Remarks on a paper by Maharaj and Wulftange (Q1946073)

Let \(f(X,Y)\) be an absolutely irreducible polynomial with coefficients in a finite field \(\mathbb{F}\) which has even degree over its prime field. If \(f\) is separable over \(\mathbb{F}[X]\) and has \(\deg_Yf>1\), it determines a tower of function fields \(\mathcal{F}=(F_0,F_1,F_2,\dots)\), where \(F_0=\mathbb{F}(x_0)\) and \(F_{i+1}=F_i(x_{i+1})\) is determined by the equation \(f(x_i,x_{i+1})=0\). The tower is called asymptotically optimal if it contains function fields of genus \(g(F_i)>1\) and \(\lim_{i\to\infty}N(F_i)/g(F_i)=\sqrt{\#\mathbb{F}}-1\), where \(N(F_i)\) is the number of rational places of \(F_i\). \textit{H. Maharaj} and \textit{J. Wulftange} found several asymptotically optimal towers by a numerical search [J. Pure Appl. Algebra 199, No. 1--3, 197--218 (2005; Zbl 1128.11054)]. In this paper it is shown that most of these towers are modular, thus providing further numerical evidence for Elkies' modularity conjecture. The modularity is established by finding explicit models of quotients of elliptic modular curves \(X_0(N)\) and/or \(X_1(N)\) by involutions and comparing with the defining equations \(f(X,Y)\) of the Maharaj and Wulftange towers.