An algorithm for computing compatibly Frobenius split subvarieties (Q414631)

Throughout \(R\) will denote a commutative ring of finite type over a perfect field \(k\) of prime characteristic \(p > 0\) or a localization of such a ring. We denote by \(F_{*}^e R\) the ring \(R\) viewed as an \(R\)-module via the \(e\)-th iteration of the Frobenius morphism \(F: R \to R\). Given an \(R\)-linear map \(\phi: F_{*}^e R \to R\) and an ideal \(J \subseteq R\), we say that \(J\) is \(\phi\)-compatible if \(\phi(F_{*}^e J) \subseteq J\). An \(R\)-linear map \(\phi: F_{*}^e R \to R\) is said to be a Frobenius splitting if \(\phi(F_{*}^e1)=1\). In this paper the authors describe an algorithm that produces all the \(\phi\)-compatible ideals of a given surjective \(R\)-linear map \(\phi: F_{*}^e R \to R\). In particular, the algorithm applies to the case of Frobenius splittings. Their work is motivated by a result (and its proof) of \textit{S. Kumar} and \textit{V. B. Mehta} [Int. Math. Res. Not. 2009, No. 19, 3595--3597 (2009; Zbl 1183.13006)] and \textit{K. Schwede} [Algebra Number Theory 3, No. 8, 907--950 (2009; Zbl 1209.13013)] who showed that there are only finitely many such \(\phi\)-compatible ideals. The authors also study a variant of the algorithm in the case when \(\phi\) is not necessarily surjective. These algorithms have been implemented in Macaulay2.