On \(p\)-adic zeta-functions associated to the positive topology of algebraic number fields (Q1364342)

Let \(K\) be an algebraic number field and \(O_K\) the ring of integers in \(K\). In his previous paper [\textit{A. Schmidt}, Invent. Math. 123, 575-610 (1996; Zbl 0867.11072)], the author introduced a new Grothendieck topology, called the positive topology, over \(O_K\). When \(K\) is finite over \(\mathbb{Q}\), \(\text{Spec} (O_K)\) with this topology behaves very similarly to the étale site over a complete curve over a finite field. Let \(p\) be an odd prime number and \(X_\infty\) the normalization of \(\text{Spec} (O_K)\) in the cyclotomic \(\mathbb{Z}_p\)-extension of \(K\). In this paper, the author defines the positive \(p\)-adic zeta function in terms of characteristic polynomials of cohomology groups of \(X_\infty\) for the positive topology. Following the analogy with the case of a curve over a finite field, he shows by the global duality theorem given in the paper cited above that if \(K\) contains a primitive \(p\)-th root of unity and if \(p\) satisfies a certain condition (which is satisfied for all but only finitely many \(p)\), then the positive \(p\)-adic zeta function satisfies a functional equation. When \(K/k\) is an abelian extension and satisfies certain technical conditions, then the author also defines the positive \(p\)-adic \(L\)-function associated to the character of the Galois group of \(K/k\). The positive \(p\)-adic zeta function of \(K\) splits into a product of positive \(p\)-adic \(L\)-functions in the usual way. Then he shows that if \(k\) is totally real and ``totally \(p\)-real'' (e.g. \(k=\mathbb{Q}\) satisfies it), the positive \(p\)-adic \(L\)-functions can be calculated at least up to Greenberg's conjecture (see [\textit{R. Greenberg}, Am. J. Math. 98, 263-284 (1976; Zbl 0334.12013)], and also [\textit{H. Ichimura} and \textit{H. Sumida}, Int. J. Math. 7, 721-744 (1996; Zbl 0881.11075)] and its references) in terms of the classical \(p\)-adic \(L\)-functions defined by Kubota-Leopoldt and Deligne-Ribet. In particular, under certain conditions, the positive \(p\)-adic zeta function of \(K\) can be written in terms of the classical \(p\)-adic zeta function of the maximal totally real subfield of \(K\).