {|g|}}}{\frac {\partial }{\partial x^{j}}}\left(g^{jk}{\sqrt {|g|}}{\frac {\partial f}{\partial x^{k}}}\right)\\&=g^{jk}{\frac {\partial ^{2}f}{\partial x^{j}\partial x^{k}}}+{\frac {\partial g^{jk}}{\partial x^{j}}}{\frac {\partial f}{\partial x^{k}}}+{\frac {1}{2}}g^{jk}g^{il}{\frac {\partial g_{il}}{\partial x^{j}}}{\frac {\partial f}{\partial x^{k}}}=g^{jk}{\frac {\partial ^{2}f}{\partial x^{j}\partial x^{k}}}-g^{jk}\Gamma ^{l}{}_{jk}{\frac {\partial f}{\partial x^{l}}}\end{aligned}}}

The divergence of an antisymmetric tensor field of type ${\displaystyle (2,0)}$  simplifies to

${\displaystyle \nabla _{k}A^{ik}={\frac {1}{\sqrt {|g|}}}{\frac {\partial (A^{ik}{\sqrt {|g|}})}{\partial x^{k}}}.}$

The Hessian of a map ${\displaystyle \phi :M\rightarrow N}$  is given by

${\displaystyle \left(\nabla \left(d\phi \right)\right)_{ij}^{\gamma }={\frac {\partial ^{2}\phi ^{\gamma }}{\partial x^{i}\partial x^{j}}}-^{M}\Gamma ^{k}{}_{ij}{\frac {\partial \phi ^{\gamma }}{\partial x^{k}}}+^{N}\Gamma ^{\gamma }{}_{\alpha \beta }{\frac {\partial \phi ^{\alpha }}{\partial x^{i}}}{\frac {\partial \phi ^{\beta }}{\partial x^{j}}}.}$

## Kulkarni–Nomizu product

The Kulkarni–Nomizu product is an important tool for constructing new tensors from existing tensors on a Riemannian manifold. Let ${\displaystyle A}$  and ${\displaystyle B}$  be symmetric covariant 2-tensors. In coordinates,

${\displaystyle A_{ij}=A_{ji}\qquad \qquad B_{ij}=B_{ji}}$

Then we can multiply these in a sense to get a new covariant 4-tensor, which is often denoted ${\displaystyle A{~\wedge \!\!\!\!\!\!\bigcirc ~}B}$ . The defining formula is

${\displaystyle \left(A{~\wedge \!\!\!\!\!\!\bigcirc ~}B\right)_{ijkl}=A_{ik}B_{jl}+A_{jl}B_{ik}-A_{il}B_{jk}-A_{jk}B_{il}}$

Clearly, the product satisfies

${\displaystyle A{~\wedge \!\!\!\!\!\!\bigcirc ~}B=B{~\wedge \!\!\!\!\!\!\bigcirc ~}A}$

## In an inertial frame

An orthonormal inertial frame is a coordinate chart such that, at the origin, one has the relations ${\displaystyle g_{ij}=\delta _{ij}}$  and ${\displaystyle \Gamma ^{i}{}_{jk}=0}$  (but these may not hold at other points in the frame). These coordinates are also called normal coordinates. In such a frame, the expression for several operators is simpler. Note that the formulae given below are valid at the origin of the frame only.

${\displaystyle R_{ik\ell m}={\frac {1}{2}}\left({\frac {\partial ^{2}g_{im}}{\partial x^{k}\partial x^{\ell }}}+{\frac {\partial ^{2}g_{k\ell }}{\partial x^{i}\partial x^{m}}}-{\frac {\partial ^{2}g_{i\ell }}{\partial x^{k}\partial x^{m}}}-{\frac {\partial ^{2}g_{km}}{\partial x^{i}\partial x^{\ell }}}\right)}$
${\displaystyle R^{\ell }{}_{ijk}={\frac {\partial }{\partial x^{j}}}\Gamma ^{\ell }{}_{ik}-{\frac {\partial }{\partial x^{k}}}\Gamma ^{\ell }{}_{ij}}$

## Under a conformal change

Let ${\displaystyle g}$  be a Riemannian metric on a smooth manifold ${\displaystyle M}$ , and ${\displaystyle \varphi }$  a smooth real-valued function on ${\displaystyle M}$ . Then

${\displaystyle {\tilde {g}}=e^{2\varphi }g}$

is also a Riemannian metric on ${\displaystyle M}$ . We say that ${\displaystyle {\tilde {g}}}$  is conformal to ${\displaystyle g}$ . Evidently, conformality of metrics is an equivalence relation. Here are some formulas for conformal changes in tensors associated with the metric. (Quantities marked with a tilde will be associated with ${\displaystyle {\tilde {g}}}$ , while those unmarked with such will be associated with ${\displaystyle g}$ .)

${\displaystyle {\tilde {g}}_{ij}=e^{2\varphi }g_{ij}}$
${\displaystyle {\tilde {\Gamma }}^{k}{}_{ij}=\Gamma ^{k}{}_{ij}+\delta _{i}^{k}\partial _{j}\varphi +\delta _{j}^{k}\partial _{i}\varphi -g_{ij}\nabla ^{k}\varphi }$

Note that the difference between the Christoffel symbols of two different metrics always form the components of a tensor.

We can also write this in a coordinate-free manner as

${\displaystyle {\tilde {\nabla }}_{F_{*}X}F_{*}Y=F_{*}{\Bigl (}\nabla _{X}Y+X(\varphi )Y+Y(\varphi )X-g(X,Y)\operatorname {grad} \varphi {\Bigr )}}$

whenever ${\displaystyle F:M\to N}$  is locally a conformal diffeomorphism, i.e. ${\displaystyle F^{*}{\tilde {g}}=e^{2\varphi }g}$  (${\displaystyle F=\operatorname {id} }$  in our case), and ${\displaystyle X,Y}$  are vector fields.

${\displaystyle d{\tilde {V}}=e^{n\varphi }dV}$

Here ${\displaystyle dV}$  is the Riemannian volume element.

${\displaystyle {\tilde {R}}_{ijkl}=e^{2\varphi }\left(R_{ijkl}-\left[g{~\wedge \!\!\!\!\!\!\bigcirc ~}\left(\nabla \partial \varphi -\partial \varphi \partial \varphi +{\frac {1}{2}}\|\nabla \varphi \|^{2}g\right)\right]_{ijkl}\right)}$

Here ${\displaystyle {~\wedge \!\!\!\!\!\!\bigcirc ~}}$  is the Kulkarni–Nomizu product defined earlier in this article. The symbol ${\displaystyle \partial _{k}}$  denotes partial derivative, while ${\displaystyle \nabla _{k}}$  denotes covariant derivative. From this formula and the orthogonal decomposition of the Riemann tensor, it follows that

${\displaystyle {\tilde {C}}_{ijkl}=e^{2\varphi }C_{ijkl}}$

and so we see that the (1,3) Weyl tensor is invariant under conformal changes:

${\displaystyle {\tilde {C}}^{i}{}_{jkl}=C^{i}{}_{jkl}}$
${\displaystyle {\tilde {R}}_{ij}=R_{ij}-(n-2)\left[\nabla _{i}\partial _{j}\varphi -(\partial _{i}\varphi )(\partial _{j}\varphi )\right]+\left(\triangle \varphi -(n-2)\|\nabla \varphi \|^{2}\right)g_{ij}}$

Beware that here the Laplacian ${\displaystyle \triangle }$  is minus the trace of the Hessian on functions,

${\displaystyle \triangle f=-\nabla ^{i}\partial _{i}f}$

Thus the operator ${\displaystyle -\triangle }$  is elliptic because the metric ${\displaystyle g}$  is Riemannian.

${\displaystyle {\tilde {\triangle }}f=e^{-2\varphi }\left(\triangle f-(n-2)\nabla ^{k}\varphi \nabla _{k}f\right)}$
${\displaystyle {\tilde {R}}=e^{-2\varphi }\left(R+2(n-1)\triangle \varphi -(n-2)(n-1)\|\nabla \varphi \|^{2}\right)}$

If the dimension ${\displaystyle n>2}$ , then the formula for ${\displaystyle {\tilde {R}}}$  simplifies to

${\displaystyle {\tilde {R}}=e^{-2\varphi }\left[R+{\frac {4(n-1)}{(n-2)}}e^{-(n-2)\varphi /2}\triangle \left(e^{(n-2)\varphi /2}\right)\right]}$

Given an immersed hypersurface ${\displaystyle N}$  with a local unit normal vector field ${\displaystyle \nu }$ , its second fundamental form ${\displaystyle \mathrm {I\!I} }$  (defined by ${\displaystyle \mathrm {I\!I} (X,Y)\nu =(\nabla _{X}^{M}Y)^{\perp }}$  for local vector fields ${\displaystyle X,Y}$  on ${\displaystyle N}$ ) becomes

${\displaystyle {\tilde {\mathrm {I\!I} }}=e^{\varphi }\mathrm {I\!I} -e^{\varphi }\nu (\varphi )g}$

while its mean curvature ${\displaystyle H}$  (i.e. the trace of ${\displaystyle \mathrm {I\!I} }$  with respect to the induced metric on ${\displaystyle N}$ , divided by ${\displaystyle n-1}$ ) changes as

${\displaystyle {\tilde {H}}=e^{-\varphi }H-e^{-\varphi }\nu (\varphi )}$

Let ${\displaystyle \omega }$  be a differential ${\displaystyle p}$ -form. Let ${\displaystyle *}$  be the Hodge star, and ${\displaystyle \delta }$  the codifferential. Under a conformal change, these satisfy

${\displaystyle {\tilde {*}}=e^{(n-2p)\varphi }*}$
${\displaystyle \left[{\tilde {\delta }}\omega \right](v_{1},v_{2},\ldots ,v_{p-1})=e^{-2\varphi }\left[\delta \omega -(n-2p)\omega \left(\nabla \varphi ,v_{1},v_{2},\ldots ,v_{p-1}\right)\right]}$