ConceptComplete

Spectral Sequences - Examples and Constructions

Classical spectral sequences appear throughout mathematics with wide-ranging applications.

ExampleSerre Spectral Sequence

For a fibration F→E→BF \to E \to B with BB simply connected and field coefficients kk: E2p,q=Hp(B;Hq(F;k))⇒Hp+q(E;k)E_2^{p,q} = H^p(B; H^q(F; k)) \Rightarrow H^{p+q}(E; k)

This computes cohomology of the total space from the base and fiber.

ExampleLeray Spectral Sequence

For a continuous map f:Xβ†’Yf: X \to Y and sheaf F\mathcal{F} on XX: E2p,q=Hp(Y,Rqfβˆ—F)β‡’Hp+q(X,F)E_2^{p,q} = H^p(Y, R^qf_*\mathcal{F}) \Rightarrow H^{p+q}(X, \mathcal{F})

This relates sheaf cohomology on XX to derived push-forward on YY.

ExampleGrothendieck Spectral Sequence

For composable functors F:Aβ†’BF: \mathcal{A} \to \mathcal{B} and G:Bβ†’CG: \mathcal{B} \to \mathcal{C}, if FF sends injectives to GG-acyclic objects: E2p,q=RpG(RqF(M))β‡’Rp+q(G∘F)(M)E_2^{p,q} = R^pG(R^qF(M)) \Rightarrow R^{p+q}(G \circ F)(M)

This computes derived functors of compositions.

ExampleLyndon-Hochschild-Serre Spectral Sequence

For a group extension 1→N→G→Q→11 \to N \to G \to Q \to 1 and GG-module MM: E2p,q=Hp(Q,Hq(N,M))⇒Hp+q(G,M)E_2^{p,q} = H^p(Q, H^q(N, M)) \Rightarrow H^{p+q}(G, M)

This is the Grothendieck spectral sequence specialized to group cohomology.

Definition7.10Hypercohomology

For a complex of sheaves Fβˆ™\mathcal{F}^\bullet on XX: Hn(X,Fβˆ™)=RnΞ“(X,Fβˆ™)\mathbb{H}^n(X, \mathcal{F}^\bullet) = R^n\Gamma(X, \mathcal{F}^\bullet)

Computed via the Čech-to-derived spectral sequence: E1p,q=HΛ‡p(X,Fq)β‡’Hp+q(X,Fβˆ™)E_1^{p,q} = \check{H}^p(X, \mathcal{F}^q) \Rightarrow \mathbb{H}^{p+q}(X, \mathcal{F}^\bullet)

ExampleEilenberg-Moore Spectral Sequence

For a pullback square of spaces, the Eilenberg-Moore spectral sequence computes cohomology of the pullback from Tor groups: E2p,q=Torp,qHβˆ—(B)(Hβˆ—(X),Hβˆ—(Y))β‡’Hp+q(XΓ—BY)E_2^{p,q} = \text{Tor}^{H^*(B)}_{p,q}(H^*(X), H^*(Y)) \Rightarrow H^{p+q}(X \times_B Y)

Remark

Spectral sequences provide a systematic framework for iterative computation, allowing us to build up global information from local data through successive approximations.