Motivic zeta functions of degenerating Calabi-Yau varieties (Q1709769)

The motivic monodromy conjecture predicts a relation between poles of the motivic zeta function and the monodromy eigenvalues of a Calabi-Yau variety over \(\mathbb C ((t))\). This conjecture was already proved for tamely ramified abelian varieties in arbitrary characteristic in [\textit{L. H. Halle} and \textit{J. Nicaise}, Adv. Math. 227, No. 1, 610--653 (2011; Zbl 1230.11076)], using techniques that rely heavily on the existence of Néron models. Galois equivariant Kulikov models form a class of models which maintain some of the key properties of Néron models, and the authors prove the motivic monodromy conjecture for Calabi-Yau varieties over \(\mathbb C((t))\) which admit such models and describe the unique pole of their motivic zeta functions. Moreover, the existence of Galois equivariant Kulikov models is proven for some classes of Calabi-Yau varieties, a remarkable one being abelian varieties, and it was shown that there exist Calabi-Yau varieties which do not admit such models, see Example 5.3.5 for an example of \(K3\) surface which does not admit a Galois equivariant Kulikov model. Finally, it is shown that, for a general Calabi-Yau variety \(X\) over \(\mathbb C((t))\), the highest order of poles is one plus the dimension of the essential skeleton of \(X\). This provides a link between the motivic zeta function and the non-archimedean SYZ conjecture suggested in [\textit{M. Kontsevich} and \textit{Y. Soibelman}, Prog. Math. 244, 321--385 (2006; Zbl 1114.14027)]. In the appendix to the paper motivic integration for algebraic spaces is developed. This was a technical requirement for some constructions, since Galois equivariant Kulikov models are, in general, algebraic spaces.