Random construction of Riemann surfaces (Q2494199)

The goal of this paper is to try to understand what a ``typical'' Riemann surface looks like. This is not easy and the authors take a novel standpoint. They use a construction of Riemann surfaces based on an idea of P. Buser. According to this one can associate with any 3-regular graph with an orientation (a combinatorial concept) a Riemann surface put together from ideal hyperbolic triangles. They interpret Belyi's theorem to say that the resulting Riemann surfaces are dense in the set of all compact Riemann surfaces. Next B. Bollobás has defined a probability measure on the set of all 3-regular graphs with orientation and with \(n\) vertices for a fixed \(n\). Buser's construction consists of two steps, the construction of a Riemann surface of finite area and then of its compactification. The first main theorem says, roughly, that as \(n\rightarrow \infty\) for every \(L\) the probability that all the cusps have a horocycle of length \(\geq L\) tends to \(1\). This is the basic result of this paper and it is used to derive a number of further results, for example estimates of the smallest eigenvalue of the Laplace-Beltrami operator, of Cheeger's constant, of the shortest geodesic or of the diameter. Some of these results were announced in [Electron. Res. Announc. Am. Math. Soc. 5, No. 11, 76--81 (1999; Zbl 0948.58003)].