1
$\begingroup$

Let $X$ be a quasi-hyperelliptic surface in characteristic 3 where the canonical bundle $K_X$ is trivial.

Question: Is there a finite étale covering $Y \rightarrow X$ such that

  • $Y$ is an abelian variety, or
  • the étale fundamental group of $Y$ is trivial?

Motivation:

The Beauville-Bogomolov decomposition theorem for $\mathbb C$ dictates that a smooth projective variety $X/\mathbb C$ with $K_X$ trivial has a finite étale cover $Y \rightarrow X$ such that $Y$ is a product of abelian varieties, "strict" Calabi–Yau varieties and irreducible symplectic varieties. The two latter has their étale fundamental groups trivial.

Though there are no universally agreed definition of irreducible symplectic varieties in char. p, we could still "disprove" Beauville-Bogomolov decomposition in char. p if the covering spaces have undesirable properties.

$\endgroup$

1 Answer 1

1
$\begingroup$

From pp. 488-489 of W.E. Lang's thesis (Ann. ENS. vol. 12, 1979) it follows that neither case can arise. For the fibration $X\to C$ has an elliptic base and all its geometric fibers are cuspidal rational curves, so that $\pi_1(X) =\pi_1(C)$ and then every etale cover of $X$ is also a quasihyperelliptic surface.

$\endgroup$
2
  • $\begingroup$ Why are the base changes induced by the covers of $C$ quasi-hyperelliptic surfaces? $\endgroup$ Commented Nov 8, 2023 at 0:22
  • 1
    $\begingroup$ From the definition of quasi-hyperelliptic as given by Lang. $\endgroup$
    – inkspot
    Commented Nov 8, 2023 at 8:49

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.