×

Geometry of the moduli of higher spin curves. (English) Zbl 1094.14504

Summary: This article treats various aspects of the geometry of the moduli \({\mathcal S}^{1/r}\) of \(r\)-spin curves and its compactification \(\overline{\mathcal S}^{1/r}\). Generalized spin curves, or \(r\)-spin curves, are a natural generalization of 2-spin curves (algebraic curves with a theta-characteristic), and have been of interest lately because of the similarities between the intersection theory of these moduli spaces and that of the moduli of stable maps. In particular, these spaces are the subject of a remarkable conjecture of E. Witten relating their intersection theory to the Gelfand-Dikii (KdVr) hierarchy. There is also a \(W\)-algebra conjecture for these spaces generalizing the Virasoro conjecture of quantum cohomology.
For any line bundle \({\mathcal K}\) on the universal curve over the stack of stable curves, there is a smooth stack \({\mathcal G}_{g,n}^{1/r}({\mathcal K})\) of triples \((X, {\mathcal L}, b)\) of a smooth curve \(X\), a line bundle \({\mathcal L}\) on \(X\), and an isomorphism \(b: {\mathcal L}^{\otimes r}\to {\mathcal K}\). In the special case that \({\mathcal K}=\omega\) is the relative dualizing sheaf, then \({\mathcal G}_{g,n}^{1/r}({\mathcal K})\) is the stack \({\mathcal G}_{g,n}^{1/r}\) of \(r\)-spin curves.
We construct a smooth compactification \(\overline{\mathcal G}_{g,n}^{1/r}({\mathcal K})\) of the stack \({\mathcal G}_{g,n}^{1/r}({\mathcal K})\), describe the geometric meaning of its points, and prove that its coarse moduli is projective.
We also prove that when \(r\) is odd and \(g>1\), the compactified stack of spin curves \(\overline{\mathcal G}^{1/r}\) and its coarse moduli space \(\overline{\mathcal S}^{1/r}\) are irreducible, and when \(r\) is even and \(g>1\), \(\overline{\mathcal S}^{1/r}\) is the disjoint union of two irreducible components. We give similar results for \(n\)-pointed spin curves, as required for Witten’s conjecture, and also generalize to the \(n\)-pointed case the classical fact that when \(g=1\), \(\overline{\mathcal S}^{1/r}\) is the disjoint union of \(d(r)\) components, where \(d(r)\) is the number of positive divisors of \(r\). These irreducibility properties are important in the study of the Picard group, and also in the study of the cohomological field theory related to Witten’s conjecture.

MSC:

14H10 Families, moduli of curves (algebraic)
PDFBibTeX XMLCite
Full Text: DOI arXiv

References:

[1] DOI: 10.1016/0370-2693(94)90721-8 · doi:10.1016/0370-2693(94)90721-8
[2] Cornalba M., World Scientific pp 560– (1989)
[3] DOI: 10.1007/BF02684599 · Zbl 0181.48803 · doi:10.1007/BF02684599
[4] DOI: 10.1007/BF01446303 · Zbl 0847.14018 · doi:10.1007/BF01446303
[5] Hain R. M., Cambridge pp 97– (1995)
[6] DOI: 10.1023/A:1000209527158 · Zbl 0912.14010 · doi:10.1023/A:1000209527158
[7] DOI: 10.2307/2006977 · Zbl 0549.57006 · doi:10.2307/2006977
[8] DOI: 10.2307/2951828 · Zbl 0881.14018 · doi:10.2307/2951828
[9] Kollár J., J. Differential Geom. 32 (1) pp 235– (1990)
[10] DOI: 10.1007/BF02099526 · Zbl 0756.35081 · doi:10.1007/BF02099526
[11] J. Algebraic Geom. 3 (2) pp 283– (1994)
[12] DOI: 10.1016/0040-9383(92)90008-6 · Zbl 0772.57025 · doi:10.1016/0040-9383(92)90008-6
[13] Pikaart M., MA pp 483– (1995)
[14] DOI: 10.1090/S0002-9939-1978-0494115-8 · doi:10.1090/S0002-9939-1978-0494115-8
[15] DOI: 10.1007/BF02698887 · Zbl 0891.14005 · doi:10.1007/BF02698887
[16] DOI: 10.1007/BF01475756 · Zbl 0502.32017 · doi:10.1007/BF01475756
[17] DOI: 10.1007/BF01388892 · Zbl 0694.14001 · doi:10.1007/BF01388892
[18] DOI: 10.1016/0550-3213(92)90235-4 · doi:10.1016/0550-3213(92)90235-4
[19] TX pp 235– (1993)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.