Take the 2-minute tour ×
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.

Consider a polynomial $P(z) = z^4 \in \mathbb{C}[z]$. Set-theoretically $P(z)$ has one root equal to zero. From algebraic point of view it has four roots: root zero has multiplicity four. Also we can't draw a curve in $\mathbb{C}$ around one of such roots but not around the others.

Now consider a Riemann surface $X$ given by $$ X = \left\{ (z,w) \in \mathbb{C}^2 \colon w^2 - z^4 + 1 = 0 \right\}. $$ $\hskip1cm$ enter image description here enter image description here

Its compactification in $\mathbb{C}P^2$ is a surface $\overline{X}$: $$ \overline{X} = \left\{ (\xi : \eta : \zeta) \in \mathbb{C}P^2 \colon \eta^2\zeta^2 - \xi^4 + \zeta^4 = 0 \right\}. $$ Points at infinity in $\mathbb{C}P^2$ are given by $\zeta = 0$. So if we want to find intersection of $\overline{X}$ with infinity we should solve a system $\eta^2 \zeta^2 - \xi^4 + \zeta^4 = 0$, $\zeta = 0$. It has a solution $\zeta = 0$, $\xi = 0$. Then set-theoretically $\overline{X}$ has one point at infinity $(0:1:0) \in \mathbb{C}P^2$.

Now take in open neighborhood $\eta \neq 0$ of point $(0:1:0)$ as coordinates $$ x = \frac{\xi}{\eta}, \;\;\;y = \frac{\zeta}{\eta}. $$ Then points at infinity are given by $y = 0$, our surface is given by $y^2 + y^4 - x^4 = 0$ and intersection of our surface with line at infinity is given by the ideal $\langle y, y^2 + y^4 - x^4 \rangle = \langle y, x^4 \rangle \subset \mathbb{C}[x,y]$. Multiplicity of intersection is $\dim_{\mathbb{C}} \frac{\mathbb{C}[x,y]}{\langle y, x^4 \rangle} = 4$. Then our point $(0 : 1 : 0)$ has a multiplicity four and intersection-theoretically there are four points at infinity.

In Dubrovin's book "Riemann surfaces and non-linear equations" author writes that a surface $X$ has two points at infinity like any surface given by equation $w^2 = P_{2m}(z)$ where $P_{2m}(z) \in \mathbb{C}[z]$, $m > 1$ and $P_{2m}(z)$ has mutually distinct roots. I would like to know which theory gives two points at infinity and if there is possible to draw a contour on $\overline{X}$ around one of such points but not around the others. Thank you for answers.

share|improve this question
    
For background see the preceding question –  Georges Elencwajg Sep 23 '12 at 11:44

1 Answer 1

up vote 1 down vote accepted

The analytic curve $\bar X$ has only one point $P$ at infinity (i.e. with $\zeta=0$) but it is not a Riemann surface because $P$ is a singularity of $\bar X$.
The right compactification of $X$ is the Riemann surface $\bar X^\nu$ obtained by normalizing $\bar X$.
It is obtained by deleting $P$ from $\bar X$, replacing it by two points $P_1, P_2$, and endowing the resulting set with a suitable structure of Riemann surface.
You then obtain a normalization morphism of analytic curves $n:\bar X^\nu \to \bar X$ such that $n^{-1}(P)=\lbrace P_1,P_2\rbrace $ and such that the retriction $n_{\text{res}}: \bar X^\nu\setminus \lbrace P_1,P_2\rbrace \to X$ is an isomorphism of Riemann surfaces.

The Riemann surface $\bar X^\nu$ has genus one and becomes an elliptic curve once you arbitrarily choose a point on it to serve as the zero point of the elliptic curve.

To answer your last question, you can certainly choose a contour around each of the two $P_i$'s not enclosing the other one, but this is a general property of Riemann surfaces, having nothing to do with the situation at hand here.

share|improve this answer
    
Thank you, Georges. Can you please advise me some book where I can read about normalization? My question is of practical interest and now I can't imagine how to draw normalized surface with a curve around one of point at infinity and not around the other. –  Nimza Sep 23 '12 at 17:11
    
Dear Nimza, a good reference for the normalization of curves is Chapter 2 of Griffiths's Introduction to Algebraic Curves . –  Georges Elencwajg Sep 24 '12 at 22:07

Your Answer

 
discard

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.