The number of zeros of certain combinations of the Eisenstein series for \(\Gamma_0^+(2)\) (Q2660312)

From MaRDI portal
scientific article
Language Label Description Also known as
English
The number of zeros of certain combinations of the Eisenstein series for \(\Gamma_0^+(2)\)
scientific article

    Statements

    The number of zeros of certain combinations of the Eisenstein series for \(\Gamma_0^+(2)\) (English)
    0 references
    0 references
    0 references
    29 March 2021
    0 references
    Let \(\Gamma_0(2)^+\) be the modular group generated by the elements of \(\Gamma_0(2)\) and the Fricke involution \(W_2\). The set \(\mathcal{F}^+=\{z\in \mathbb C:|z|>1/\sqrt{2},-1/2<\Re(z)<0\}\cup\{z\in \mathbb C:|z|\ge 1/\sqrt{2},0\le\Re(z)\le 1/2\}\) is its canonical fundamental domain. The Eisenstein series of weight \(k\) for \(\Gamma_0(2)^+\) is defined by \(E_k^+(z)=(E_k(z)+2^{k/2}E_k(2z))/(1+2^{k/2})\) with the Eisenstein series \(E_k(z)\) of weight \(k\) for \(\mathrm{SL}_2(\mathbb Z)\). It is proved that the zeros of the Eisenstein series for \(\Gamma_0(2)^+\) lie on the arc boundary of \(\mathcal{F}^+\). Let us consider a combination of Eisenstein series \(\Delta_{k,\ell}^+=E_k^+E_\ell^+-E_{k+\ell}^+\) for positive integers \(k\ge\ell\ge 4\), which are not identically zero. In this article, the authors give lower bounds of the numbers of the zeros \(A_{k,\ell}^+\) and \(B_{k,\ell}^+\) of \(\Delta_{k,\ell}^+\) on the arc boundary \(C_1=\{z_1(t)=e^{it}/\sqrt{2}:\pi/4\le t\le\pi/2\}\) and on the vertical boundary \(C_2=\{z_2(t)=1/2+it: t\ge 1/2\}\) of \(\mathcal{F}^+\) respectively. The upper bound of the sum \(A_{k,\ell}^++B_{k,\ell}^+\) is given by \((k+\ell)/8-1\). The function \(F_{1,k,\ell}(t)=e^{i(k+\ell)t/2}\Delta_{k,\ell}^+(z_1(t))\) on \(C_1\) and the function \(F_{2,k,\ell}(t)=\Delta_{k,\ell}^+(z_2(t))\) on \(C_2\) are both real-valued continuous functions of \(t\). For each \(i=1,2\), by giving a finite sequence of points \(z_i(t_{i,j})\) on \(C_i\) such that the finite series \(\{t_{i,j}\}~(j=1,2,\cdots)\) of real numbers is strictly monotonic and \(F_{i,k,\ell}(z_i(t_{i,j}))F_{i,k,\ell}(z_i(t_{i,j+1}))<0\), they obtain a lower bound from the intermediate value theorem. To determine the sign of \(F_{i,k,\ell}(z_i(t_{i,j}))\), they use a new estimate of \(\Delta_{k,\ell}^+\) with sufficiently small error term at \(z_i(t_{i,j})\). The main results are as follows; Let \(n\in\mathbb Z^+\). If \(k\ge 2\ln(200/13)/\ln(1+\varepsilon (n))\ge 28\) and \(k\ge\ell\ge 18\) with \(\varepsilon (n)=\pi/(8n+6)\), then \(A_{k,\ell}^+\ge n+\eta\), where \(\eta=0,1,-1\), depending on the pair \((k,\ell) \mod 8\). Suppose \(k\ge \ell\ge L\) where \(L\) is large. Then \(B_{k,\ell}^+\ge \lfloor \ell/6\rfloor+\eta'\), where \(\eta'=0,-1\), depending on the pair \((k,\ell) \mod 6\). The precise statements are given in Theorems 1.1 and 1.2.
    0 references
    0 references
    weakly holomorphic modular form
    0 references

    Identifiers