3
$\begingroup$

G\"ortz and Wedhorn, in their book ``Algebraic Geometry I'' at page 84, give the following definitions. Let $(X, \mathscr{O}_X)$ be a scheme.

(1) A closed subscheme of $X$ is given by a closed subset $Z \subseteq X$ (let $i : Z \longrightarrow X$ be the inclusion) and a sheaf $\mathscr{O}_Z$ on $Z$, such that $(Z, \mathscr{O}_Z)$ is a scheme, and such that the sheaf $i_*\mathscr{O}_Z$ is isomorphic to $\mathscr{O}_X/\mathscr{J}$ for a sheaf of ideals $\mathscr{J} \subseteq \mathscr{O}_X$.

(2) A morphism $i : Z \longrightarrow X$ of schemes is called a closed immersion, if the underlying continuous map is a homeomorphism between $Z$ and a closed subset of $X$, and the sheaf homomorphism $i^\flat : \mathscr{O}_X \longrightarrow i_*\mathscr{O}_Z$ is surjective.

Then they wrote~: if $Z \subseteq X$ is closed subscheme as in (1), then the morphism $(i, i^\flat)$ is a closed immersion.

I think that the morphism $i^\flat$ is the composition of $\mathscr{O}_X \longrightarrow \mathscr{O}_X/\mathscr{J}$ and the isomorphism $\mathscr{O}_X/\mathscr{J} \simeq i_*\mathscr{O}_Z$. But, I don't see why $(i, i^\flat)$ is a morphism of schemes.

$\endgroup$
1
$\begingroup$

You've done the hard part in understanding the sheaf homorphism $i^{\flat}$. All you need to make the pair an morphism of schemes is check that $i$ is continuous. This is clear because $Z$ is endowed with the subspace topology.

The fact that $Z$ is endowed with the subspace topology is implicit in "... such that $(Z, \mathcal{O}_Z)$ is a scheme ...". A scheme is a locally ringed space, which is a pair of a topological space $Z$ and a sheaf of rings $\mathcal{O}_Z$. That means before $(Z, \mathcal{O}_Z)$ is a conditioned on being a scheme $Z$ is endowed with a topology. The obvious implication is that $Z$ was given the subspace topology.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.