3
$\begingroup$

Let $X,Y$ be smooth irreducible algebraic varieties over $\mathbb{C}$, let $\pi: X \to Y$ be a morphism which is injective and surjective on closed points of $X,Y$. Then how to show $\pi$ is an isomorphism of $X,Y$ as varieties?

This result is not true if $X,Y$ are not smooth. One can consider the example $\mathbb{P}^1 \to C$, where $C$ is a plane curve defined by $y^2-x^3=0$, because the map $t \mapsto (t^2,t^3)$ satisfies the condition, but $\mathbb{P}^1$ cannot isomporhic to $C$ which is singular at origin.

$\endgroup$
5
  • 1
    $\begingroup$ Use (a special case of) Zariski's Main Theorem. $\endgroup$
    – anon
    Mar 16, 2013 at 19:46
  • $\begingroup$ The map must be birational. Since Y is normal, it has to be an isomorphism. $\endgroup$ Mar 16, 2013 at 21:15
  • 1
    $\begingroup$ It is in EGA IV, 3ème partie, Théorème 8.12.6, page 45, it is called " ,,Main Theorem'' de Zariski." It says that your map decomposes into an open immersion, followed by a finite morphism. In your case, you should be able to see that the both are isomorphisms. $\endgroup$ Mar 16, 2013 at 21:22
  • $\begingroup$ @Ray Hoobler, could you give more details on how the map is birational? $\endgroup$
    – Li Yutong
    Mar 16, 2013 at 21:54
  • 1
    $\begingroup$ Since it is injective, it is birational with its image (compute the degree of field extension: it is $1$). Because it is surjective, it is birational. $\endgroup$ Mar 17, 2013 at 9:58

1 Answer 1

1
$\begingroup$

One way to prove this is to show that for each closed point $x\in X$ the differential $T_\pi(x):T_xX\rightarrow T_{\pi(x)}Y$ is surjective, that is $\pi$ is a smooth morphism.

Let $\pi:X\rightarrow Y$ be a morphism or relative dimension $r$ of smooth varieties over an algebraically closed field. Assume that the relative cotangent sheaf $\Omega_{X/Y}$ is locally free of rank $r$ on $X$. We have an exact sequence $$\pi^{*}\Omega_{Y}\rightarrow\Omega_{X}\rightarrow\Omega_{X/Y}\mapsto 0.$$ Let $k(x)$ be the residue field at a closed point $x$. Tensorizing we get
$$\pi^{*}\Omega_{Y}\otimes k(x)\rightarrow\Omega_{X}\otimes k(x)\rightarrow\Omega_{X/Y}\otimes k(x)\mapsto 0.$$ Since $X$ and $Y$ are smooth and $\Omega_{X/Y}$ is locally free of rank $r$ these three vector spaces are of dimension $dim(Y),dim(X),r$ respectively. So the first map is injective and we have $$0\mapsto\pi^{*}\Omega_{Y}\otimes k(x)\rightarrow\Omega_{X}\otimes k(x)\rightarrow\Omega_{X/Y}\otimes k(x)\mapsto 0.$$ For any closed point $x\in X$ we have $k(x)\cong k$. Therefore we can identify the injective map $\pi^{*}\Omega_{Y}\otimes k\rightarrow\Omega_{X}\otimes k$ with the map between the cotangent spaces $\mathfrak{m}_y/\mathfrak{m}^2_y\rightarrow\mathfrak{m}_x/\mathfrak{m}^2_x$ where $y = \pi(x)$. Dualizing we have that the differential $T_\pi(x):T_xX\rightarrow T_yY$ is surjective. Therefore $\pi$ is a smooth morphism.

In particular, if any fiber of $\pi$ is just one point, then $\Omega_{X/Y}$ is locally free of rank $r = 0$ and $\pi$ is a smooth morphism of relative dimension zero. Finally, since $\pi$ is surjective it is an isomorphism.

$\endgroup$
6
  • $\begingroup$ Thank you! Why $\Omega_{X/Y}$ is locally free when the fibre of $\pi$ is just one point? Your argument let me thing that flatness of $\pi$ is enough to show the isomorphism --- is that right? $\endgroup$
    – Li Yutong
    Apr 2, 2014 at 14:31
  • $\begingroup$ I am assuming that the fiber of $\pi$ is reduced. Therefore $\Omega_{X/Y}$ is locally free. In general flat is not enough. For instance project a conic $C$ in $\mathbb{P}^2$ from a point $p\notin C$. You have two points $x,y\in C$ where the lines through $p$ are tangent. The projection $\pi:C\rightarrow\mathbb{P}^1$ is flat. However $T_{\pi}(x):T_xC\rightarrow T_{\pi(x)}\mathbb{P}^1$ is not surjective, and $T_{\pi}(y)$ is not surjective as well. As points of the fiber of $\pi$ both $x,y$ have multiplicity two. $\endgroup$
    – Puzzled
    Apr 2, 2014 at 16:17
  • $\begingroup$ Thank you again! Is the statement that "The fibre of $\pi$ is reduced then $\Omega_{X/Y}$ is locally free" somewhere in standard text book? I have never seen it before:( $\endgroup$
    – Li Yutong
    Apr 3, 2014 at 0:33
  • $\begingroup$ Besides, I might be understand wrong, but in your example, isn't the fibre of $\pi$ always two points? So it does not satisfy the requirement that $\pi$ is injective? $\endgroup$
    – Li Yutong
    Apr 3, 2014 at 0:34
  • $\begingroup$ I thought flatness is okay because if $\pi$ is proper, then injective of $\pi$ would imply $\pi$ to be finite, and flatness would imply it is an isomorphism (because locally, it is a rank one free module). Is there anything wrong in that argument? $\endgroup$
    – Li Yutong
    Apr 3, 2014 at 0:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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