Inversion of adjunction for arithmetic log pairs

Let C=Spec(OK)C=\operatorname{Spec}(\mathcal{O}_{K}) be the spectrum of the ring of integers of a number field, let π ⁣:XC\pi\colon\mathcal{X}\to C be as in the preceding notation, and let Xc=π1(c)\mathcal{X}_{c}=\pi^{-1}(c) for a closed point cCc\in C. Let DD be an effective Q\mathbb{Q}-Cartier Q\mathbb{Q}-divisor whose support contains no component of Xc\mathcal{X}_{c}, and let t0t\geq 0. Inversion of adjunction. Each of the following equivalences holds: (X,Xc+tD)(\mathcal{X},\mathcal{X}_{c}+tD) is log-canonical in a neighborhood of Xc\mathcal{X}_{c} if and only if (Xc,DXc)(\mathcal{X}_{c},D|_{\mathcal{X}_{c}}) is geometrically semi-log-canonical, meaning semi-log-canonical after base change to an algebraic closure of κ(c)\kappa(c); and (X,Xc+tD)(\mathcal{X},\mathcal{X}_{c}+tD) is purely log terminal in a neighborhood of Xc\mathcal{X}_{c} if and only if (Xc,DXc)(\mathcal{X}_{c},D|_{\mathcal{X}_{c}}) is log-terminal. This is proposed because inversion of adjunction in the arithmetic setting is not established, motivating the formulation via log pairs; no resolution status is supplied.

Sources & referencesView supporting material

Primary source

Yuji Odaka, “Canonical Kahler metrics and Arithmetics – Generalising Faltings heights”, arXiv:1508.07716 (2016).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.