On Mordell-Weil groups and congruences between derivatives of twisted Hasse-Weil \(L\)-functions (Q1692369): Difference between revisions

Let \(A\) be an abelian variety over a number field \(k\) and let \(F/k\) be a finite Galois extension with Galois group \(G\). Let \(A_F\) denote the base change of \(A\) to \(F\). Then the ETNC for the pair \((h^1(A_F)(1), \mathbb Z[G])\), as formulated by \textit{D. Burns} and \textit{M. Flach} [Doc. Math., J. DMV 6, 501--570 (2001; Zbl 1052.11077)], may be seen as an equivariant refinement of the Birch and Swinnerton-Dyer conjecture for \(A\) for all intermediate fields of \(F/k\). This formulation, however, is rather inexplicit in nature, and the main purpose of this article is to give an explicit interpretation of the `\(p\)-part' (\(p\) an odd prime) of the ETNC under certain restrictions on the variety \(A\) and the ramification in \(F/k\). These hypotheses include certain standard conjectures as the finiteness of the Tate-Shafarevich group and a bunch of technical assumptions. The most restrictive seems to be hypothesis (h) that \(p\) is unramified in \(F/ \mathbb Q\), which is assumed from section 4.4 onwards. Based on their results on the Galois structure of Selmer groups established in [Int. Math. Res. Not. 2015, No. 22, 11909--11933 (2015; Zbl 1334.11043)], this reformulation is then used to verify the \(p\)-part of the ETNC for a family of elliptic curves over \(\mathbb Q\) and (generalized) dihedral extensions \(F/\mathbb Q\). Here, the rank of the Mordell-Weil group might be positive, and \(p\) divides the order of \(G\) (which is more interesting and more difficult than the case \(p \nmid |G|\)). Moreover, the authors discuss the case of elliptic curves and \(S_3\)-extensions in more detail (\(S_3\) the symmetric group on \(3\) letters), and use their reformulation to numerically verify the ETNC in a few more cases with the help of computer (using Sage and Magma). Note that there already is a subsequent article of \textit{W. Bley} and the second author [J. Reine Angew. Math. 722, 105--135 (2017; Zbl 1453.11142)].