Kernels of \(L\)-functions of cusp forms (Q2267761)

Let \(f\) be a cusp form of weight \(k \in \mathbb{N}\) for the full modular group \(\Gamma = \mathrm{SL}_2(\mathbb{Z})\). In particular, \(f\) is a holomorphic function on the upper half-plane \(\mathbb{H} = \{z \in \mathbb{C}; \, \text{Im}(z)>0\}\), satisfies the transformation condition associated to the group \(\Gamma\) and weight \(k\), and admits the expansion \[ f(z) = \sum_{n=1}^\infty a_f(n) \, e^{2\pi inz} \qquad (z \in \mathbb{H}). \] The space of such cusp forms is denoted by \(S_k(\Gamma)\). The to \(f\) associated \(L\)-function is \[ L(f,s) := \sum_{n=1}^\infty a_f(n) \, n^{-s} \] for all \(s\in \mathbb{C}\) with \(\text{Re}(s) \gg 0\) large enough. The completed \(L\)-function is given by \[ L^\star(f,s) := (2\pi)^{-s} \, \Gamma(s) \, L(f,s) = \int_0^\infty f(iy)\, y^{s-1} \,dy. \] It is analytic for all \(s \in \mathbb{C}\). For integers \(0 \leq n \leq k-2\) we define the \(n\)-th period of \(f\) as \[ r_n(f) := L^\star(f,n+1). \] Recall the Petersson inner product \[ \langle f,g \rangle := \int_{\Gamma \backslash \mathbb{H}} f(z) \, \overline{g(z)} \, y^{k-2} \,dz \] for \(f,g \in S_k(\Gamma)\). It is known that there exist a kernel function \(R_n \in S_k(\Gamma)\) such that \[ \langle f, R_n \rangle = r_n(f) \] holds for all \(f \in S_k(\Gamma)\). Also, an explicit description of \(\langle R_m, R_n \rangle\) is known [\textit{W.\ Kohnen} and \textit{D.\ Zagier}, Modular forms, Symp.\ Durham/Engl.\ 1983, 197--249 (1984; Zbl 0618.10019)]. The first main result of the authors is the extension of the known formula for \(\langle R_m, R_n \rangle\) to values of \(n, m\) with ar non-integer. To make it precise, the authors show: For every \(s \in \mathbb{C}\) there exists a kernel function \(\mathcal{D}_k(\cdot,s) \in S_k(\Gamma)\) such that \[ \langle \mathcal{D}_k(\cdot,s), f \rangle = L^\star(\overline{f},s) \] holds for all \(f \in S_k(\Gamma)\). Let \(T_l\) denote the \(l\)-th Hecke operator acting on the space \(S_k(\Gamma)\). The authors give an explicit representation of \[ \langle T_l \mathcal{D}_k(\cdot,s), \mathcal{D}_k(\cdot,\overline{w}) \rangle \] with \(s,w \in \mathbb{C}\), in terms of a Petersson inner product of Poincaré series and non-holomorphic Eisenstein series. They also have a specialized version where \(s\) and \(w\) are integers of opposite parity. Their last main result is the discussion of the function \[ \mathcal{C}_k(z,s):= \sum_{\gamma \in \Gamma} \frac{1}{(\gamma z)^s \, j(\gamma,z)^k} \qquad (z \in \mathbb{H}, \; 1 < \text{Re}(s) < k-1) \] which was introduced by Kohnen and Zagier [loc. cit.] on page 203 for integer \(s\) between \(2\) and \(k-2\). Here \(\gamma z\) and \(j(\gamma,z)\) are defined as \[ {a \; b \choose c \; d} z := \frac{az+b}{cz+d} \qquad \text{and} \qquad j\left({a \; b \choose c \; d} ,z\right) = cz+d \] for all \({a \; b \choose c \; d} \in \Gamma\) and \(z \in \mathbb{H}\).