2
$\begingroup$

It is known that the moduli space $\overline{M}_{g}$ of genus $g$ curves is of general type for $g\geq 24$.

By Theorem 2.4 of

Logan, Adam The Kodaira dimension of moduli spaces of curves with marked points. Amer. J. Math. 125 (2003), no. 1, 105–138.

all but finitely many of the $\overline{M}_{g,n}$ with $g>3$ are of general type. Furthermore in Theorem 5.1 of the same paper for any $4\leq g\leq 23$ an $n$ such that $\overline{M}_{g,n}$ is of general type is determined.

Is there an inequality in terms of $g,n$ which implies that $\overline{M}_{g,n}$ is of general type ?

$\endgroup$
5
  • $\begingroup$ I am not sure what you are asking. In any case you must impose $g\geq 4$ since the result is not known for $\bar{M}_{2,n}$ and $\bar{M}_{3,n}$. Then you may observe that Logan's inequalities imply (trivially) that $\bar{M}_{g,n}$ is of general type for $g+n\geq 30$. Does that answer your question? $\endgroup$
    – abx
    Nov 3, 2013 at 15:02
  • $\begingroup$ How do you see it for instance for $g=22, n=9$ and for $g=23, n = 7$ ? $\endgroup$
    – Puzzled
    Nov 3, 2013 at 15:16
  • $\begingroup$ According to Logan, $n=8$ is enough for $g=22$, and $n=1$ for $g=23$. $\endgroup$
    – abx
    Nov 3, 2013 at 16:23
  • $\begingroup$ Thank you very much. Do you have a reference for the statement: general type on the base + general type on the general fiber = the variety is of general type ? $\endgroup$
    – Puzzled
    Nov 3, 2013 at 16:54
  • $\begingroup$ I read the comment down the page. Thanks again. $\endgroup$
    – Puzzled
    Nov 3, 2013 at 16:55

1 Answer 1

3
$\begingroup$

So it seems your question is: if we know that $\overline M_{g,n}$ is of general type, is the same true for $\overline M_{g,n'}$ with $n' \geq n$? The answer is yes, and is a special case of the main theorem of: J. Kollár, Subadditivity of the Kodaira dimension : fibers of general type.

As far as I know there are no explicit counterexamples to the stronger statement that if $\overline M_{g,n}$ is of general type, then the same true for $\overline M_{g',n}$ with $g' \geq g$ and $n' \geq n$, but it is almost certainly false: for instance, Farkas has proved that $\overline M_{22}$ is of general type and also that $\kappa(\overline M_{23}) \geq 2$, which he however conjectured to be sharp. Also note that Logan's function $f(g)$ such that $\overline M_{g,n}$ is of general type for $n \geq f(g)$ is not monotone in $g$.

$\endgroup$
5
  • $\begingroup$ Thank you. I see but the function $f(g)$ is just on finitely many $g$. Furthermore for any such $g$ he find an $n$ such that $\overline{M}_{g,n}$ is of general type but he does not state it for any $n\geq f(g)$. $\endgroup$
    – Puzzled
    Nov 3, 2013 at 16:13
  • $\begingroup$ I guess that now my question becomes this: let $f:X\rightarrow Y$ be a fibration where $Y$ and the general fiber of $f$ are of general type. Is $X$ of general type? $\endgroup$
    – Puzzled
    Nov 3, 2013 at 16:24
  • $\begingroup$ Yes. So if $\bar{M}_{g,n}$ is of general type, $\bar{M}_{g,n'}$ for $n'\geq n$ is of general type: it maps onto $\bar{M}_{g,n}$ with fibers of general type. $\endgroup$
    – abx
    Nov 3, 2013 at 16:26
  • $\begingroup$ So this more general fact is true: let $f:X\rightarrow Y$ be a fibration where $Y$ and the general fiber of $f$ are of general type. Then $X$ of general type. Do you have a reference for this? $\endgroup$
    – Puzzled
    Nov 3, 2013 at 16:28
  • 1
    $\begingroup$ I think this is due to Viehweg : Die Additivität der Kodaira Dimension für projektive Faserräume über Varietäten des allgemeinen Typs. J. Reine Angew. Math. 330 (1982), 132-142. $\endgroup$
    – abx
    Nov 3, 2013 at 16:54

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.