Toric rigid spaces (Q2463803): Difference between revisions

Let \(K\) be a field complete with respect to a non-Archimedean valuation, with integral subring \(R\) and residue characteristic~\(p\). The authors mimic the usual construction of toric varieties so as to get rigid spaces: for a convex rational polyhedral cone \(\sigma\in N\otimes{\mathbb R}\), where \(N\cong {\mathbb Z}^n\) and \(M=N^\vee\), they put \(S_\sigma=\sigma^\vee \cap M\) and consider the affinoid algebra \[ K\langle S_\sigma\rangle =K \otimes_R\lim\limits_\leftarrow R/p^nR [S_\sigma] \] and \(U_\sigma={\text{ Sp}}(K\langle S_\sigma\rangle)\). By gluing such rigid spaces \(U_\sigma\) for \(\sigma\) in a fan \(\Delta\) one obtains a rigid space called \(R(\Delta)\). It is simple to prove that \(R(\Delta)\) is separated and integral. However, it is not usually equal to the rigid analytification \(RT(\Delta)\) of the toric scheme \(T(\Delta)\) associated with \(\Delta\). The authors show that \(R(\Delta)=RT(\Delta)\) if \(\Delta\) is a complete fan: they attribute both the statement and the proof to Strauch. They also describe \(R(\Delta)\) in some simple examples, and investigate reductions. They show that the reduction of \(R(\Delta)\) is the toric scheme \(T(\Delta)\) over the residue field of \(K\).