6
$\begingroup$

Consider a holomoprhic map from a Riemann surface $$ f: \Sigma_g \to \mathbb{CP}^n. $$ This is given by some homogeneous polynomials in some variables.

How can we show that the homogeneous degree $d$ of these polynomials coincides with the definition of degree from fundamental homology class, namely $d = f_* ([\Sigma_g]) \in H_2(\mathbb{CP}^n;\mathbb{Z})$?

$\endgroup$
4
$\begingroup$

Let me rename $C=\Sigma_g$, as the genus is unimportant here. To give a map from a smooth complex projective curve $C$ to $\mathbb P^n$ is the same as to give a line bundle $L$ on $C$. This line bundle can be recovered up to isomorphism as $$L=f^\ast(\mathscr O_{\mathbb P^n}(1))=g^\ast(\mathscr O_{\mathbb P^n}(1)|_D),$$ where I have denoted by $g$ the same map $f$, but restricted to the image $D=f(C)$; in other words we factor $f$ as follows: $$C\overset{g}{\longrightarrow}D\hookrightarrow\mathbb P^n.$$ On $\mathbb P^n$, the one-cycle $f_\ast[C]$ has degree $$\deg f_\ast[C]=\deg g_\ast[C]=(\deg g)\cdot(\deg D).$$ On the other hand, the degree of the polynomials defining $f$ is the degree of $L$, which is $$\deg L=(\deg g)\cdot \deg \mathscr O_{\mathbb P^n}(1)|_D=(\deg g)\cdot(\deg D).$$

$\endgroup$
  • $\begingroup$ Thanks for the answer. You mean two-cycle, right? So, requiring that $f$ maps to a general quintic in $\mathbb{CP}^4$ would amount to take $L=f^*(\mathcal{O}_{\mathbb{CP}^4}(5))$, correct? Moreover, is it possible to understand what is the effect, at the level of line bundles (or divisors) of making $f$ an embedding (->very ample?), or making $f$ equivariant with respect to some involutions on $C$ and $\mathbb{CP}^n$, or again deforming one $f$ into another, e.g. when smoothing a node? $\endgroup$ – jj_p Mar 18 '14 at 12:15
  • 1
    $\begingroup$ Sorry for the confusion but I called "curve" what you called "surface"! so yes, two-cycles is fine. And yes, your $L$ looks fine (it has the correct degree). "Very ample" is the key word. Sorry I do not feel able to say more about your comment! Maybe post a new question and someone will! :) $\endgroup$ – Brenin Mar 18 '14 at 22:13

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.