Tate modules of isocrystals and good reduction of Drinfeld modules (Q2036837)

Let \(\mathbb{F}_q\) be a finite field with \(q\) elements and \(F\) a fixed function field over \(\mathbb{F}_q\). Let us also fix a place \(\infty\) of \(F\) and set \(A\subset F\) to be the ring of elements integral outside of \(\infty\). We set \(K\) to be a field over \(\mathbb{F}_q\) and \(E\) a Drinfeld \(A\)-module of rank \(r\) defined over Spec\(K\). Consider a valuation \(v:K^{\times}\to \mathbb{Z}\) so that \(R\) denotes the ring consisting of elements with non-negative valuation. We finally define the morphism \(\iota:A\to K\) induced by the action of \(A\) on the target space of \(E\) at \(0\) and assume that \(\iota(A)\subset R\). Inspired by known results on the reduction theory of elliptic curves, over the years, there have been several works to describe a criterion for good reduction for Drinfeld \(A\)-modules. For instance, \textit{T. Takahashi}'s result [J. Math. Soc. Japan 34, 475--487 (1982; Zbl 0476.14010)], in this direction, determines the good reduction by analyzing ramification properties of the \(\mathfrak{p}\)-adic Tate module of \(E\), where \(\mathfrak{p}\subset A\) is a non-zero prime ideal of \(A\). On the other hand, one can also see \textit{U. Hartl} and \textit{S. Hüsken}'s result [Ann. Sc. Norm. Super. Pisa, Cl. Sci. (5) 15, Spec. Iss., 25--43 (2016; Zbl 1417.11112)] which gives a criterion for good reduction in the language of local shtukas. Let \(M\) be the Anderson \(t\)-motive of \(E\) and \(M_{\infty}\) be the \(\infty\)-adic completion of \(F\otimes_{A}M\) which gives rise to a pure isocrsytal and hence the \(\infty\)-adic Tate module \(T(M_{\infty})\) (see Section 6 and Section 9). The paper under review, which takes a different path than the aforementioned results, provides a criterion for good reduction in terms of ramification properties of \(T(M_{\infty})\) of \(E\). In particular, it is proved in Theorem 9.5.4 that \(E\) has good reduction over Spec\(R\) if and only if \(T(M_{\infty})\) is unramified at the valuation \(v\). In what follows, we briefly explain the idea of the proof of Theorem 9.5.4. The main tool used is the construction of the Tate module of a given pure isocrystal. Then in Theorem 8.3.6, the author gives an equivalent condition for a pure isocrystal having good reduction in terms of a certain property of the \(\infty\)-adic Tate module. Later on, by constructing a fully faithful functor between pure isocrystals of good reduction and \(\sigma\)-bundles (Theorem 8.4.7) as well as using the work of \textit{V. Drinfeld} [Elliptic modules, Mat. Sb. (N.S.), 94, 594--627 (1974)] and \textit{F. Gardeyn} [J. Number Theory, No. 97, 447--471 (2002; Zbl 1053.11054)] and [Math. Ann. 325, No. 4, 795--828 (2003; Zbl 1108.11046)], the proof of Theorem 9.5.4 is concluded. It is important to mention that the author is also able to describe a certain representation \(\rho_{\infty}\) of the Weil group \(W_{K}\) of \(K\) constructed by \textit{J. K. Yu} [Comp. Math. 138, No. 2 140, 189--197 (2003; Zbl 1039.11039)] in terms of the action of \(W_{K}\) on \(T(M_{\infty})\) (Theorem 9.2.2) and hence the good reduction criterion obtained in this paper can be interpreted via \(\rho_{\infty}\) (see Theorem 1.1.1).