Canonical: https://knowlpedia.com/differential-geometry/riemannian-manifold/

# Riemannian manifold

A smooth manifold equipped with a smoothly varying positive-definite inner product on its tangent spaces.

Let \(M\) be a finite-dimensional [smooth manifold](https://knowlpedia.com/fiber-bundles/smooth-manifold/). A **Riemannian metric** on \(M\) is a smooth symmetric covariant \(2\)-tensor \(g\) such that \(g_p\) is a positive-definite [inner product](https://knowlpedia.com/linear-algebra/inner-product/) on \(T_pM\) for every \(p\in M\). Equivalently, \(g\) is a [bundle metric](https://knowlpedia.com/fiber-bundles/bundle-metric/) on the [tangent bundle](https://knowlpedia.com/fiber-bundles/tangent-bundle/) \(TM\). A **Riemannian manifold** is a pair \((M,g)\). In local coordinates, \(g=g_{ij}\,dx^i\otimes dx^j\), where the matrix \((g_{ij}(p))\) is symmetric positive definite at every point and each coefficient \(g_{ij}\) is smooth.

## Metric and volume constructions

The metric assigns each tangent vector a length \(\lVert v\rVert_g=\sqrt{g(v,v)}\) and each piecewise smooth curve \(\gamma\) a length
\[
L_g(\gamma)=\int \sqrt{g_{\gamma(t)}(\dot\gamma(t),\dot\gamma(t))}\,dt.
\]
Taking the infimum of curve lengths gives the Riemannian distance on each [connected component](https://knowlpedia.com/topology/connected-component/). The metric also identifies \(TM\) with the [cotangent bundle](https://knowlpedia.com/fiber-bundles/cotangent-bundle/) and determines a canonical volume density; an orientation turns this density into a [volume form](https://knowlpedia.com/differential-geometry/volume-form/).

## Canonical connection and curvature

Every Riemannian metric has a unique torsion-free, metric-compatible connection, the [Levi–Civita connection](https://knowlpedia.com/fiber-bundles/levicivita-connection-connection/). It determines geodesics, parallel transport, and [curvature](https://knowlpedia.com/fiber-bundles/curvature/). These structures depend on derivatives of \(g\), whereas lengths and angles are pointwise data.

## Examples and scope

The Euclidean metric \(\sum_i dx^i\otimes dx^i\) is the standard example on \(\mathbb R^n\). An immersion into [Euclidean space](https://knowlpedia.com/linear-algebra/euclidean-space/) pulls this metric back to a Riemannian metric when its differential is injective. Positive definiteness distinguishes Riemannian metrics from pseudo-Riemannian metrics, whose nondegenerate symmetric forms may have mixed signature. Smoothness is essential: an arbitrary choice of an inner product on each [tangent space](https://knowlpedia.com/differential-geometry/tangent-space/) does not necessarily define a Riemannian metric.

## References

1. John M. Lee, *Introduction to Riemannian Manifolds*, 2nd ed., Graduate Texts in Mathematics 176, Springer, 2018. [Publisher record](https://doi.org/10.1007/978-3-319-91755-9). Relevant: Chapters 2–5.
2. Manfredo P. do Carmo, *Riemannian Geometry*, Mathematics: Theory & Applications, Birkhäuser, 1992. [Publisher record](https://doi.org/10.1007/978-1-4757-2201-7). Relevant: Chapters 0–3.
