Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

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'm interested in finite galois covers $\varphi: Y \rightarrow X$ between smooth proper curves over an algebraically closed field of characteristic zero, which are étale outside a prescribed finite set $D$ of points of $X$.

More precisely, I'd like to understand how the étale fundamental group $\widehat\pi_1(U_{X})$ of $U_{X}:=X \setminus D$ determines the ramification indices $e_{y}$ of points $y$ of $Y$.

As a starting point I have the following more specific question:

Suppose that $X=\mathbb{P}^{1}_{k}$, where $k$ is an algebraically closed field of characteristic zero, and that $D=\{0,1,\infty\}$. The étale fundamental group of $U_{X}=\mathbb{P}^{1}_{k}\setminus \{0,1,\infty\}$ is then the profinite group generated by three elements $\sigma_1,\sigma_2,\sigma_3$, subject to the only relation $$ \sigma_1 \cdot \sigma_2 \cdot \sigma_3=1 $$ Finite quotients of this group classify finite galois coverings $\varphi: Y \rightarrow \mathbb{P}^{1}_{k}$, which are étale outside of $\{0,1,\infty\}$.

Is it possible to choose the covering $\varphi:Y \rightarrow \mathbb{P}^{1}_{k}$ such that every point $y \in \varphi^{-1}(\{0,1,\infty\})$ has ramification index $e$, where $e$ is a fixed positive integer, $e \geq 2$? If so, which are the properties of the finite quotient of $\widehat{\pi}_1(\mathbb{P}^{1}_{k}\setminus\{0,1,\infty\})$, associated to $\varphi$, which reflect the order of the ramification indices?

I think that at least my first question might be related to (or possibly even be answered by) the Riemann existence theorem, but I was unable to figure out how.

share|cite|improve this question
I think by Riemann-Hurwitz you can numerically rule out some cases of being able to find a covering $X\to \mathbb{P}^1$ so that every point has the same ramification index. I should probably wait until morning to double check this, but it seems you'd get $2g(X)-2=-2d+3(e-1)$ where $d$ is the degree of the covering. So maybe for even $e$ no such cover can exist? – Matt Oct 15 '12 at 5:29
Dear @Matt, how do you get the $3$ in your formula using Riemann-Hurwitz? – Nils Matthes Oct 15 '12 at 8:52
Ah. You're right. I was making the simplifying assumption that there was only one "point" over each of $0, 1$ and $\infty$. You could of course have 2 over each, and that fixes the problem because there would be a 6 in front. – Matt Oct 15 '12 at 23:53
@Matt: yes, that's what I was thinking. Thanks for your response, nonetheless. I wish there were more of them. – Nils Matthes Oct 16 '12 at 8:06
up vote 3 down vote accepted

The relevant reference seems to be Exposé XIII, Corollaire 2.12. of SGAI. For the specific example $\Bbb{P}^1_k \setminus \{0,1,\infty\}$, with $k$ algebraically closed of characteristic zero, it says that giving a finite Galois cover

$$ \varphi: Y \rightarrow \Bbb{P}^1_k $$

with (finite) Galois group $G$, which is étale outside of $\{0,1,\infty\}$, is equivalent to giving a surjective group homomorphism

$$ \phi: \widehat{\pi}_1(\Bbb{P}^1_k \setminus \{0,1,\infty\}) \rightarrow G $$ Also it is shown in loc.cit. that $\widehat{\pi}_1(\Bbb{P}^1_k \setminus \{0,1,\infty\})$ is the pro-finite group generated by elements $\sigma_1,\sigma_2,\sigma_3$, subject only to the single relation $$ \sigma_1 \cdot \sigma_2 \cdot \sigma_3=1 $$ Moreover, if we set $x_1=0$, $x_2=1$, $x_3=\infty$ (the actual ordering of the three points $0,1,\infty$ is irrelevant) then the ramification index of every point in $\varphi^{-1}(x_i)$ equals the order of $\phi(\sigma_i)$ in $G$.

So putting everything in context of the question, if I want a finite Galois cover $$\varphi: Y \rightarrow \Bbb{P}^1_k$$ which is étale outside of $D=\{0,1,\infty\}$ and has ramification index equal to $e$ at every point of $D$, I just have to find a finite group $G$, which can be generated by three elements $\overline{\sigma}_1,\overline{\sigma}_2,\overline{\sigma}_3$, each of order $e$ such that $$ \overline{\sigma}_1 \cdot \overline{\sigma}_2 \cdot \overline{\sigma}_3 = 1 $$ It is clear that for a prime number $e \geq 3$, $G=\Bbb{Z}/e\Bbb{Z}$ does the job. This is in fact sufficient for my purposes.

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.