ScalingStacks

Example 4.3 [035L]

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

Example 4.3

Let T=𝔾mrT={\mathbb{G}}_{m}^{r} be the split multiplicative torus of rank rr with coordinates z1,…,zrz_{1},\dots,z_{r}. Then a point xx of Tan{T^{\rm an}} could be visualized by the coordinates z1​(x),…,zr​(x)∈ℋ⁡(x)z_{1}(x),\dots,z_{r}(x)\in{\mathscr{H}}(x) and the multiplicative seminorm corresponding to xx is given by |f⁡(x)|=|f⁡(z1​(x),…,zr​(x))||f(x)|=|f(z_{1}(x),\dots,z_{r}(x))| for every Laurent polynomial ff on TT. Conversely, every field extension L/KL/K with an absolute value extending the given absolute value on KK and every (β1,…,βr)∈Lr(\beta_{1},\dots,\beta_{r})\in L^{r} give rise to a point x∈Tanx\in{T^{\rm an}} by |f⁡(x)|:=|f⁡(β1,…,βr)||f(x)|:=|f(\beta_{1},\dots,\beta_{r})|. Note that LL and (β1,…,βr)(\beta_{1},\dots,\beta_{r}) are not uniquely determined by xx.

In particular, we get an inclusion of T⁡(K)T(K) into Tan{T^{\rm an}}. For every x∈T⁡(K)x\in T(K), we have d⁡(x)=0d(x)=0. However, there can be also other points with d⁡(x)=0d(x)=0. If T=𝔾m1T={\mathbb{G}}_{m}^{1}, then precisely the points of type 1 (i.e. the KK-rational points) and the points of type 4 satisfy d⁡(x)=0d(x)=0 (see [Be90], 1.4.4).

Returning to the case T=𝔾mrT={\mathbb{G}}_{m}^{r}, there are some distinguished points of Tan{T^{\rm an}} which behave completely different than KK-rational points. For positive real numbers s1,…,srs_{1},\dots,s_{r}, we define the associated weighted Gauss norm on K⁡[T]K[T] by

|f|𝐬:=max𝐦∈ℤr⁡|α𝐦|​𝐬𝐦|f|_{\mathbf{s}}:=\max_{{\mathbf{m}}\in{\mathbb{Z}}^{r}}|\alpha_{\mathbf{m}}|{\mathbf{s}}^{\mathbf{m}}

for every Laurent polynomial f=∑𝐦∈ℤrα𝐦​𝐳𝐦∈K⁡[T]=K⁡[z1±1,…,zr±1]f=\sum_{{\mathbf{m}}\in{\mathbb{Z}}^{r}}\alpha_{\mathbf{m}}{\mathbf{z}}^{\mathbf{m}}\in K[T]=K[z_{1}^{\pm 1},\dots,z_{r}^{\pm 1}]. It follows from the Gauss Lemma that the weighted Gauss norm is a multiplicative seminorm giving rise to a point η𝐬∈Tan\eta_{\mathbf{s}}\in{T^{\rm an}}. The set S(Tan):={η𝐬∣s1>0,…,sr>0}S({T^{\rm an}}):=\{\eta_{\mathbf{s}}\mid s_{1}>0,\dots,s_{r}>0\} is called the skeleton of Tan{T^{\rm an}}. Every point η𝐬∈S⁡(Tan)\eta_{\mathbf{s}}\in S({T^{\rm an}}) satisfies d⁡(η𝐬)=rd(\eta_{\mathbf{s}})=r (see [Du12], (0.12) and (0.13)).

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