Hodge polynomials of the moduli spaces of rank 3 pairs (Q960041): Difference between revisions

From MaRDI portal
RedirectionBot (talk | contribs)
Changed an Item
Import240304020342 (talk | contribs)
Set profile property.
Property / MaRDI profile type
 
Property / MaRDI profile type: MaRDI publication profile / rank
 
Normal rank

Revision as of 01:44, 5 March 2024

scientific article
Language Label Description Also known as
English
Hodge polynomials of the moduli spaces of rank 3 pairs
scientific article

    Statements

    Hodge polynomials of the moduli spaces of rank 3 pairs (English)
    0 references
    0 references
    16 December 2008
    0 references
    Given a compact Riemann surface \(X\), a holomorphic triple of type \((n_1,n_2,d_1,d_2)\) is a triple \((E_1,E_2,\phi)\), where \(E_i\) is a holomorphic vector bundle of rank \(n_i\) and degree \(d_i\), and \(\phi\) is a holomorphic map \(\phi:E_2\to E_1\). There is a (poly,semi)stability condition for these objects, depending on a real parameter \(\sigma\), and which allows the construction of the moduli spaces \(\mathcal{N}_\sigma=\mathcal{N}_\sigma(n_1,n_2,d_1,d_2)\) of \(\sigma\)-semistable triples of the given type. As \(\sigma\) varies, the nature of the \(\sigma\)-semistability condition only changes for a discrete (finite if \(n_1\neq n_2\)) number of so-called critical values. \(\mathcal{N}_{\sigma_1}\) and \(\mathcal{N}_{\sigma_2}\) are, therefore, isomorphic if and only if there is no critical value between \(\sigma_1\) and \(\sigma_2\). Moreover, when a critical value \(\sigma_c\) is crossed, then \(\mathcal{N}_{\sigma_c-\epsilon}\) and \(\mathcal{N}_{\sigma_c+\epsilon}\) (where \(\epsilon\) is small enough so that \(\sigma_c\) is the only one critical value between \(\sigma_c-\epsilon\) and \(\sigma_c+\epsilon\)) are related by a transformation, which is restricted to certain subspaces of \(\mathcal{N}_{\sigma_c-\epsilon}\) and of \(\mathcal{N}_{\sigma_c+\epsilon}\), called flip loci [see \textit{S. B. Bradlow, O. García-Prada, P. B. Gothen}, Math. Ann. 328, No. 1-2, 299--351 (2004; Zbl 1041.32008)]. In this paper, the author computes the Hodge polynomials of \(\mathcal{N}_\sigma(3,1,d_1,d_2)\), using the theory of mixed Hodge structures [introduced by \textit{P. Deligne} in Actes Congr. Internat. Math. 1970, 1, 425--430 (1971; Zbl 0219.14006); Publ. Math., Inst. Hautes Étud. Sci. 40, 5--57 (1971; Zbl 0219.14007); and Publ. Math., Inst. Hautes Étud. Sci. 44, 5--77 (1974; Zbl 0237.14003)] and which can be defined in any complex algebraic variety, not necessarily smooth, compact or irreducible. For that, the author carries out a detailed study of the flip loci and of their Hodge polynomials, by making use, among other things, of the additivity of these polynomials with respect to finite disjoint unions of algebraic subvarieties. One consequence of this computation for the case of \(\mathcal{N}_\sigma\), where \(\sigma\) is the minimum critical value, is the recovery of the Hodge polynomial of the moduli space of semistable vector bundles of rank \(3\) and degree \(d\), when \(d\) is not a multiple of \(3\). This is, hence, done using different methods from the ones of \textit{R. Earl} and \textit{F. Kirwan} [Q. J. Math. 51, No.4, 465-483 (2000; Zbl 0971.14011)].
    0 references
    moduli space
    0 references
    complex curve
    0 references
    vector bundle
    0 references
    stable triple
    0 references
    Hodge polynomial
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references