5
$\begingroup$

I'm reading "The Geometry of Moduli Spaces of Sheaves" by Huybrechts and Lehn, and I am stuck with the construction of the Grassmann variety (Example 2.2.2). I'd appreciate any help, so here goes:

Let V be a be a finite dimensional vector space over a field $k$. Let $0 \leq r \leq dim(V)$. Define the Grassmann functor as $\underline{Grass}(V,r):(Sch/k)^o \rightarrow (Sets)$ associating to any $k$-scheme $S$ of finite type the set of all subsheaves $K \subset \mathcal{O}_S \otimes V$ with locally free quotient $ F = \mathcal{O}_S \otimes V/K$ of rank $r$.

For each $r$-dimensional linear subspace $W \subset V$ they consider a subfunctor $\mathcal{G}_W$ of $\underline{Grass}(V,r)$, associating to each $k$-scheme S those locally free quotients $F$ for which the composition $\mathcal{O}_S \otimes W \rightarrow \mathcal{O}_S \otimes V \rightarrow F$ is an isomorphism, giving a splitting of the inclusion $W \subset V$. Now they make the unclear jump to saying that from this one can conclude that $\mathcal{G}_W$ is represented by $\text{Spec}(S^*\text{Hom}(V,W)^{\vee})$, "corresponding to homomorphisms that split the inclusion map $W \subset V$". Now, if I am interpreting this correctly $\text{Spec}(S^*\text{Hom}(V,W)^{\vee})$ is an affine scheme over the vector space of polynomial functions on the vector space $\text{Hom}(V,W)$. But I don't see how an element of $\mathcal{G}_W(S)$ gives a map of schemes $S \rightarrow \text{Spec}(S^*\text{Hom}(V,W)^{\vee})$. Additionally, what is the reverse map, given such a morphism of schemes $S \rightarrow \text{Spec}(S^*\text{Hom}(V,W)^{\vee})$, how does one obtain a locally free quotient in $\mathcal{G}_W(S)$?

$\endgroup$
1
$\begingroup$

Using the isomorphism $F\cong \mathcal{O}_S\otimes W$ implies that an $S$-point of $\mathcal{G}_W$ gives a map $\mathcal{O}_S\otimes V \to \mathcal{O}_S\otimes W$ which splits the inclusion. This is the desired $S$-point of the scheme $\mathrm{Spec}(S^*\mathrm{Hom}(V,W)^{\vee})$, though we should note that it is not this whole space that represents $\mathcal{G}_W$ but just the affine subspace consisting of splittings.

$\endgroup$

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.