On the cohomological dimension of the moduli space of Riemann surfaces (Q2012193)

Let \(S_{g, n}\) be a compact Riemann surface of genus \(g\geq 2\) with \(n\geq 0\) points removed. Denote by \(T(g, n)\), \(M(g, n)\) and \(\mathrm{Mod}(g, n)\) the Teichmüller space, Riemann moduli space and Teichmüller modular group (or mapping class group) of \(S_{g, n}\), respectively. It is well known that \(\mathrm{Mod}(g, n)\) acts discontinuously on the complex manifold \(T(g, n)\) so that \(M(g, n)=T(g, n)/\mathrm{Mod}(g, n)\) is a complex analytic orbifold. So it makes sense to speak of Dolbeault cohomology with coefficients in a holomorphic vector bundle and of Dolbeault cohomological dimension \(\text{coh-dim}_{\text{Dol}}(M(g, n))\). Looijenga conjectured that \[ \text{coh-dim}_{\text{Dol}}(M(g, n))=g-2+\zeta_n, \] with \(\zeta_0=0\) and \(\zeta_n=1\) for \(n>0\). The main result of the paper is the following estimate \[ \text{coh-dim}_{\text{Dol}}(M(g, n))\leq (g-2+\zeta_n)+g. \]