Integration on manifolds

Intuition [Beginner]

Integration on a curved space means adding tiny local contributions without losing track of sign. On a flat rectangle, the coordinate grid already tells you how to add pieces. On a manifold, many overlapping coordinate patches are needed.

Orientation is the agreement rule. It tells each patch which direction counts as positive, so local measurements add into one global number.

The basic idea is ordinary area or volume with two upgrades: the space may be curved, and the quantity being added may carry a direction-sensitive sign.

Visual [Beginner]

Each coordinate patch contributes a signed local piece. The orientation makes those signs compatible across overlaps.

Without a consistent orientation, local signed measurements cannot be assembled into one global signed integral.

Worked example [Beginner]

Take the unit sphere. Its surface area can be computed by cutting it into two coordinate regions, one avoiding the north pole and one avoiding the south pole.

Each region is measured using flat coordinates. Where the regions overlap, the two measurements agree because the coordinate change preserves the chosen orientation.

What this tells us: integration on manifolds is local calculation plus a global rule for consistent signs.

Check your understanding [Beginner]

Formal definition [Intermediate+]

Let $M$ be an oriented smooth $n$-manifold. A compactly supported top-degree form $\omega \in {\Omega }_c^n(M)$ can be integrated over $M$.

If $\omega$ is supported in one positively oriented coordinate chart $\phi :U\to {\mathbb{R}}^n$, write

$$\omega =f(x)d{x}^1\wedge \cdots \wedge d{x}^n.$$

Then define

$${\int }_M\omega ={\int }_{\phi (U)}f(x)d{x}^1\cdots d{x}^n.$$

For a general compactly supported form, choose a partition of unity $\{{\rho }_i\}$ subordinate to oriented coordinate charts and set

$${\int }_M\omega =\sum _i{\int }_M{\rho }_i\omega .$$

The change-of-variables theorem ensures that this definition is independent of the chosen oriented atlas and partition of unity ^{[Bott-Tu §I.3]}.

Key theorem with proof [Intermediate+]

Theorem (orientation reversal changes sign). Let $M$ be an oriented $n$-manifold and let $-M$ denote the same manifold with the opposite orientation. For every compactly supported top-degree form $\omega$,

$${\int }_{-M}\omega =-{\int }_M\omega .$$

Proof. It is enough to check the statement on forms supported in a single coordinate chart, because partitions of unity reduce the general case to a finite or locally finite sum of such pieces.

Choose an oriented chart for $M$ in which

$$\omega =f(x)d{x}^1\wedge \cdots \wedge d{x}^n.$$

The opposite orientation is represented by any chart with negative Jacobian relative to the original orientation. Under such a coordinate change, the top-degree coordinate form acquires the sign of the Jacobian determinant. Since the determinant sign is negative, the local integral changes by a factor of $-1$.

Applying this to every partition-of-unity term gives

$${\int }_{-M}\omega =-{\int }_M\omega .$$

$\square$

Bridge. The construction here builds toward 03.04.06 (de rham cohomology), where the same data is upgraded, and the symmetry side is taken up in 03.04.08 (variational calculus on manifolds). The defining pattern appears again in those units in a sharpened form, where the local data is glued or quotiented. Putting these together, the foundational insight is that the data of this unit gives the structural signature that the rest of the strand reads off.

Exercises [Intermediate+]

Lean formalization [Intermediate+]

lean_status: none — Mathlib has measure theory and manifolds, but not the top-degree-form integration package in the curriculum form used here.

The formalization path requires oriented smooth manifolds, top-degree differential forms, partitions of unity, compact support, change of variables, boundary orientations, and Stokes' theorem.

Advanced results [Master]

Stokes' theorem relates integration to exterior differentiation:

$${\int }_{M}d\eta ={\int }_{\partial M}\eta$$

for compactly supported $(n-1)$-forms on an oriented manifold with boundary ^{[Spivak §11]}. This theorem identifies exact top-degree forms with boundary contributions and underlies de Rham cohomology 03.04.06.

Integration also gives the pairing

$$H_{\mathrm{dR}}^k(M)\times H_k(M;\mathbb{R})\to \mathbb{R},$$

obtained by integrating closed forms over cycles. This pairing is the analytic entry point to de Rham's theorem and to characteristic-number computations.

Synthesis. This construction generalises the pattern fixed in 03.02.01 (smooth manifold), with the symmetric data replaced by its skew or twisted analogue. Read in the opposite direction, the construction is dual to the metric story: complements and orthogonality are taken with respect to the bilinear datum of this unit, not a metric, and the resulting category of subobjects is the one the rest of the strand classifies. The central insight is that this datum identifies algebra with geometry: functions become vector fields, subspaces become quotients, and invariants become cohomology classes — and that identification is the engine driving every theorem downstream.

Full proof set [Master]

Independence of partition of unity. Let $\{{\rho }_i\}$ and $\{{\psi }_j\}$ be two partitions of unity subordinate to oriented charts. Then

$$\sum _i\int {\rho }_i\omega =\sum _i\sum _j\int {\rho }_i{\psi }_j\omega =\sum _j\int {\psi }_j\omega .$$

Local finiteness justifies the rearrangement on the compact support of $\omega$, and chart independence follows from change of variables on each overlap.

Boundary sign. The boundary orientation is fixed by the outward-normal-first convention. With this convention, Stokes' theorem on a half-space reduces to the one-variable fundamental theorem of calculus in the normal coordinate.

Connections [Master]

• De Rham cohomology 03.04.06 — integration pairs de Rham classes with cycles.

• Variational calculus 03.04.08 — action functionals integrate Lagrangian densities over manifolds.

• Yang-Mills action 03.07.05 — the Yang-Mills functional integrates curvature energy over the base manifold.

• Chern-Weil homomorphism 03.06.06 — characteristic numbers are obtained by integrating characteristic forms.

Historical & philosophical context [Master]

The invariant integration of differential forms grew from the change-of-variables theorem and the coordinate-free development of exterior calculus. Stokes' theorem unified the fundamental theorem of calculus, Green's theorem, Gauss' theorem, and the classical Stokes theorem.

de Rham's work identified integration of closed forms over cycles as a bridge between differential forms and topology. Bott and Tu use this pairing as the computational core of de Rham cohomology ^{[fast-track Bott-Tu-Differential-Forms-in-AlgTop-645x1024__a28a653851.jpg]}.

Bibliography [Master]

• Spivak, M., A Comprehensive Introduction to Differential Geometry, Vol. I, Publish or Perish, 1970. §11.
• Bott, R. and Tu, L. W., Differential Forms in Algebraic Topology, Springer, 1982. §I.3.
• Madsen, I. and Tornehave, J., From Calculus to Cohomology, Cambridge University Press, 1997. §10.
• de Rham, G., Variétés différentiables, Hermann, 1955.

---

Wave 2 Phase 2.4 unit #4. Produced as the integration prerequisite for variational calculus and Yang-Mills.