Iwasawa invariants of some non-cyclotomic \(\mathbb{Z}_p\)-extensions (Q1708065)

Let \(p\) be prime, let \(K\) be a number field, and let \(K_\infty/K\) be a \(\mathbb{Z}_p\)-extension of \(K\). Let \(K_n\) be the subfield of \(K_\infty\) that has degree \(p^n\) over \(K\) and let \(p^{e_n}\) be the power of \(p\) dividing the class number of \(K_n\). A well-known result of Iwasawa says that there exist integers \(\lambda, \mu, \nu\), independent of \(n\), such that \[ e_n = \mu p^n+\lambda n+\nu \] for all sufficiently large \(n\). When \(K_\infty\) is the cyclotomic \(\mathbb{Z}_p\)-extension of \(K\), it is conjectured that \(\mu=0\), and this has been proved by \textit{B. Ferrero} and the second author in [Ann. Math. (2) 109, 377--395 (1979; Zbl 0443.12001)] when \(K/\mathbb{Q}\) is abelian. In [in: Number Theory, algebr. Geom., commut. Algebra, in Honor of Yasuo Akizuki, 1--11 (1973; Zbl 0281.12005)], \textit{K. Iwasawa} showed that there exist non-cyclotomic \(\mathbb{Z}_p\)-extensions with \(\mu>0\). A natural question is what are the actual values of the Iwasawa invariants for these \(\mathbb{Z}_p\)-extensions ? In the paper under review, the authors give examples of Iwasawa's construction when \(p=3\) and= \(p=2\), and they show that in many cases it is possible to evaluate the invariants \(\lambda, \mu, \nu\). Several explicit examples are given in Sections 9 and 11. In order to evaluate the Iwasawa invariants, the authors develop in Sections 5 and 6 some techniques that could prove useful for studying Iwasawa invariants in situations other than number fields, especially in cases where \(\mu\) does not necessarily vanish. The eventual goal is that by obtaining some exact values for \(\mu\) when it is nonzero, they are able to understand better what controls \(\mu\). In Section 10, the authors work in this direction by studying the effects of the ambiguous ideal classes and the higher ambiguous groups.