Positivity of cotangent bundles

DOI10.1307/MMJ/1260475697MaRDI QIDQ846229zbMATH OpenOpenAlexFDO

Publication date 2 February 2010

Published in Michigan Mathematical Journal (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/0803.0622

ample vector bundle cotangent bundle Kodaira spencer map smooth fibration

Differentials and other special sheaves; D-modules; Bernstein-Sato ideals and polynomials (14F10) Families, moduli, classification: algebraic theory (14J10) (n)-folds ((n>4)) (14J40)

Abstract: In this paper we prove a generalization of a theorem of Schneider, which gives a criterion for a projective surface over the complex numbers to have an ample cotangent bundle. After reviewing different notions of positivity, we introduce a slightly weaker notion of ampleness, which we call quasi-ample, and then are able to extend Schneider's result to higher dimensions.

Review text

Let \(X\) be a complex projective manifold. While there has been a lot of research on the positivity of the canonical bundle \(K_X\), relatively few things are known about the positivity of the cotangent bundle \(\Omega_X\). If \(\Omega_X\) is ample this gives important restrictions on the geometry of \(X\): all the subvarieties of \(X\) are of general type, moreover \(X\) is hyperbolic, i.e. does not admit a holomorphic non-constant map \(\mathbb C \rightarrow X\). Although manifolds with ample cotangent bundle are expected to be reasonably abundant, it is quite hard to check this property in general. \textit{O. Debarre} [Compos. Math. 141, No. 6, 1445--1459 (2005; Zbl 1086.14038)] has shown that a general complete intersection of sufficiently ample line bundles in an abelian variety has ample cotangent bundle and one expects the same property for general complete intersections of high degree in \(\mathbb P^n\). In a different direction [\textit{M. Schneider}, Complex analysis and algebraic geometry, Proc. Conf., Göttingen 1985, Lect. Notes Math. 1194, 150--157 (1986; Zbl 0592.14016)] proved that if a projective surface \(S\) admits a smooth fibration such that the Kodaira-Spencer map is injective in every point, then \(\Omega_S\) is ample. \newline The aim of the paper under review is to give a weak generalization of Schneider's result to towers of smooth fibrations by curves of general type. More precisely let \(f_i:X^i \rightarrow X^{i-1}\) be a tower of fibrations where each \(X^i\) is a projective manifold of dimension \(i\) and each \(f_i\) is smooth with maximal variation, i.e. the Kodaira-Spencer map has maximal rank at a general point. Then \(\Omega_{X^n}\) is nef and big. Moreover \(\Omega_X\) is quasi-ample on some open set \(U\), i.e. for every smooth curve \(C \rightarrow X\) that meets \(U\), the restriction to \(C\) is ample. Generalizing the classical construction due to Kodaira, the author constructs examples of varieties satisfying the conditions of the main theorem. Using sheaves of one-forms with logarithmic poles along a divisor, she also proves a positivity result when the fibrations are not smooth.

Recommendations

Cites work

Cited in

(11)

This page was built for publication: Positivity of cotangent bundles

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q846229)