Big symplectic or orthogonal monodromy modulo

DOI10.1215/S0012-7094-08-14115-8MaRDI QIDQ2465261zbMATH OpenOpenAlexFDO

Publication date 9 January 2008

Published in Duke Mathematical Journal (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/math/0608718

Big monodromy hyperelliptic and elliptic curves inverse Galois problem sheaves

Elliptic curves over global fields (11G05) Abelian varieties of dimension (> 1) (11G10) Inverse Galois theory (12F12) Structure of families (Picard-Lefschetz, monodromy, etc.) (14D05) Jacobians, Prym varieties (14H40) Elliptic surfaces, elliptic or Calabi-Yau fibrations (14J27) Arithmetic ground fields for abelian varieties (14K15)

Abstract: Let k be a field not of characteristic two and L be a set of almost all rational primes invertible in k. Suppose we have a variety X/k and strictly compatible system {M_ell -> X : ell in L} of constructible F_ell-sheaves. If the system is orthogonally or symplectically self-dual, then the geometric monodromy group of M_ell is a subgroup of a corresponding isometry group G_ell over F_ell, and we say it has big monodromy if it contains the derived subgroup DG_ell=[G_ell,G_ell]. We prove a theorem which gives sufficient conditions for M_ell to have big monodromy. We apply the theorem to explicit systems arising from the middle cohomology of families of hyperelliptic curves and elliptic surfaces to show that the monodromy is uniformly big as we vary ell and the system.

Review text

Let \(k\) be a field, \(C/k\) a proper smooth geometrically connected curve, \(K=k(C)\) and \(E/K\) an elliptic curve with nonconstant \(j\)-invariant. \textit{E. Kowalski} [Int. J. Number Theory 2, No. 2, 267--288 (2006; Zbl 1115.11030)] considered if one could prove uniform big \({\mathbb F}_ {\ell}\)--monodromy results for certain sequences of quadratic twists of \(E\) when \(k={\mathbb F} _ {\ell}\) and showed how a sufficiently strong uniform bound as \(\ell\) varies, allows one to prove a variant of Goldfeld's average-rank conjecture. The goal of this paper is to complete Kowalski's proof by providing bounds of a sort that he requires. Also, the author shows that his methods can be used to give a new proof of a result of Yu which had given evidence of the existence of sufficient bounds required by Kowalski. The main result states that if \(V\) is an \({\mathbb F}_ {\ell}\)-vector space with a perfect pairing \(\varphi: V\times V\to {\mathbb F}_ {\ell}\), \(G\leq \roman{GL}(V)\) is an irreducible primitive subgroup preserving \(\varphi\). If \(\varphi\) is symmetric, \(G\) contains a reflection and an isotropic shear, and if \(\ell \geq 5\), then \(G\) is either: \({\mathcal O}(V)\), the orthogonal group; the kernel of the spinor norm; or the kernel of the product of the spinor norm and the determinant. If \(\varphi\) is alternating, \(G\) contains a transvection, and \(\ell\geq 3\), then \(G\) is all of the symplectic group \(\text{Sp}(V)\). The result of Yu proved is that the \(\bmod \ell\) geometric monodromy of hyperelliptic curves is \(\text{Sp}(2g,{\mathbb F}_ {\ell})\) for \(\ell>2\). In the last section, the author gives applications of the group theorem to families of quadratic twists of elliptic curves, and derives new results for the inverse Galois problem showing how big subgroups of orthogonal groups arise as Galois groups over \({\mathbb Q}(t)\).

Recommendations

Cites work

Cited in

(33)

This page was built for publication: Big symplectic or orthogonal monodromy modulo \(\ell\)

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2465261)