1
$\begingroup$

Let $X$ be a separated irreducible variety which is regular in codimension $1$ (I want to talk about Weil divisors), and write $\mathbb{P}^n_X = \mathbb{P}^n_k \times X$ for the projective space over $X$. Now if $Y \subset \mathbb{P}^n_X$ is a prime Weil divisor, i.e. an irreducible closed subvariety of codimension $1$, we can consider the generic fiber $Y_K$ (here $K$ is the function field of $X$).

Since closed embeddings are stable under base change, $Y_K$ is a closed subscheme of $\mathbb{P}^n_K$, and clearly $Y_K \neq \varnothing$ if and only if the projection of $Y$ to $X$ is dense. My question: in this case ($Y_K \neq \varnothing$), can we conclude that $Y_K$ is an integral subvariety of $\mathbb{P}^n_K$ of codimension $1$? Hopefully this is true, since then we will have a well-defined degree map $\text{Pic } \mathbb{P}^n_X \to \mathbb{Z}$.

Edit: Maybe I should be asking a more general question: if $Y \to X$ is a dominant morphism of (separated, if necessary) irreducible varieties and $K$ denotes the function field of $X$, is $Y_K = Y \times_X \text{Spec } K$ an irreducible variety over $K$? The codimension $1$ assertion seems easier, since it is local, but I don't know a good criterion for irreducibility.

$\endgroup$
  • $\begingroup$ What about $Y_K$ being reduced? This seems to be wrong in general (but I haven't given it much thought). $\endgroup$ – Martin Brandenburg May 29 '12 at 19:15
1
$\begingroup$

I guess you mean by "irreducible varieties" integral scheme of finite type. Then the anwser is yes.

First the local rings of $Y_K$ are localizations of local rings of $Y$ (can replace $X$ by an affine dense open subvariety, then $K$ is a localization of $O(X)$), so $Y_K$ is locally integral.

Second, the generic point of $Y$ belongs to $Y_K$, so it is also the generic point of $Y_K$. This implies that $Y_K$ is irreducible.

Edit Codimension of $Y_K$. Let $d=\dim X$. Then $\dim Y=n+d-1$, and we have $$\dim Y_K=\dim Y - \dim X=n-1$$ (dimension formula for irreducible algebraic varieties). So $Y_K$ is of codimension $1$ in $\mathbb P^n_K$.

$\endgroup$
  • $\begingroup$ What about the codimension $1$ question? $\endgroup$ – Justin Campbell May 29 '12 at 22:41
  • $\begingroup$ @JustinCampbell: I didn't see the codimension 1 question. $\endgroup$ – user18119 May 30 '12 at 9:32

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.