1
$\begingroup$

Let $C$ be a curve in $\mathbb{P}^3$ which is of local complete intersection is some smooth surface in $\mathbb{P}^3$. Assume $C$ is non-reduced. What can we say about $h^0(\mathcal{O}_C)$? When can we say that $h^0(\mathcal{O}_C) \le 1$? The curve that I have is mind is of the form $2l+C'$ (seen as a divisor in a smooth surface in $\mathbb{P}^3$) where $l$ is a line and $C'$ is a reduced plane curve lying on the same plane as $l$. Any idea/reference for the direction of approaching this problem will be most appreciated.

EDIT Note that $h^0(\mathcal{O}_C)) \ge 1$ and is equal to $1$ if and only if $h^1(\mathcal{O}_X(-C))=h^1(\mathcal{O}_X(C)(d-4))$ vanish.

$\endgroup$
1
  • 1
    $\begingroup$ Why the down vote? Is it a trivial question or is there something very unclear? $\endgroup$ Dec 2, 2013 at 8:38

1 Answer 1

1
$\begingroup$

Your question is probably not well formulated. What I understand is that the curve $C$ is actually a plane curve of the form $2l+C'$, with $l$ and $C'$ in $\mathbb{P}^2$. Then of course $h^0(\mathcal{O}_C)=1$, because of the exact sequence $\ 0\rightarrow \mathcal{O}_{\mathbb{P}^2}(-C)\rightarrow \mathcal{O}_{\mathbb{P}^2}\rightarrow \mathcal{O}_C\rightarrow 0 $.

$\endgroup$
3
  • $\begingroup$ $C$ is not necessarily a plane curve. I do not know how else to write it. I meant that the curve is the scheme associated to an effective divisor on a surface. Using the adjunction formula one sees that the arithmetic genus depends on the degree of the surface. For example take a smooth surface containing $l, C'$. This will also contain the curve $2l+C'$ as a Weil divisor. But the genus depends on the degree of the surface. $\endgroup$ Dec 2, 2013 at 8:37
  • $\begingroup$ Sorry I don't get it. You say that "$C'$ is a reduced plane curve lying on the same plane as $l$". Doesn't that mean that $C$ is the curve $2l+C'$ in that plane? $\endgroup$
    – abx
    Dec 2, 2013 at 8:51
  • $\begingroup$ No. The ideal of $2l+C'$ does not contain the ideal of the plane. See "Le Schema de Hilbert des Courbes gauches localement Cohen-Macualay n'est (presque) jamais reduit" by M Martin-Descamps and D. Perrin Proposition $0.6$ for such examples. $\endgroup$ Dec 2, 2013 at 9:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.