Loading web-font TeX/Math/Italic

The Stacks project

Lemma 97.5.5. Let S be a scheme. Let \kappa = \text{size}(T) for some T \in \mathop{\mathrm{Ob}}\nolimits ((\mathit{Sch}/S)_{fppf}). Let f : \mathcal{X} \to \mathcal{Y} be a 1-morphism of categories fibred in groupoids over (\mathit{Sch}/S)_{fppf} such that

  1. \mathcal{Y} \to (\mathit{Sch}/S)_{fppf} is limit preserving on objects,

  2. for an affine scheme V locally of finite presentation over S and y \in \mathop{\mathrm{Ob}}\nolimits (\mathcal{Y}_ V) the fibre product (\mathit{Sch}/V)_{fppf} \times _{y, \mathcal{Y}} \mathcal{X} is representable by an algebraic space of size \leq \kappa 1,

  3. \mathcal{X} and \mathcal{Y} are stacks for the Zariski topology.

Then f is representable by algebraic spaces.

Proof. Let V be a scheme over S and y \in \mathcal{Y}_ V. We have to prove (\mathit{Sch}/V)_{fppf} \times _{y, \mathcal{Y}} \mathcal{X} is representable by an algebraic space.

Case I: V is affine and maps into an affine open \mathop{\mathrm{Spec}}(\Lambda ) \subset S. Then we can write V = \mathop{\mathrm{lim}}\nolimits V_ i with each V_ i affine and of finite presentation over \mathop{\mathrm{Spec}}(\Lambda ), see Algebra, Lemma 10.127.2. Then y comes from an object y_ i over V_ i for some i by assumption (1). By assumption (3) the fibre product (\mathit{Sch}/V_ i)_{fppf} \times _{y_ i, \mathcal{Y}} \mathcal{X} is representable by an algebraic space Z_ i. Then (\mathit{Sch}/V)_{fppf} \times _{y, \mathcal{Y}} \mathcal{X} is representable by Z \times _{V_ i} V.

Case II: V is general. Choose an affine open covering V = \bigcup _{i \in I} V_ i such that each V_ i maps into an affine open of S. We first claim that \mathcal{Z} = (\mathit{Sch}/V)_{fppf} \times _{y, \mathcal{Y}} \mathcal{X} is a stack in setoids for the Zariski topology. Namely, it is a stack in groupoids for the Zariski topology by Stacks, Lemma 8.5.6. Then suppose that z is an object of \mathcal{Z} over a scheme T. Denote g : T \to V the morphism corresponding to the projection of z in (\mathit{Sch}/V)_{fppf}. Consider the Zariski sheaf \mathit{I} = \mathit{Isom}_{\mathcal{Z}}(z, z). By Case I we see that \mathit{I}|_{g^{-1}(V_ i)} = * (the singleton sheaf). Hence \mathcal{I} = *. Thus \mathcal{Z} is fibred in setoids. To finish the proof we have to show that the Zariski sheaf Z : T \mapsto \mathop{\mathrm{Ob}}\nolimits (\mathcal{Z}_ T)/\cong is an algebraic space, see Algebraic Stacks, Lemma 94.8.2. There is a map p : Z \to V (transformation of functors) and by Case I we know that Z_ i = p^{-1}(V_ i) is an algebraic space. The morphisms Z_ i \to Z are representable by open immersions and \coprod Z_ i \to Z is surjective (in the Zariski topology). Hence Z is a sheaf for the fppf topology by Bootstrap, Lemma 80.3.11. Thus Spaces, Lemma 65.8.5 applies and we conclude that Z is an algebraic space2. \square

[1] The condition on size can be dropped by those ignoring set theoretic issues.
[2] To see that the set theoretic condition of that lemma is satisfied we argue as follows: First choose the open covering such that |I| \leq \text{size}(V). Next, choose schemes U_ i of size \leq \max (\kappa , \text{size}(V)) and surjective étale morphisms U_ i \to Z_ i; we can do this by assumption (2) and Sets, Lemma 3.9.6 (details omitted). Then Sets, Lemma 3.9.9 implies that \coprod U_ i is an object of (\mathit{Sch}/S)_{fppf}. Hence \coprod Z_ i is an algebraic space by Spaces, Lemma 65.8.4.

Comments (0)


Your email address will not be published. Required fields are marked.

In your comment you can use Markdown and LaTeX style mathematics (enclose it like $\pi$). A preview option is available if you wish to see how it works out (just click on the eye in the toolbar).

Unfortunately JavaScript is disabled in your browser, so the comment preview function will not work.

All contributions are licensed under the GNU Free Documentation License.