Higher Chow cycles on abelian surfaces and a non-Archimedean analogue of the Hodge-\(\mathcal{D}\)-conjecture (Q2877498)

Consider a simple, principally polarized, abelian surface \(A\) over a \(p\)-adic local field \(\mathcal K_p\). Let \(\mathcal A\) be the Néron model of \(A\) over \(\mathcal O_{{\mathcal K_p}}\) where \(p \neq 2\) is an odd prime of good non-supersingular reduction and let \(\mathcal A_p\) be the special fibre. Aim of the paper is the proof of the surjectivity of the boundary arrow \( \partial : \text{CH}^2(A,1)\otimes{\mathbb Q} {\rightarrow} \mathrm{CH}^1(\mathcal A_p) \otimes \mathbb Q\). Recall that the `Archimedean' Hodge-\(\mathcal D\)-conjecture of Beilinson asserts that the regulator map to Deligne cohomology is surjective. The author explains how \(\partial\) can be understood as a non-Archimedean analogue of the Beilinson regulator map and therefore he refers to his result as a proof of the \textit{non-Archimedean Hodge-\(\mathcal D\)-conjecture} in the present setting.NEWLINENEWLINETo prove surjectivity of \(\partial\) one needs to have geometrical control of the rank of the Néron-Severi group of the non singular surface \(\mathcal A_p\). This can be done concretely, by looking to the related Kummer surface and to the associated plane. Obstruction to the surjectivity could come only from unexpected, new classes in \(\mathrm{NS}(\mathcal A_p)\). The author shows that those cycles correspond to the union of two rational curves on the Kummer. Deformation theory of such couples is at the heart of the clever construction of the element in \(\mathrm{CH}^2(A,1)\) whose boundary gives back the new class one had started with.