Relation between long exact sequence and associated graded

261 Views Asked by At

I'm reading these notes by Hutchings on spectral sequences. In the first section, he motivates spectral sequences with the long exact sequence in relative homology. Given a chain complex $C_*$ and a subcomplex $F_0C_*$, we form a long exact sequence in homology given by

$$\dotsb \to H_{k}(F_0C_*) \to H_k (C_*) \to H_k(C_*/F_0C_*) \to H_{k-1}(F_0C_*) \to \dotsb$$

Isolating at $H_k(C_*)$, we have a short exact sequence

$$0 \to \operatorname{cok} \delta_{k+1} \to H_k(C) \to \operatorname{ker}\delta_k \to 0$$

He then refers to $\operatorname{cok} \delta_{k+1}$ and $\operatorname{ker} \delta_k$ as $G_0$ and $G_1$ respectively--the associated graded modules of the homology of $C_*$.

It's clear to me why $\operatorname{cok} \delta_{k+1}$ is $G_0$. It's less clear to me why $\operatorname{ker} \delta_k$ is $G_1$. I would describe $\operatorname{ker} \delta_k$ as the subgroup of the relative $k$th homology classes whose boundaries in $F_0C_*$ are also boundaries of elements in $F_0C_*$. I'm not seeing why that is the same thing as $G_1H_k$, which I've learned is those $k$ homology classes modded out by those which have a representative in $F_0C_*$.

Can someone help me understand?

1

There are 1 best solutions below

0
On BEST ANSWER

By exactness of the long exact sequence, $\ker(\delta_k)$ is equal to the image of the map $H_k(C_*)\to H_k(C_*/F_0C_*)$, which is in turn the quotient of $H_k(C_*)$ by the image of $H_k(F_0C_*)\to H_k(C_*)$. This is exactly your description of $G_1$.