Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I was doing some exercises in Liu's book on Algebraic Geometry. I am currently trying to solve a problem by showing the following:

Let $U \subset \mathbb{P}^n_k$, k a field, be an affine open subset.
Show that the irreducible components of $\mathbb{P}^n_k-U$ all have dimension n-1.

I would appreciate any help / hint here. I have some problems understanding $\mathbb{P }^n_k$ at a deep (even semi-deep) level. I suspect that one could maybe show that the dimension of such an affine open should be of dimension n (or am I wrong here?), since we can compute the dimension of X on any open set. We should be able to write the complement as $V_+(I)$ for some homogenous ideal . However, I don't see how to get from this to that the irreducible components of the component has dimension n-1.

Thank you for looking at my question and please forgive me if it's a naive one.

share|cite|improve this question
I don't think it is naive. Here is something related:… – Andrew Dec 19 '12 at 0:07
As Andrew pointed out, this result holds in a fairly general situation (affine open subset in a noetherian separated scheme, this is also Exercise 4.1.15). But in the UDF case as in $\mathbb P^n_k$, the proof is much easier: use the previous parts of Exericse 2.5.12. – user18119 Dec 19 '12 at 7:06
QiL: I'm afraid I don't see an obvious application of either of the previous ones. Should I use something like that on $D(x_i)$ it is principal? The wonders of technology, by the way, allowing one to talk directly to the author of the book. Thank you very much for your answers! – Dedalus Dec 19 '12 at 10:12
@Dedalus: it seems like the author didn't give enough details for this exercise :). Consider the intersection with the affine spaces in $\mathbb P^n$ and use (b). If this is not enough I will post an answer. – user18119 Dec 19 '12 at 13:22
@QiL: I don't think the details are neccesarily left out (it might be that I'm simply dense here). If we let U be the open affine. if we go to the intersections of U with the principal affine opens $D(x_i)$, we get that in each such affine, $U \cap D(x_i)$ is principal. The complement is $V(f)$, and it has irreduble components of dimension n-1 (right?) . So, can we from this conclude that each irreducible component of the complement of U in $\mathbb{P}^n_k$ is irreducible of dimension n-1? – Dedalus Dec 19 '12 at 13:55
up vote 3 down vote accepted

For the sake of an answer. Cover $\mathbb P^n_k$ (with coordinates $x_0, \dots, x_n$) by affines spaces $U_0, \dots, U_n$ with $U_i=D_+(x_i)$. Then $$ \mathbb P^n_k\setminus U=\cup_i (U_i\setminus (U\cap U_i))$$ and it is enough to show $U_i\setminus (U\cap U_i)$ is empty or pure of dimension $n-1$ for all $i\le n$. By a previous part of the exercice, we know that $U_i\setminus (U\cap U_i)$ is a principal closed subset of $U_i$, hence it is empty or pure of dimension $n-1$.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.