One-dimensional families of Riemann surfaces of genus g with \(4\mathrm {g}+4\) automorphims (Q1671963): Difference between revisions

Suppose that for all \(g \geq 2\) there exists an equisymmetric uniparametric family of Riemann surfaces of genus \(g\) with \(ag + b\) automorphisms, where \(a, b\) are integers with \(a > 0\). Then, the pair \((a,b)\) is said to be admissible. Theorem 1 of this paper gives the main result, which is that \((4,4)\) is the admissible pair which gives generically the highest value of \(ag+b\). Theorem 2 is devoted to the description of a family corresponding to this upper bound, that is to say, for each \(g\), Theorem 2 describes a family of surfaces of genus \(g\), having \(4g+4\) automorphisms. This family is denoted by \(\mathcal{A}_g\). The authors prove that, if \(X \in \mathcal{A}_g\), then the group \(D_{g+1} \times C_2\) is a subgroup of \(\mathrm{Aut}(X)\), the quotient \(X/(D_{g+1} \times C_2) = S\) is the Riemann sphere, and the regular covering \(X \rightarrow S\) has four branched points, of orders \(2,2,2\) and \(g+1\). For some values of \(g\), there are more families satisfying these conditions. For instance, for \(g \equiv 3\) (mod 4), Theorem 4 gives another family of surfaces \(\mathcal{K}_g\), which admit an automorphism group isomorphic to the central product of \(D_4\) with \(C_{g+1}\), of order \(4g+4\). For the small values \(g = 3,5,7\), all existing families are shown. In the case \(g=3\), there are three families, \(\mathcal{A}_3, \mathcal{K}_3\) and a third family having \(\langle 4,4 \;|\;2,2\rangle\) as automorphism group. There are two families for \(g=5\), \(\mathcal{A}_5\) and another one with automorphism group \(\langle 4,6 \;|\;2,2\rangle\). Finally, for \(g=7\), there exist \(\mathcal{A}_7, \mathcal{K}_7\) and a third family with group \(\langle 4,8 \;|\;2,2\rangle\). The surfaces in \(\mathcal{A}_g\) are hyperelliptic while those in \(\mathcal{K}_g\) are non-hyperelliptic. The other surfaces appearing in genera \(3, 5\) and \(7\) are also non-hyperelliptic. All throughout the paper, the main tool is the uniformization of Riemann surfaces and their automorphism groups by means of Fuchsian groups. A final feature of the paper is the announcement of two results, which state that the surfaces in the families \(\mathcal{A}_g\) and \(\mathcal{K}_g\) have an anticonformal involution whose fixed point sets consist of three arcs, and hence those surfaces are real Riemann surfaces. Those results will be proved elsewhere, according to the authors.