Toric co-Higgs sheaves (Q2022447)

For a complex toric variety \(X_{\Sigma}\) given by a fan \(\Sigma\subseteq N_{\mathbb R}=N\otimes_{\mathbb Z}{\mathbb R}\) for a lattice \(N\) and acting torus \(T=N\otimes_{\mathbb Z}{\mathbb C}\), by \textit{A. Klyachko} [Math. USSR, Izv. 35, No. 2, 337--375 (1990; Zbl 0706.14010)] a \textit{toric sheaf} \({\mathcal E}\), (that is, an \({\mathcal O}_X\)-module with an action of the torus \(T\) which is linear on the fibers and is compatible with the \(T\)-action of \(X\)) corresponds to the complex vector space \(E={\mathcal E}_1/{\mathfrak m}_{X,1}{\mathcal E}_1\) (where \({\mathcal E}_1\) is the stalk or \({\mathcal E}\) at \(1\in T\subseteq X\) and \({\mathfrak m}_{X,1}\) is the maximal ideal of \({\mathcal O}_{X,1}\)) together with a decreasing \({\mathbb Z}\)-filtration \(E^{\bullet}_{\rho}\) indexed by the rays \(\rho\in \Sigma(1)\). Following \textit{I. Biswas} et al., [Ill. J. Math. 65, No. 1, 181--190 (2021; Zbl 1465.14048)] a \textit{toric co-Higgs sheaf} is a pair \(({\mathcal E},\Phi)\) consisting of a toric sheaf \({\mathcal E}\) on a toric variety \(X_{\Sigma}\) and a \textit{Higgs field}, i.e., a \(T\)-equivariant \({\mathcal O}_X\)-morphism \(\Phi:{\mathcal E}\rightarrow{\mathcal E}\otimes_{{\mathcal O}_X}{\mathcal T}_X\) such that \(\Phi\wedge\Phi=0\), where \({\mathcal T}_X\) is the tangent sheaf of \(X\). Dropping the integrability condition \(\Phi\wedge\Phi=0\), the pair \(({\mathcal E},\Phi)\) is called a \textit{toric pre-co-Higgs sheaf} and \(\Phi\) a \textit{pre-co-Higgs field}. The paper under review considers general co-Higgs fields and not only \(M\)-homogeneous ones as in [Biswas et al., loc. cit]. To study these general co-Higgs sheaves, the authors start characterizing pre-co-Higgs fields using Klyachko's formalism by means of associated contractions in Theorem 8 and then show that every co-Higgs field defines a commutative finitely generated \({\mathbb C}[M]\)-algebra, the \textit{Higgs algebra}. Next, the authors introduce some combinatorial invariants: First, using that a pre-co-Higgs field is a direct sum \(\Phi=\sum \Phi^r\) of maps \(\Phi^r:{\mathcal E}\rightarrow {\mathcal E}\otimes_{{\mathcal O}_X}{\mathcal T}_X\) of degree \(r\in M\), they define the corresponding \textit{Higgs polytope} of \(\Phi\) as the convex hull in \(M_{\mathbb R}\) of its support \(\text{supp}(\Phi)=\{r\in M: \Phi^r\neq 0\}\subseteq M\). The convex hull of the totality of degrees \(r\in M\) of all possible toric pre-co-Higgs fields defines a second polytope, the \textit{Higgs range}. After proving some properties of these polytopes, in the last two sections of the paper they are calculated for several smooth toric surfaces: They compute the Higgs range of the projective plane and Hirzebruch and Fano surfaces, and they also compute the Higgs polytope for some del Pezzo surfaces. The whole paper includes many illustrative examples with explicit calculations nicely complementing the developments.