42.16 Preparation for principal divisors
Some of the material in this section partially overlaps with the discussion in Divisors, Section 31.26.
Lemma 42.16.1. Let (S, \delta ) be as in Situation 42.7.1. Let X be locally of finite type over S. Assume X is integral.
If Z \subset X is an integral closed subscheme, then the following are equivalent:
Z is a prime divisor,
Z has codimension 1 in X, and
\dim _\delta (Z) = \dim _\delta (X) - 1.
If Z is an irreducible component of an effective Cartier divisor on X, then \dim _\delta (Z) = \dim _\delta (X) - 1.
Proof.
Part (1) follows from the definition of a prime divisor (Divisors, Definition 31.26.2) and the definition of a dimension function (Topology, Definition 5.20.1). Let \xi \in Z be the generic point of an irreducible component Z of an effective Cartier divisor D \subset X. Then \dim (\mathcal{O}_{D, \xi }) = 0 and \mathcal{O}_{D, \xi } = \mathcal{O}_{X, \xi }/(f) for some nonzerodivisor f \in \mathcal{O}_{X, \xi } (Divisors, Lemma 31.15.2). Then \dim (\mathcal{O}_{X, \xi }) = 1 by Algebra, Lemma 10.60.13. Hence Z is as in (1) by Properties, Lemma 28.10.3 and the proof is complete.
\square
Lemma 42.16.2. Let f : X \to Y be a morphism of schemes. Let \xi \in Y be a point. Assume that
X, Y are integral,
Y is locally Noetherian
f is proper, dominant and R(Y) \subset R(X) is finite, and
\dim (\mathcal{O}_{Y, \xi }) = 1.
Then there exists an open neighbourhood V \subset Y of \xi such that f|_{f^{-1}(V)} : f^{-1}(V) \to V is finite.
Proof.
This lemma is a special case of Varieties, Lemma 33.17.2. Here is a direct argument in this case. By Cohomology of Schemes, Lemma 30.21.2 it suffices to prove that f^{-1}(\{ \xi \} ) is finite. We replace Y by an affine open, say Y = \mathop{\mathrm{Spec}}(R). Note that R is Noetherian, as Y is assumed locally Noetherian. Since f is proper it is quasi-compact. Hence we can find a finite affine open covering X = U_1 \cup \ldots \cup U_ n with each U_ i = \mathop{\mathrm{Spec}}(A_ i). Note that R \to A_ i is a finite type injective homomorphism of domains such that the induced extension of fraction fields is finite. Thus the lemma follows from Algebra, Lemma 10.113.2.
\square
Comments (2)
Comment #5092 by Klaus Mattis on
Comment #5301 by Johan on