Generation of jets and Fujita's jet ampleness conjecture on toric varieties

From MaRDI portal
Publication:2664595



Abstract: Jet ampleness of line bundles generalizes very ampleness by requiring the existence of enough global sections to separate not just points and tangent vectors, but also their higher order analogues called jets. We give sharp bounds guaranteeing that a line bundle on a projective toric variety is k-jet ample in terms of its intersection numbers with the invariant curves, in terms of the lattice lengths of the edges of its polytope, in terms of the higher concavity of its piecewise linear function and in terms of its Seshadri constant. For example, the tensor power k+n−2 of an ample line bundle on a projective toric variety of dimension ngeq2 always generates all k-jets, but might not generate all (k+1)-jets. As an application, we prove the k-jet generalizations of Fujita's conjectures on toric varieties with arbitrary singularities.


Let \(X\) be a projective variety and let \(k\) be a non-negative integer. A line bundle \(L\) on \(X\) is said to be \(k\)-jet ample if for any set \(x_1, \dots , x_r\) of distinct closed points of \(X\), and any positive integers \(k_1, \dots , k_r\) such that \(\sum_{i=1}^r k_i= k + 1\), the evaluation map \(H^0(X, L) \to H^0\big(X, L \otimes \mathcal O_X/(m_{x_1}^{k_1}\otimes \dots \otimes m_{x_r}^{k_r})\big)\) is surjective, where \(m_{x_i}\) denotes the ideal sheaf of the point \(x_i\). This notion, which generalizes those of spannedness (\(k=0\)) and very ampleness (\(k=1\)), is due to \textit{M. Beltrametti} et al. [Duke Math. J. 58, No. 2, 425--439 (1989; Zbl 0702.14010)] and has been largely studied in the literature, also in comparison with other notions of higher order embedding. The problem of finding sufficient conditions for jet ampleness has been studied for several classes of varieties. In the case of a smooth projective toric variety \(X\), Di Rocco showed that a line bundle \(L\) is \(k\)-jet ample if \(L \cdot C \geq k\) for every torus-invariant curve \(C \subset X\) [\textit{S. Di Rocco}, Math. Z. 231, No. 1, 169--188 (1999; Zbl 0941.14020)]. In the paper under review the authors show that on a singular toric variety \(X\) of dimension \(n \geq 2\), \(L\) is \(k\)-jet ample provided that \(L \cdot C \geq k+n-2\) for every \(C\) as above. This follows from a more subtle formulation of the previous condition through a constant \(\Gamma_X\) associated with the toric variety, which satisfies \(0 \leq \Gamma_X \leq n-2\) and is \(0\) for smooth \(X\). Moreover, in line with what Di Rocco proved in the smooth case the authors show that the above condition on \(L\) can be equivalently rephrased in terms of the lattice lengths of the edges of the associated polytope, in terms of the higher concavity of its piecewise linear function and in terms of its Seshadri constant. Furthermore, the \((k + n - 2)\)th tensor power of an ample line bundle on a projective toric variety of dimension \(n \geq 2\) is always \(k\)-jet ample, but might not be \((k+1)\)-jet ample, which shows the sharpness of the bound above. As an application, the authors prove a jet ampleness variant of Fujita's conjectures for singular toric varieties. In particular, they show that given an ample line bundle \(L\) on a projective \(n\)-dimensional Gorenstein toric variety \(X\), not isomorphic to \(\mathbb P^n\), the adjoint bundle \(K_X+(n+k)L\) is \(k\)-jet ample for every \(k \geq 0\).



Cites work









This page was built for publication: Generation of jets and Fujita's jet ampleness conjecture on toric varieties

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2664595)