A rank inequality for the Tate conjecture over global function fields

DOI10.1016/J.EXMATH.2008.10.008MaRDI QIDQ1012899zbMATH OpenOpenAlexWikidataFDO

Publication date 23 April 2009

Published in Expositiones Mathematicae (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/0812.0094, https://resolver.caltech.edu/CaltechAUTHORS:20090911-153504218

zbMATH Keywords

\(\ell\)-adic representations \(L\)-functions \(p\)-adic cohomology algebraic cycles automorphic representations Langlands correspondence Tate conjecture varieties over global fields

Mathematics Subject Classification ID

Representation-theoretic methods; automorphic representations over local and global fields (11F70) Varieties over global fields (11G35) (L)-functions of varieties over global fields; Birch-Swinnerton-Dyer conjecture (11G40) Langlands-Weil conjectures, nonabelian class field theory (11R39) Algebraic cycles (14C25) (p)-adic cohomology, crystalline cohomology (14F30) Global ground fields in algebraic geometry (14G25)

Abstract: Following D. Ramakrishnan, we explain how L. Lafforgue's modularity theorem and an analytic theorem of H. Jacquet and J. Shalika can be applied to prove the following result related to the Tate Conjecture: for a smooth, projective, geometrically-connected variety defined over a global function field, the algebraic rank is less than or equal to the analytic rank. Also discussed is the analogous (open) question for number fields and an easy extension of Lafforgue's theorem to remove the "finite-order character" assumption. All results are likely "known to the experts", but don't appear to be written down.

Review text

Let \(X\) be a smooth, projective, geometrically connected variety over a global function field \(k\). Denote by \(\mathbb F_q\) the constant field of \(k\), let \(\overline k\) be its separable closure, fix a prime number \(\ell\neq \text{char}\,k\) and an integer \(0\leq m\leq\dim X\), and finally consider the \(2m\)-th \(\ell\)-adic étale cohomology group \(V_\ell:= H^{2m}_{\text{ét}}(X\times_k \overline k,\mathbb Q_\ell)\). Then the natural action of \(\Gamma_k:=\text{Gal} (\overline k/k)\) on \(\overline k\) induces a continuous representation \(\rho_\ell:\Gamma_k \to\Aut_{\mathbb Q}(V_\ell)\), which in turn gives rise to a certain \(L\)-function, to the so-called \(m\)-th Tate twist \(V_\ell(m)\) of \(V\), and to a special algebraic cycle of codimension \(m\) on \(X\). These invariants define particular integers \(r^{(m)}_{an,k}\), \(r^{(m)}_{\ell,k}\) and \(r^{(m)}_{\text{alg},k}\), respectively, which were introduced by J. Tate, more than 45 years ago, in his seminal paper ``Algebraic cycles and poles of zeta functions [cf.: \textit{J. Tate}, in: Arithmetical Algebraic Geometry, , Proc. Conf. Purdue Univ. 1963, Harper and Row, New York, 93--110 (1965; Zbl 0213.22804). In this context, the famous, long-standing Tate conjecture predicts that, in fact, these three quantities coincide. In the exposition under review, the author reports on some very recent progress concerning the still unproven Tate conjecture. The main result (Theorem 2.1) presented here states that, indeed, the following relations are valid: \[ r^{(m)}_{\ell,k}=r^{(m)}_{an,k}\quad\text{and}\quad r^{(m)}_{alg, k}\leq r^{(m)}_{an,k} \] for any \(0\leq m\leq\dim X\). This spectacular result, which must be seen as a major step towards a complete proof of the Tate conjecture, is based on an observation pointed out by D. Ramakrishnan during a lecture given at a workshop in July 2007. In the current paper, the author elaborates this idea of proof rigorously and in full detail, thereby using a wealth of highly advanced, partly very recently developed methods and results. In this regard, the main ingredients of proof are the global Langlands correspondence for the groups \(\text{GL}_n\) over global function fields (à la \textit{G. Lafforgue}, [Invent. Math. 147, No. 1, 1--241 (2002; Zbl 1038.11075)], on the one hand, and an earlier analytic result of H. Jacquet and J. A. Shalika (1981) on automorphic \(L\)-functions for \(\text{GL}_n\), on the other hand, which are specifically used to derive effective bounds for \(r^{(m)}_{alg,k}\) and \(r^{(m)}_{an,k}\), respectively. The author concludes his enlightening discussion with several hints and remarks concerning the analogue of the Tate conjecture for number fields, especially for abelian varieties which are potentially CM-type ones.

Recommendations

Cites work

Cited in

(4)

This page was built for publication: A rank inequality for the Tate conjecture over global function fields

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1012899)