The closed cone of a rational series is rational polyhedral (Q401947): Difference between revisions

Let \(X\) be a projective algebraic variety over a field \(k\). For each non-negative integer \(d\), and a homology class \(d\) in \(H_{2p}(X,\mathbb Z)/\text{torsion}\), denote \(C_{p,d}(X)\) the Chow variety parametrizing effective \(p\)-cycles of degree \(d\) in \(X\). \textit{J. Elizondo} [Compos. Math. 94, No. 3, 297--310 (1994; Zbl 0845.14005)] defined the Euler-Chow series of \(p\)-dimensional cycles on \(X\) as \[ EC_{X,p}(t) = \sum_{d\in H_{2p}(X,\mathbb Z)/\text{torsion}} e(C_{p,d}(X))t^d, \] where \(e(C_{p,d}(X))\) is the Euler characteristic to \(C_{p,d}(X)\). Elizondo showed that the Euler-Chow series \(EC_{X,p}(t)\) is rational for any simplicial projective toric variety \(X\) and any \(p\). He asked whether the Euler-Chow series for a general variety \(X\) is rational or not. He suggested to consider the blow-up \(X\) of \(\mathbb P^2\) at \(9\) general points in general position for a counterexample. In this setting the pseudo-effective cone of \(X\) is not a finite rational polyhedral cone by \textit{M. Nagata} [Mem. Coll. Sci., Univ. Kyoto, Ser. A 33, 271--293 (1960; Zbl 0100.16801)]. In the current paper under review, the authors shows that for a series \(f\) to be rational the cone generated by the exponents of the monomials with nonzero coefficients of \(f\) has to be a rational cone. As an application, they show that the suggestion of Elizondo does give a counterexample to the rationality of the Euler-Chow series.