The rational cohomology ring of the moduli space of abelian 3-folds. (Q1864713)

From MaRDI portal
scientific article
Language Label Description Also known as
English
The rational cohomology ring of the moduli space of abelian 3-folds.
scientific article

    Statements

    The rational cohomology ring of the moduli space of abelian 3-folds. (English)
    0 references
    0 references
    8 May 2003
    0 references
    Let \({\mathcal A}_g\) denote the moduli space of principally polarised abelian varieties of dimension \(g\), defined over the complex numbers. The rational cohomology ring of \({\mathcal A}_1\) is well known and is trivial except in degree \(0\), while that of \({\mathcal A}_2\) has been determined by \textit{J. Igusa} [Am. J. Math. 84, 175--200 (1962; Zbl 0133.33301)]. The object of the present paper is to determine the rational cohomology ring of the \(3\)-dimensional variety \({\mathcal A}_3\). The author also computes the rational cohomology rings of the Satake compactifications of \({\mathcal A}_2\) and \({\mathcal A}_3\). The method of proof is to analyse the spectral sequence corresponding to the extension \(1\to{\widehat T}_g\to\Gamma_g\to PSp_g({\mathbb Z})\to1\), where \(\Gamma_3\) is the mapping class group and \({\widehat T}_3\) is the extended Torelli group. This requires a knowledge of the cohomology of \({\widehat T}_3\) and of \(\Gamma_3\) (or equivalently of the moduli space \({\mathcal M}_3\) of curves of genus \(3\)). The cohomology of \({\mathcal M}_3\) has been computed by \textit{E. Looijenga} [in: Mapping class groups and moduli spaces of Riemann surfaces. Contemp. Math. 150, 205--228 (1993; Zbl 0814.14029)], while that of \({\widehat T}_3\) is computed in the present paper using the stratified Morse theory of \textit{M. Goresky} and \textit{R. MacPherson} [Stratified Morse theory. (Ergebnisse der Mathematik und ihrer Grenzgebiete (3), 14, Springer-Verlag, Berlin), (1988; Zbl 0639.14012)]. There is no obvious way of extending these arguments to \({\mathcal A}_4\), since the cohomology of \({\mathcal M}_4\) is not known and it does not seem tractable to compute that of \({\widehat T}_4\).
    0 references
    0 references
    0 references
    0 references
    0 references
    principally polarised abelian variety
    0 references
    moduli space
    0 references
    cohomology of moduli spaces
    0 references
    mapping class group
    0 references
    Torelli group
    0 references
    Morse theory
    0 references
    0 references
    0 references