MMP for co-rank one foliations on threefolds (Q2049901): Difference between revisions

This paper is an important step toward the Minimal Model Program for foliated varieties. Partially from the introduction: The classical MMP predicts that a complex projective manifold is either uniruled, or admits a minimal model: it is birational to a (possibly singular) projective variety with nef canonical divisor. After the works of [\textit{L. G. Mendes}, Bol. Soc. Bras. Mat., Nova Sér. 31, No. 2, 127--143 (2000; Zbl 0979.32017); \textit{M. Brunella}, Birational geometry of foliations. Rio de Janeiro: Instituto Nacional de Matemática Pura e Aplicada (IMPA) (2004; Zbl 1082.32022); \textit{M. McQuillan}, Pure Appl. Math. Q. 4, No. 3, 877--1012 (2008; Zbl 1166.14010)] we expect a similar picture holds true for foliated varieties. More specifically, assuming that \(X\) is a normal complex projective variety and \(\mathcal F\) is a foliation with mild singularities, it is conjectured that either \(\mathcal F\) is uniruled, i.e. \(X\) is covered by rational curves which are tangent to \(\mathcal F\), or \(\mathcal F\) admits a minimal model, i.e. \(X\) is birational to a projective variety \(Y\) such that the transformed foliation on \(Y\) admits a nef canonical divisor. Many of the main goals of the program were carried out successfully in the case of rank one foliations by M. McQuillan and, in any rank, it is expected to follow the main steps of Mori's program. The goal of this paper is to show the existence of flips for foliations of co-rank one on a complex projective threefold and present several applications, under some natural assumptions on the singularities. The canonical bundle \(K_{\mathcal F}\) of volume forms along leaves of \(\mathcal F\) plays a central role. The authors consider foliated pairs \((\mathcal F,\delta)\) where \(\Delta\) is typically a positive \(\mathbb Q\)-diviseur such that \(K_{\mathcal F}+\Delta\) is \(\mathbb Q\)-Cartier. Then the condition F-dlt (foliated divisorial log terminal) imposes some restriction of the possible type of singularities for \(X\) and \(\mathcal F\), in particular that they are log canonical. The first result shows the existence of flips in great generality, therefore extending results of the second author in [\textit{C. Spicer}, Compos. Math. 156, No. 1, 1--38 (2020; Zbl 1428.14025)]. Next, the authors provide the existence of a minimal model for F-dlt pairs \((\mathcal F,\delta)\) under some positivity assumption on \(\Delta\). The second part of the paper deals with applications when \(X\) is normal of dimension at most three (existence of F-dlt modifications and F-terminalisations).