ScalingStacks

3.4 NA Monge-Ampère measure [00B8]

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 · Original author HTML

3.4 NA Monge-Ampère measure

The NA Monge-Ampère measure [11] is defined through intersection theory in a somewhat counterintuitive manner. As a motivation, we consider the complex analytic setting of an snc model 𝒳\mathcal{X} over an algebraic curve, equipped with a Hermitian line bundle (ℒ,h)(\mathcal{L},h) with curvature form θ\theta in the class c1​(ℒ)c_{1}(\mathcal{L}). Then θn\theta^{n} defines a family of nn-forms on XtX_{t}, such that ∫Xtθn\int_{X_{t}}\theta^{n} equals the intersection number (Ln)(L^{n}). The question is to describe the limit of these nn-forms, when we view XtX_{t} as converging to the dual intersection complex Δ𝒳\Delta_{\mathcal{X}} (cf. section 3.1).

We write X0=∑i∈Ibi​EiX_{0}=\sum_{i\in I}b_{i}E_{i}. Recall that the regions on XtX_{t} corresponding to the faces in the dual intersection complex are from the algebraic perspective only small neighbourhoods of EJE_{J}. Thus the limit of θn|Xt\theta^{n}|_{X_{t}} can only be supported at the vertices of Δ𝒳\Delta_{\mathcal{X}}, which correspond to the components EiE_{i}. The amount of delta masses concentrated at the vertices are

bi​∫Eiθn=bi​ℒn⋅Ei,b_{i}\int_{E_{i}}\theta^{n}=b_{i}\mathcal{L}^{n}\cdot E_{i},

where bib_{i} appears due to the multiplicity of the sheets. Reassuringly,

∑ibi​ℒn⋅Ei=(Ln)\sum_{i}b_{i}\mathcal{L}^{n}\cdot E_{i}=(L^{n})

gives the correct total mass.

Back to the NA setting, given a model ℚ\mathbb{Q}-line bundle ℒ→𝒳\mathcal{L}\to\mathcal{X} for L→XKL\to X_{K}, we write 𝒳0=∑ibi​Ei\mathcal{X}_{0}=\sum_{i}b_{i}E_{i}, and denote the divisorial points associated to EiE_{i} as qiq_{i}. We can then define the NA Monge-Ampère measure for the model metric ‖⋅‖ℒ\left\lVert\cdot\right\rVert_{\mathcal{L}} as the following signed atomic measure supported at qi∈XKa​nq_{i}\in X_{K}^{an}:

M​A​(‖⋅‖ℒ)=∑Eibi​(ℒn⋅Ei)​δqiMA(\left\lVert\cdot\right\rVert_{\mathcal{L}})=\sum_{E_{i}}b_{i}(\mathcal{L}^{n}\cdot E_{i})\delta_{q_{i}}

This definition is compatible with pullback of line bundles by the projection formula, and ensures the total mass is the intersection number (Ln)(L^{n}). If ‖⋅‖ℒ\left\lVert\cdot\right\rVert_{\mathcal{L}} is furthermore semipositive, then the intersection numbers are non-negative, so M​A​(‖⋅‖ℒ)MA(\left\lVert\cdot\right\rVert_{\mathcal{L}}) is a measure. Now a general continuous semipositive metric on LL is the uniform limit of a sequence of continuous semipositive model metrics [2, Cor. 8.8], and its NA MA measure can be defined as the unique limiting Radon measure of the NA MA measures for the sequence [4, Cor. 3.5].

The theory of NA MA measures bears strong resemblance to the complex pluripotential theory for complex MA measures [4][5]. The difficulty of working with them lies in the highly nonlocal appearance of the above definition, ultimately caused by the tension between the algebraic limit 𝒳0\mathcal{X}_{0} and the tropical limit Δ𝒳\Delta_{\mathcal{X}}. The recent result of Vilsmeier [45] offers a more concrete perspective:

Proposition 3.7.

(NA MA-real MA comparison) Let (𝒳,ℒ)(\mathcal{X},\mathcal{L}) be a semistable snc model of (XK,L)(X_{K},L), and Int​(ΔJ)\text{Int}(\Delta_{J}) be an nn-dimensional open face of Δ𝒳\Delta_{\mathcal{X}}. Recall the retraction map r𝒳:XKa​n→Δ𝒳r_{\mathcal{X}}:X_{K}^{an}\to\Delta_{\mathcal{X}}. Let ϕ∈C0​(XKa​n)\phi\in C^{0}(X_{K}^{an}) be the potential of a semipositive metric ‖⋅‖ℒ​e−ϕ\left\lVert\cdot\right\rVert_{\mathcal{L}}e^{-\phi}, and suppose ϕ=ϕ∘r𝒳\phi=\phi\circ r_{\mathcal{X}} on r𝒳−1​(ΔJ)r_{\mathcal{X}}^{-1}(\Delta_{J}), then on Int​(ΔJ)\text{Int}(\Delta_{J}) the pushforward of the NA MA measure

r𝒳∗MA(‖⋅‖e−ϕ)=n!MAℝ(ϕ|Int​(ΔJ))r_{\mathcal{X}*}MA(\left\lVert\cdot\right\rVert e^{-\phi})=n!MA_{\mathbb{R}}(\phi|_{\text{Int}(\Delta_{J})})

equals the real MA measure of the convex function ϕ|Int​(ΔJ)\phi|_{\text{Int}(\Delta_{J})} up to a factor n!n!.

The rigorous proof of this comparison uses intersection theory. We present a heuristic calculation that hopefully makes the relation between NA MA measure and real MA measure more intuitive to differential geometers. Consider an snc model 𝒳\mathcal{X} over an algbebraic curve as in the motivation, and assume furthermore that it is semistable. Recall our heuristic dictionary that a metric ‖⋅‖\left\lVert\cdot\right\rVert on L→XKL\to X_{K} should encode a family of Hermitian metrics hth_{t} on L→XtL\to X_{t}, such that ht1/|log⁡|t||→‖⋅‖2h_{t}^{1/|\log|t||}\to\left\lVert\cdot\right\rVert^{2} in the hybrid topology, and the NA MA measure of ‖⋅‖\left\lVert\cdot\right\rVert should be the limit of the measures associated to the curvature forms of h|Xth|_{X_{t}}. We now focus on the neighbourhood of an nn-dimensional open face Int​(ΔJ)⊂Δ𝒳⊂Δ𝒳⊔X\text{Int}(\Delta_{J})\subset\Delta_{\mathcal{X}}\subset\Delta_{\mathcal{X}}\sqcup X, where we have local coordinates z0,…​znz_{0},\ldots z_{n} with ∏0nzi=t\prod_{0}^{n}z_{i}=t, and xi=log⁡|zi|log⁡|t|x_{i}=\frac{\log|z_{i}|}{\log|t|}. In the local picture we identify metrics with potentials, so ‖⋅‖∼e−ϕ\left\lVert\cdot\right\rVert\sim e^{-\phi}, and after ignoring C0C^{0}-fluctuation effects ht1/|log⁡|t||∼e−2ϕ∘Log𝒳h_{t}^{1/|\log|t||}\sim e^{-2\phi\circ\text{Log}_{\mathcal{X}}}. Imposing more smoothness assumptions, the curvature form of hth_{t} is approximately

|log⁡|t||​d​dc​ϕ∘Log𝒳=−12​π​∑1≤i,j≤n∂2ϕ∂xi​∂xj​d​xi∧d​arg​(zj).|\log|t||dd^{c}\phi\circ\text{Log}_{\mathcal{X}}=\frac{-1}{2\pi}\sum_{1\leq i,j\leq n}\frac{\partial^{2}\phi}{\partial x_{i}\partial x_{j}}dx_{i}\wedge d\text{arg}(z_{j}).

The NA MA measure should agree with the limiting pushforward measure

limt→0Log𝒳∗(|log|t||ddcϕ∘Log𝒳)n=n!det(D2ϕ)|dx1…dxn|=n!MAℝ(ϕ)\lim_{t\to 0}\text{Log}_{\mathcal{X}*}(|\log|t||dd^{c}\phi\circ\text{Log}_{\mathcal{X}})^{n}=n!\det(D^{2}\phi)|dx_{1}\ldots dx_{n}|=n!\text{MA}_{\mathbb{R}}(\phi)

which equals the real MA measure up to the factor n!n!.

Remark 3.8.

In this heuristic calculation, the assumption for ϕ\phi to factor through the retraction map allows us to replace the hybrid space X⊔XKa​nX\sqcup X_{K}^{an} by its finite approximation X⊔Δ𝒳X\sqcup\Delta_{\mathcal{X}}.

Remark 3.9.

The above calculation is similar to the formalism developed by Chambert-Loir and Ducros, see [12, Lemma 5.7.1].

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