Equations of low-degree projective surfaces with three-divisible sets of cusps (Q1771973)

The authors consider algebraic surfaces \(Y\subset\mathbb{P}_3(\mathbb{C})\). A cusp (=singularity \(A_2\)) on \(Y\) is a singularity near which the surface is given in local (analytic) coordinates \(x\), \(y\) and \(z\), centered at the singularity, by an equation \(xy- z^3= 0.\) This is an isolated quotient singularity \(\mathbb{C}^2/\mathbb{Z}_3\). A set \(P_1,\dots, P_n\) of cusps on \(Y\) is called 3-divisible, if there is a cyclic global triple cover of \(Y\) branched, precisely over these cusps. Equivalently: If \(\pi: X\to Y\) is the minimal desingularization introducing two \((-2)\)-curves \(E_\nu'\), \(E^{\prime\prime}_\nu\) over each cusp, then there is a way to label these curves such that the divisor class of \[ \sum^n_{\nu=1} (2E_\nu+ E^{\prime\prime}_\nu) \] is divisible by 3 in NS\((X)\) [cf. \textit{S.-L. Tan}, Cusps on some algebraic surfaces and plane curves. Preprint, 1999, \texttt{http://www.math.ecnu.cn/\(\sim\)sitan/tan-cusp.pdf}] where applications of 3-divisible sets can be found. The aim of this note is to determine the equations for surfaces \(Y\subset\mathbb{P}_3\) of degrees \(\leq 6\) carrying a minimal, non-empty set of 3-divisible cusps.