The Stacks project

Lemma 59.73.14. Let $X$ be a quasi-compact and quasi-separated scheme. Let $\eta \in X$ be a generic point of an irreducible component of $X$.

  1. Let $\mathcal{F}$ be a torsion abelian sheaf on $X_{\acute{e}tale}$ whose stalk $\mathcal{F}_{\overline{\eta }}$ is zero. Then $\mathcal{F} = \mathop{\mathrm{colim}}\nolimits \mathcal{F}_ i$ is a filtered colimit of constructible abelian sheaves $\mathcal{F}_ i$ such that for each $i$ the support of $\mathcal{F}_ i$ is contained in a closed subscheme not containing $\eta $.

  2. Let $\Lambda $ be a Noetherian ring and $\mathcal{F}$ a sheaf of $\Lambda $-modules on $X_{\acute{e}tale}$ whose stalk $\mathcal{F}_{\overline{\eta }}$ is zero. Then $\mathcal{F} = \mathop{\mathrm{colim}}\nolimits \mathcal{F}_ i$ is a filtered colimit of constructible sheaves of $\Lambda $-modules $\mathcal{F}_ i$ such that for each $i$ the support of $\mathcal{F}_ i$ is contained in a closed subscheme not containing $\eta $.

Proof. Proof of (1). We can write $\mathcal{F} = \mathop{\mathrm{colim}}\nolimits _{i \in I} \mathcal{F}_ i$ with $\mathcal{F}_ i$ constructible abelian by Lemma 59.73.2. Choose $i \in I$. Since $\mathcal{F}|_\eta $ is zero by assumption, we see that there exists an $i'(i) \geq i$ such that $\mathcal{F}_ i|_\eta \to \mathcal{F}_{i'(i)}|_\eta $ is zero, see Lemma 59.71.8. Then $\mathcal{G}_ i = \mathop{\mathrm{Im}}(\mathcal{F}_ i \to \mathcal{F}_{i'(i)})$ is a constructible abelian sheaf (Lemma 59.71.6) whose stalk at $\eta $ is zero. Hence the support $E_ i$ of $\mathcal{G}_ i$ is a constructible subset of $X$ not containing $\eta $. Since $\eta $ is a generic point of an irreducible component of $X$, we see that $\eta \not\in Z_ i = \overline{E_ i}$ by Topology, Lemma 5.15.15. Define a new directed set $I'$ by using the set $I$ with ordering defined by the rule $i_1$ is bigger or equal to $i_2$ if and only if $i_1 \geq i'(i_2)$. Then the sheaves $\mathcal{G}_ i$ form a system over $I'$ with colimit $\mathcal{F}$ and the proof is complete.

The proof in case (2) is exactly the same and we omit it. $\square$


Comments (0)


Post a comment

Your email address will not be published.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.




In order to prevent bots from posting comments, we would like you to prove that you are human. You can do this by filling in the name of the current tag in the following input field. As a reminder, this is tag 0F0M. Beware of the difference between the letter 'O' and the digit '0'.