The Chow group of the moduli space of marked cubic surfaces (Q2505610)

From MaRDI portal
scientific article
Language Label Description Also known as
English
The Chow group of the moduli space of marked cubic surfaces
scientific article

    Statements

    The Chow group of the moduli space of marked cubic surfaces (English)
    0 references
    0 references
    0 references
    27 September 2006
    0 references
    \textit{I. Naruki} [Proc. Lond. Math. Soc. (3) 45, 1--30 (1982; Zbl 0508.14005)] constructed a desingularisation \({\mathcal C}\) of the moduli space \({\mathcal M}\) of marked cubic surfaces. The variety \({\mathcal C}\) is referred to as the cross-ratio variety because it can be described in terms of certain cross-ratios studied by Cayley, but the description of it given by Naruki is as a particular birational modification of the toric variety associated with the Weyl chamber decomposition of the root system \(D_4\). It carries an action of the Weyl group \(W(E_6)\). This paper uses that description to compute the Chow groups and Chern classes of \({\mathcal C}\). The authors use this to write down Riemann-Roch for \({\mathcal C}\) explicitly. They obtain information about the pullback (called \(nH\)) to \({\mathcal C}\) of \({\mathcal O}(n)\) of the projective embedding of \({\mathcal M}\) in \(\mathbb P^9\) discovered by Coble (who regarded \({\mathcal M}\) as parametrising sets of six points in \(\mathbb P^2\)), and rediscovered by Allcock and Freitag using modular forms. It turns out that \(\chi(nH)\) coincides with the Hilbert function of \(\mathbb{C}[Y_0,\dots,Y_9]/I\), where \(I\) is the homogeneous ideal of \({\mathcal M}\) in \(\mathbb P^9\). This suggests that Sym\(^n H^0(H)\) surjects onto \(H^0(nH)\) and \(H^i(nH)\) vanishes for \(i,n>0\). The structure of the paper is well described in the introduction: ``We determine the Chow groups of \({\mathcal C}\) and the spaces of \(W(E_6)\)-invariant cycles in section 2. In section 3 we consider the exceptional divisors in \({\mathcal C}\) over the cusps in \({\mathcal M}\) , which we call cusp divisors (these are actually very simple varieties, being the product of three \(\mathbb P^1\)s. We study the 36 boundary divisors, which parametrize nodal cubic surfaces, in section 4. The image of such a divisor in \({\mathcal M}\) is isomorphic to the Segre cubic in \(\mathbb P^4\). We use the information obtained on these divisors to compute the Chern classes of \({\mathcal C}\) in section 5. In section 6.3 we consider the 45 tritangent divisors, these parametrize cubics with an Eckardt point (a point on the cubic surface through which 3 lines pass). All these divisors correspond, in a \(W(E_6)\)-equivariant way, to points in a finite projective geometry.''
    0 references
    0 references
    0 references
    0 references