Loading web-font TeX/Main/Regular

The Stacks project

38.28 Grothendieck's Existence Theorem, IV

This section continues the discussion in Cohomology of Schemes, Sections 30.24, 30.25, and 30.27. We will work in the following situation.

Situation 38.28.1. Here we have an inverse system of rings (A_ n) with surjective transition maps whose kernels are locally nilpotent. Set A = \mathop{\mathrm{lim}}\nolimits A_ n. We have a scheme X separated and of finite presentation over A. We set X_ n = X \times _{\mathop{\mathrm{Spec}}(A)} \mathop{\mathrm{Spec}}(A_ n) and we view it as a closed subscheme of X. We assume further given a system (\mathcal{F}_ n, \varphi _ n) where \mathcal{F}_ n is a finitely presented \mathcal{O}_{X_ n}-module, flat over A_ n, with support proper over A_ n, and

\varphi _ n : \mathcal{F}_ n \otimes _{\mathcal{O}_{X_ n}} \mathcal{O}_{X_{n - 1}} \longrightarrow \mathcal{F}_{n - 1}

is an isomorphism (notation using the equivalence of Morphisms, Lemma 29.4.1).

Our goal is to see if we can find a quasi-coherent sheaf \mathcal{F} on X such that \mathcal{F}_ n = \mathcal{F} \otimes _{\mathcal{O}_ X} \mathcal{O}_{X_ n} for all n.

Lemma 38.28.2. In Situation 38.28.1 consider

K = R\mathop{\mathrm{lim}}\nolimits _{D_\mathit{QCoh}(\mathcal{O}_ X)}(\mathcal{F}_ n) = DQ_ X(R\mathop{\mathrm{lim}}\nolimits _{D(\mathcal{O}_ X)}\mathcal{F}_ n)

Then K is in D^ b_{\mathit{QCoh}}(\mathcal{O}_ X) and in fact K has nonzero cohomology sheaves only in degrees \geq 0.

Proof. Special case of Derived Categories of Schemes, Example 36.21.5. \square

Lemma 38.28.3. In Situation 38.28.1 let K be as in Lemma 38.28.2. For any perfect object E of D(\mathcal{O}_ X) we have

  1. M = R\Gamma (X, K \otimes ^\mathbf {L} E) is a perfect object of D(A) and there is a canonical isomorphism R\Gamma (X_ n, \mathcal{F}_ n \otimes ^\mathbf {L} E|_{X_ n}) = M \otimes _ A^\mathbf {L} A_ n in D(A_ n),

  2. N = R\mathop{\mathrm{Hom}}\nolimits _ X(E, K) is a perfect object of D(A) and there is a canonical isomorphism R\mathop{\mathrm{Hom}}\nolimits _{X_ n}(E|_{X_ n}, \mathcal{F}_ n) = N \otimes _ A^\mathbf {L} A_ n in D(A_ n).

In both statements E|_{X_ n} denotes the derived pullback of E to X_ n.

Proof. Proof of (2). Write E_ n = E|_{X_ n} and N_ n = R\mathop{\mathrm{Hom}}\nolimits _{X_ n}(E_ n, \mathcal{F}_ n). Recall that R\mathop{\mathrm{Hom}}\nolimits _{X_ n}(-, -) is equal to R\Gamma (X_ n, R\mathop{\mathcal{H}\! \mathit{om}}\nolimits (-, -)), see Cohomology, Section 20.44. Hence by Derived Categories of Schemes, Lemma 36.30.7 we see that N_ n is a perfect object of D(A_ n) whose formation commutes with base change. Thus the maps N_ n \otimes _{A_ n}^\mathbf {L} A_{n - 1} \to N_{n - 1} coming from \varphi _ n are isomorphisms. By More on Algebra, Lemma 15.97.3 we find that R\mathop{\mathrm{lim}}\nolimits N_ n is perfect and that its base change back to A_ n recovers N_ n. On the other hand, the exact functor R\mathop{\mathrm{Hom}}\nolimits _ X(E, -) : D_\mathit{QCoh}(\mathcal{O}_ X) \to D(A) of triangulated categories commutes with products and hence with derived limits, whence

R\mathop{\mathrm{Hom}}\nolimits _ X(E, K) = R\mathop{\mathrm{lim}}\nolimits R\mathop{\mathrm{Hom}}\nolimits _ X(E, \mathcal{F}_ n) = R\mathop{\mathrm{lim}}\nolimits R\mathop{\mathrm{Hom}}\nolimits _ X(E_ n, \mathcal{F}_ n) = R\mathop{\mathrm{lim}}\nolimits N_ n

This proves (2). To see that (1) holds, translate it into (2) using Cohomology, Lemma 20.50.5. \square

Proof. Combinging Lemma 38.28.3 and Derived Categories of Schemes, Lemma 36.34.3 we see that R\Gamma (X, K \otimes ^\mathbf {L} E) is pseudo-coherent in D(A) for all pseudo-coherent E in D(\mathcal{O}_ X). Thus the lemma follows from More on Morphisms, Lemma 37.69.4. \square

Lemma 38.28.5. In Situation 38.28.1 let K be as in Lemma 38.28.2. For any quasi-compact open U \subset X we have

R\Gamma (U, K) \otimes _ A^\mathbf {L} A_ n = R\Gamma (U_ n, \mathcal{F}_ n)

in D(A_ n) where U_ n = U \cap X_ n.

Proof. Fix n. By Derived Categories of Schemes, Lemma 36.33.4 there exists a system of perfect complexes E_ m on X such that R\Gamma (U, K) = \text{hocolim} R\Gamma (X, K \otimes ^\mathbf {L} E_ m). In fact, this formula holds not just for K but for every object of D_\mathit{QCoh}(\mathcal{O}_ X). Applying this to \mathcal{F}_ n we obtain

\begin{align*} R\Gamma (U_ n, \mathcal{F}_ n) & = R\Gamma (U, \mathcal{F}_ n) \\ & = \text{hocolim}_ m R\Gamma (X, \mathcal{F}_ n \otimes ^\mathbf {L} E_ m) \\ & = \text{hocolim}_ m R\Gamma (X_ n, \mathcal{F}_ n \otimes ^\mathbf {L} E_ m|_{X_ n}) \end{align*}

Using Lemma 38.28.3 and the fact that - \otimes _ A^\mathbf {L} A_ n commutes with homotopy colimits we obtain the result. \square

Lemma 38.28.6. In Situation 38.28.1 let K be as in Lemma 38.28.2. Denote X_0 \subset X the closed subset consisting of points lying over the closed subset \mathop{\mathrm{Spec}}(A_1) = \mathop{\mathrm{Spec}}(A_2) = \ldots of \mathop{\mathrm{Spec}}(A). There exists an open W \subset X containing X_0 such that

  1. H^ i(K)|_ W is zero unless i = 0,

  2. \mathcal{F} = H^0(K)|_ W is of finite presentation, and

  3. \mathcal{F}_ n = \mathcal{F} \otimes _{\mathcal{O}_ X} \mathcal{O}_{X_ n}.

Proof. Fix n \geq 1. By construction there is a canonical map K \to \mathcal{F}_ n in D_\mathit{QCoh}(\mathcal{O}_ X) and hence a canonical map H^0(K) \to \mathcal{F}_ n of quasi-coherent sheaves. This explains the meaning of part (3).

Let x \in X_0 be a point. We will find an open neighbourhood W of x such that (1), (2), and (3) are true. Since X_0 is quasi-compact this will prove the lemma. Let U \subset X be an affine open neighbourhood of x. Say U = \mathop{\mathrm{Spec}}(B). Choose a surjection P \to B with P smooth over A. By Lemma 38.28.4 and the definition of relative pseudo-coherence there exists a bounded above complex F^\bullet of finite free P-modules representing Ri_*K where i : U \to \mathop{\mathrm{Spec}}(P) is the closed immersion induced by the presentation. Let M_ n be the B-module corresponding to \mathcal{F}_ n|_ U. By Lemma 38.28.5

H^ i(F^\bullet \otimes _ A A_ n) = \left\{ \begin{matrix} 0 & \text{if} & i \not= 0 \\ M_ n & \text{if} & i = 0 \end{matrix} \right.

Let i be the maximal index such that F^ i is nonzero. If i \leq 0, then (1), (2), and (3) are true. If not, then i > 0 and we see that the rank of the map

F^{i - 1} \to F^ i

in the point x is maximal. Hence in an open neighbourhood of x inside \mathop{\mathrm{Spec}}(P) the rank is maximal. Thus after replacing P by a principal localization we may assume that the displayed map is surjective. Since F^ i is finite free we may choose a splitting F^{i - 1} = F' \oplus F^ i. Then we may replace F^\bullet by the complex

\ldots \to F^{i - 2} \to F' \to 0 \to \ldots

and we win by induction on i. \square

Lemma 38.28.7. In Situation 38.28.1 let K be as in Lemma 38.28.2. Let W \subset X be as in Lemma 38.28.6. Set \mathcal{F} = H^0(K)|_ W. Then, after possibly shrinking the open W, the support of \mathcal{F} is proper over A.

Proof. Fix n \geq 1. Let I_ n = \mathop{\mathrm{Ker}}(A \to A_ n). By More on Algebra, Lemma 15.11.3 the pair (A, I_ n) is henselian. Let Z \subset W be the support of \mathcal{F}. This is a closed subset as \mathcal{F} is of finite presentation. By part (3) of Lemma 38.28.6 we see that Z \times _{\mathop{\mathrm{Spec}}(A)} \mathop{\mathrm{Spec}}(A_ n) is equal to the support of \mathcal{F}_ n and hence proper over \mathop{\mathrm{Spec}}(A/I). By More on Morphisms, Lemma 37.53.9 we can write Z = Z_1 \amalg Z_2 with Z_1, Z_2 open and closed in Z, with Z_1 proper over A, and with Z_1 \times _{\mathop{\mathrm{Spec}}(A)} \mathop{\mathrm{Spec}}(A/I_ n) equal to the support of \mathcal{F}_ n. In other words, Z_2 does not meet X_0. Hence after replacing W by W \setminus Z_2 we obtain the lemma. \square

Lemma 38.28.8. Let A = \mathop{\mathrm{lim}}\nolimits A_ n be a limit of a system of rings whose transition maps are surjective and with locally nilpotent kernels. Let S = \mathop{\mathrm{Spec}}(A). Let T \to S be a monomorphism which is locally of finite type. If \mathop{\mathrm{Spec}}(A_ n) \to S factors through T for all n, then T = S.

Proof. Set S_ n = \mathop{\mathrm{Spec}}(A_ n). Let T_0 \subset T be the common image of the factorizations S_ n \to T. Then T_0 is quasi-compact. Let T' \subset T be a quasi-compact open containing T_0. Then S_ n \to T factors through T'. If we can show that T' = S, then T' = T = S. Hence we may assume T is quasi-compact.

Assume T is quasi-compact. In this case T \to S is separated and quasi-finite (Morphisms, Lemma 29.20.15). Using Zariski's Main Theorem (in the form of More on Morphisms, Lemma 37.43.3) we choose a factorization T \to W \to S with W \to S finite and T \to W an open immersion. Write W = \mathop{\mathrm{Spec}}(B). The (unique) factorizations S_ n \to T may be viewed as morphisms into W and we obtain

A \longrightarrow B \longrightarrow \mathop{\mathrm{lim}}\nolimits A_ n = A

Consider the morphism h : S = \mathop{\mathrm{Spec}}(A) \to \mathop{\mathrm{Spec}}(B) = W coming from the arrow on the right. Then

T \times _{W, h} S

is an open subscheme of S containing the image of S_ n \to S for all n. To finish the proof it suffices to show that any open U \subset S containing the image of S_ n \to S for some n \geq 1 is equal to S. This is true because (A, \mathop{\mathrm{Ker}}(A \to A_ n)) is a henselian pair (More on Algebra, Lemma 15.11.3) and hence every closed point of S is contained in the image of S_ n \to S. \square

Theorem 38.28.9 (Grothendieck Existence Theorem). In Situation 38.28.1 there exists a finitely presented \mathcal{O}_ X-module \mathcal{F}, flat over A, with support proper over A, such that \mathcal{F}_ n = \mathcal{F} \otimes _{\mathcal{O}_ X} \mathcal{O}_{X_ n} for all n compatibly with the maps \varphi _ n.

Proof. Apply Lemmas 38.28.2, 38.28.3, 38.28.4, 38.28.5, 38.28.6, and 38.28.7 to get an open subscheme W \subset X containing all points lying over \mathop{\mathrm{Spec}}(A_ n) and a finitely presented \mathcal{O}_ W-module \mathcal{F} whose support is proper over A with \mathcal{F}_ n = \mathcal{F} \otimes _{\mathcal{O}_ W} \mathcal{O}_{X_ n} for all n \geq 1. (This makes sense as X_ n \subset W.) By Lemma 38.17.1 we see that \mathcal{F} is universally pure relative to \mathop{\mathrm{Spec}}(A). By Theorem 38.27.5 (for explanation, see Lemma 38.27.6) there exists a universal flattening S' \to \mathop{\mathrm{Spec}}(A) of \mathcal{F} and moreover the morphism S' \to \mathop{\mathrm{Spec}}(A) is a monomorphism of finite presentation. Since the base change of \mathcal{F} to \mathop{\mathrm{Spec}}(A_ n) is \mathcal{F}_ n we find that \mathop{\mathrm{Spec}}(A_ n) \to \mathop{\mathrm{Spec}}(A) factors (uniquely) through S' for each n. By Lemma 38.28.8 we see that S' = \mathop{\mathrm{Spec}}(A). This means that \mathcal{F} is flat over A. Finally, since the scheme theoretic support Z of \mathcal{F} is proper over \mathop{\mathrm{Spec}}(A), the morphism Z \to X is closed. Hence the pushforward (W \to X)_*\mathcal{F} is supported on W and has all the desired properties. \square


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.