Let $X$ be a projective variety over a field $k$, and $\mathcal L$ a vector bundle on $X$, i.e. a locally free $\mathcal O_X$-module of finite rank. For each $n\geq 0$, $\text{Sym}^n \mathcal L$ is a vector bundle on $X$, hence $H^0(X, \text{Sym}^n \mathcal L)$ is a finite-dimensional $k$-vector space. Consider the graded $k$-algebra

$$(\text{Sym } \mathcal L) (X) := \bigoplus_{n=0}^\infty H^0(X, \text{Sym}^n \mathcal L).$$

What kind of conditions on $X$ and $\mathcal L$ ensure that this algebra is finitely-generated? Is it always the case?


The ring you describe is the ring of global sections of the total space of the vector bundle $\mathcal{L}$, i.e. $\mathbf{Spec}(\mathrm{Sym}^\bullet(\mathcal{L}))$. (This notation is the "global Spec" construction for a sheaf of $\mathcal{O}_X$-algebras.)

In general, this does not need to be a finitely-generated $k$-algebra (see this shocking counterexample, for a rank-2 vector bundle over an elliptic curve: http://math.stanford.edu/~vakil/files/nonfg.pdf).

It is true for (very?) ample line bundles, since then it is the homogeneous coordinate ring of $X$ under the corresponding embedding. (Slight white lie here -- it's the integral closure of the homogeneous coordinate ring.)

It is true for arbitrary line bundles on toric varieties apparently: http://arxiv.org/pdf/alg-geom/9608034.pdf.

| cite | improve this answer | |
  • 3
    $\begingroup$ Very interesting answer! $\endgroup$ – Georges Elencwajg Jan 24 '15 at 9:28
  • 2
    $\begingroup$ Thanks Jake, resourceful as always! :) $\endgroup$ – Bruno Joyal Feb 3 '15 at 3:51

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.