ScalingStacks

Proof. [05BI]

Original official author HTML, exact retained edition. Historical TeX conversion verdicts remain unchanged. Cited-edition alignment and mathematical self-containment are not assessed.

Complete original source context Β· Original author HTML

Proof.

Let β„‚K\mathbb{C}_{K} be the completion of an algebraic closure of KK. Then there are exactly deg⁑(S)\Deg(S) points in the special fibre of 𝔛ℂK\mathfrak{X}_{\mathbb{C}_{K}} mapping to SS, hence there are precisely deg⁑(S)\Deg(S) open faces in the skeleton associated to 𝔛ℂK\mathfrak{X}_{\mathbb{C}_{K}} lying over Ο„\tau. As the base change induces an isomorphism of each of these faces with Ο„\tau, we have ΞΉβˆ—β€‹MA⁑(ΞΉβˆ—β€‹h)=deg⁑(S)​MA⁑(h)\iota_{\ast}\MA(\iota^{\ast}h)=\Deg(S)\MA(h). Using this and the invariance of the non-archimedean Monge-AmpΓ¨re measure under base change we may assume K=β„‚KK=\mathbb{C}_{K}. As in the proof of Proposition 5.9 we choose a strongly nondegenerate strictly polystable formal scheme 𝔛′\mathfrak{X}^{\prime} and a surjective Γ©tale morphism Ο†:𝔛′→𝔛\varphi:\mathfrak{X}^{\prime}\rightarrow\mathfrak{X}. Let Ο„β€²\tau^{\prime} be an open face of the skeleton associated to 𝔛′\mathfrak{X}^{\prime} lying over Ο„\tau. As we have seen, Ο†\varphi induces an isomorphism pπ”›β€²βˆ’1​(Ο„β€²)​→~​pπ”›βˆ’1​(Ο„)p_{\mathfrak{X}^{\prime}}^{-1}(\tau^{\prime})\tilde{\rightarrow}p_{\mathfrak{X}}^{-1}(\tau). As in the proof of Corollary 5.7 there is a sequence of rational piecewise affine linear convex functions (hiβ€²)iβˆˆβ„•(h^{\prime}_{i})_{i\in\mathbb{N}} on Ο„β€²\tau^{\prime} converging locally uniformly to hβˆ˜Ο†anh\circ\varphi^{\textup{an}}. Let hih_{i} be the piecewise affine linear functions on Ο„\tau such that hiβˆ˜Ο†an=hiβ€²h_{i}\circ\varphi^{\textup{an}}=h^{\prime}_{i}. By Proposition 5.9 the metrics induced by the hih_{i} are semipositive piecewise β„š\mathbb{Q}-linear metrics on pπ”›βˆ’1​(Ο„)p_{\mathfrak{X}}^{-1}(\tau) which implies that the metric induced by hh is locally semipositive. As the restriction of Ο†\varphi to pπ”›β€²βˆ’1​(Ο„β€²)p_{\mathfrak{X}^{\prime}}^{-1}(\tau^{\prime}) is an isomorphism onto pπ”›βˆ’1​(Ο„)p_{\mathfrak{X}}^{-1}(\tau) we have

c1​(π’ͺΒ―hi∘p𝔛)n=(Ο†|pπ”›β€²βˆ’1​(Ο„β€²))βˆ—β€‹c1​((Ο†|pπ”›β€²βˆ’1​(Ο„β€²))βˆ—β€‹π’ͺΒ―hi∘p𝔛)nc_{1}\left(\overline{\mathcal{O}}^{h_{i}\circ p_{\mathfrak{X}}}\right)^{n}=\left(\varphi\Big|_{p_{\mathfrak{X}^{\prime}}^{-1}(\tau^{\prime})}\right)_{\ast}c_{1}\left(\left(\varphi\Big|_{p_{\mathfrak{X}^{\prime}}^{-1}(\tau^{\prime})}\right)^{\ast}\overline{\mathcal{O}}^{h_{i}\circ p_{\mathfrak{X}}}\right)^{n}

By Corollary 5.5 we have

c1​((Ο†|pπ”›β€²βˆ’1​(Ο„β€²))βˆ—β€‹π’ͺΒ―hi∘p𝔛)n=c1​(π’ͺΒ―hiβˆ˜Ο†an∘p𝔛′)n=n!β‹…MA⁑(hiβˆ˜Ο†an|Ο„β€²).c_{1}\left(\left(\varphi\Big|_{p_{\mathfrak{X}^{\prime}}^{-1}(\tau^{\prime})}\right)^{\ast}\overline{\mathcal{O}}^{h_{i}\circ p_{\mathfrak{X}}}\right)^{n}=c_{1}\left(\overline{\mathcal{O}}^{h_{i}\circ\varphi^{\textup{an}}\circ p_{\mathfrak{X}^{\prime}}}\right)^{n}=n!\cdot\MA\left(h_{i}\circ\varphi^{\textup{an}}\Big|_{\tau^{\prime}}\right).

Hence

c1​(π’ͺΒ―hi∘p𝔛)n=(Ο†|pπ”›β€²βˆ’1​(Ο„β€²))βˆ—β€‹(n!β‹…MA⁑(hiβˆ˜Ο†an|Ο„β€²))=n!β‹…MA⁑(hi).c_{1}\left(\overline{\mathcal{O}}^{h_{i}\circ p_{\mathfrak{X}}}\right)^{n}=\left(\varphi\Big|_{p_{\mathfrak{X}^{\prime}}^{-1}(\tau^{\prime})}\right)_{\ast}\left(n!\cdot\MA\left(h_{i}\circ\varphi^{\textup{an}}\Big|_{\tau^{\prime}}\right)\right)=n!\cdot\MA(h_{i}).

It is easily seen that in Proposition 4.13 we can replace uniform convergence by locally uniform convergence. The claim follows from this fact and continuity of the real Monge-Ampère operator. ∎

Original mathematics by the credited authors. Source-backed reader collection; mathematical self-containment is not assessed.