Counter-example to global Torelli problem for irreducible symplectic manifolds (Q1865612): Difference between revisions
From MaRDI portal
Changed an Item |
Normalize DOI. |
||
Property / DOI | |||
Property / DOI: 10.1007/s00208-002-0344-2 / rank | |||
Property / DOI | |||
Property / DOI: 10.1007/S00208-002-0344-2 / rank | |||
Normal rank |
Latest revision as of 10:53, 16 December 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Counter-example to global Torelli problem for irreducible symplectic manifolds |
scientific article |
Statements
Counter-example to global Torelli problem for irreducible symplectic manifolds (English)
0 references
27 March 2003
0 references
An irreducible symplectic manifold is a simply connected compact Kähler manifold with everywhere nondegenerate holomorphic 2-form. An irreducible symplectic manifold \(X\) is of even-complex dimension and has a canonical symmetric form \(q_X\) on \(H^2(X, Z)\). O'Grady and Huybrechts independently proved that if irreducible symplectic manifolds \(X\) and \(Y\) are bimeromorphically equivalent, then there is a natural Hodge isometry between \((H^2(X, Z), q_X)\) and \((H^2(Y, Z), q_Y)\). The converse is the so-called bimeromorphic Torelli Problem (BTP). The paper under review aims mainly to construct a counter-example to BTP. Note that Debarre constructs two bimeromorphically equivalent irreducible symplectic manifolds which are not isomorphic. Let \(T\) be a complex torus of dimension 2 such that its dual torus \(T^*\) is not isomorphic to \(T\) and the Neron-Severi group \(NS(T)\) is trivial. Let \(K^2(T)\) be the fiber over \(0\in T\) of the composite maps of a Hilbert-Chow map \(h\text{Hilb}^3 \to \text{Sym}^3T\) and a holomorphic map \(\alpha: \text{Sym}^3T \to T\) sending \([p_1, p_2, p_3]\) to \(p_1+p_2+p_3\). Beauville showed that both, \(K^2(T)\) and \(K^2(T^*)\) are irreducible symplectic manifolds. The identification \(H^1(T, Z) \cong H_1(T^*, Z)\) gives a canonical isomorphism \(\alpha_T: H^2(T^*, Z) = \wedge^2 H^1(T^*, Z) \cong H^2(T, Z)\), and Shioda showed that \(\alpha_T\) is a Hodge isometry. The upshot is to show that there are no bimeromorphic maps from \(K^2(T)\) to \(K^2(T^*)\). Assume the contrary: \(f: K^2(T) \to K^2(T^*)\) bimeromorphic. Then \(f^*: H^2(K^2(T^*), Z) \cong H^2(K^2(T), Z)\), \(\text{Pic}(K^2(T^*)) \cong \text{Pic}(K^2(T))\) with \(f^*(\delta^*) = \delta\) since they are Kähler, and their exceptional divisors \(E, E^*\) and resolutions \(F, F^*\) are bimeromorphic. Therefore their Albanese varieties \(\text{Alb}(F) = T\) and \(\text{Alb}(F^*) = T^*\) are isomorphic which contradicts the choice of \(T\). By the same method, the author constructs a counter-example to the polarized Torelli problem. The paper ends with an interesting question on whether there is an equivalence between the bounded derived categories of two irreducible symplectic manifolds with Hodge isometry.
0 references
irreducible symplectic manifold
0 references
bimeromorphic Torelli Problem
0 references
Hodge Isometry
0 references