Boundary Andreotti–Mayer conjecture for tangentially degenerate loci

About 19 years old · traced to

Let (B,Ξ)(B,\Xi) be a simple principally polarized abelian variety of dimension gg, and suppose that (B,Ξ)∉Ng,i(B,\Xi)\notin N_{g,i} for every i≥k≥1i\geq k\geq1. Here Nk(B,Ξ)⊂BN_k(B,\Xi)\subset B is the locus of points bb such that Ξ\Xi and its translate Ξb\Xi_b are tangentially degenerate along a subvariety of dimension kk. Let ZZ be an irreducible component of Nk(B,Ξ)N_k(B,\Xi), and let codim⁡B(Z)\operatorname{codim}_B(Z) denote its codimension in BB. Boundary Andreotti–Mayer conjecture. There is such a component ZZ with

codim⁡B(Z)=k+1\operatorname{codim}_B(Z)=k+1

if and only if either

g≥2,k=g−2,andB is a hyperelliptic Jacobian,g\geq2,\quad k=g-2,\quad\text{and}\quad B\text{ is a hyperelliptic Jacobian},

or

g≥3,k=g−3,andB is a Jacobian.g\geq3,\quad k=g-3,\quad\text{and}\quad B\text{ is a Jacobian}.

This is proposed as a boundary version of the Andreotti–Mayer conjecture and as a conjectural characterization of Jacobians and hyperelliptic Jacobians among simple principally polarized abelian varieties; the supplied text gives no resolution.

References

Primary source

Ciro Ciliberto and Gerard van der Geer, “Andreotti-Mayer loci and the Schottky problem”, arXiv:math/0701353 (2007).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.