MathOverflow is a question and answer site for professional mathematicians. 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

A map from an algebraic variety $X$ to a projective space is the same thing as a globally generated line bundle on $X$. What geometric object on $X$ corresponds to a map to a weighted projective space?

share|cite|improve this question
A line bundle $L$ on $X$ and a set of sections $(s_1, s_2, \ldots, s_k)$ where $s_i \in H^0(X, \, L^{a_i})$? (If you want a morphism, you must also require that the common zero locus of these sections is empty). – Francesco Polizzi Jul 12 '14 at 8:33
@FrancescoPolizzi That makes sense! Maybe you can write it as an answer. I was thinking that it should be some kind of orbi-bundle, but the fact that $X$ has no orbifold structure was making me confused. – Dan Petersen Jul 12 '14 at 9:59
up vote 15 down vote accepted

Let me expand my comment in a short answer.

One of the most common way to build a rational map $f \colon X \dashrightarrow \mathbb{P}(a_1, \ldots, a_n)$ is to consider a line bundle $\mathscr{L}$ on $X$ together with sections $$\sigma_1 \in H^0(X, \, \mathscr{L}^{a_1}), \quad \sigma_2 \in H^0(X, \, \mathscr{L}^{a_2}), \ldots, \sigma_n \in H^0(X, \, \mathscr{L}^{a_n}),$$ and to define $f(x) := [\sigma_1(x): \ldots : \sigma_n(x)].$

Clearly, $f$ is a morphism if and only if the intersection of the zero loci of the sections $\sigma_i$ is empty.

The basic example is given by hyperelliptic curves. Let $X$ be a smooth hyperelliptic curve of genus $g$ and take as $\mathscr{L}$ the hyperelliptic involution, i.e. the unique $g^1_2$ of $X$. Then it is no difficult to show that there exist $x_0, \, x_1 \in H^0(X, \, \mathscr{L})$ and $y \in H^0(X, \mathscr{L}^{g+1})$ that satisfy a polynomial relation of type $$y^2 = F_{2g+2}(x_0, \, x_1), \quad (\sharp)$$ where $F_{2g+2}$ is a homogeneous form of degree $2g+2$.

This precisely means that we can write $X$ as a double cover of $\mathbb{P}^1$, branched at $2g+2$ points. Trying to homogenize the equation $(\sharp)$ in the usual way introduces a complicate singularity at infinity. Instead, it is more convenient to see $(\sharp)$ as the global equation of $X$ into the weighted projective plane $\mathbb{P}(1, \, 1,\, g+1)$.

In fact, it is easy to see that the map $f \colon X \longrightarrow \mathbb{P}(1, \, 1, \, g+1)$ defined by the sections $x_0, \, x_1, \, y$ is an embedding (notice that the image does not pass through the singular point $[0: 0: 1]$ of the weighted projective plane).

share|cite|improve this answer
Thanks! (blank space) – Dan Petersen Jul 12 '14 at 15:04
It would be interseting to describe the converse. Given a morphism f:X⇢ℙ(a1,…,an) , what are the corresponding line bundle and sections ? In other words, what is the functor of a weighted projective space ? Of course Dan Petersen' question was only one side (sufficient case). I am asking a new and more general question ! ( May I post it as a question ?) – Al-Amrani Oct 28 '15 at 13:56

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.