Modular form representation for periods of hyperelliptic integrals (Q726535)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Modular form representation for periods of hyperelliptic integrals
scientific article

    Statements

    Modular form representation for periods of hyperelliptic integrals (English)
    0 references
    0 references
    11 July 2016
    0 references
    Given a hyperelliptic curve of genus \(g \geq 2\), say \(C: \; y^{2}=f(x)=\sum_{i=0}^{2g+2} \lambda_{i} x^{i}\), there is a natural homology basis \(\alpha_{1},\dots,\alpha_{g}, \beta_{1},\dots,\beta_{g}\) (which can be described geometrically in the \(x\)-plane) and there is a natural collection of meromorphic \(1\)-forms \(\tau_{j}=x^{j-1}dx/y\), where \(j=1,\dots, 2g\). The first \(g\) form a basis for the holomorphic differential forms. The second \(g\) meromorphic forms can be normalized to be of the form \(\rho_{j}=\sum_{k=j}^{2g+1-j}(k+1-j) \lambda_{k+j+1} x^{k} \; dx\), for \(j=1,\dots,g\). The \(g \times g\) period matrix \(2\omega\) (respectively, \(2\eta\)) is given by integration of the holomorphic forms \(\tau_{1},\dots,\tau_{g}\) (respectively, \(\rho_{1},\dots,\rho_{g}\)) on the loops \(\alpha_{j}\)'s. The matrix \(\chi=\eta \omega^{-1}/2\) turns out to be symmetric. The paper concerns with a representation of \(\chi\) as sum of two matrices, one of them \(\Lambda_{g}\) (having its coefficients in terms of the parameters \(\lambda_{j}\), for \(g=1,\dots,6\)) and the second one whose coefficients are defined in terms of some \(\theta\)-constants. For the general case, some information is provided which permits the authors to conjecture that the coefficients of \(\Lambda_{g}\) consists of integer multiples of the parameters \(\lambda_{j}\).
    0 references
    periods of second kind differentials
    0 references
    theta-constants
    0 references
    modular forms
    0 references

    Identifiers