Closed geodesics and bounded gaps (Q2412545)

Motivated by \textit{Y. Zhang}'s theorem [Ann. Math. (2) 179, No. 3, 1121--1174 (2014; Zbl 1290.11128)] on bounded gaps between primes, the author looks at the analogues for other types of zeta function. For a strictly increasing sequence \(x_1<x_2<\ldots\) of positive real numbers one defines the ``limit-gap'' (LG) to be the liminf of the differences \(x_{n+1}-x_n\). If the zeta function has a product representation \(\prod(1-x_n^{-s})^{-1}\) with distinct \(x_n\) then one can ask whether the limit-gap of the \(x_n\) is finite or not. For example, the Ihara zeta function of a finite metric graph has a product representation \[ Z(s)=\prod_c(1-N(c)^{-s})^{-1}, \tag{*} \] in which \(c\) runs over all closed geodesics, and \(N(c)=e^{l(c)}\), where \(l(c)\) is the length of \(c\). In this case it is shown that the limit-gap is zero unless all edges are in rational proportion, in which case the limit-gap is infinite. (However it is not clear to the reviewer whether the lengths \(l(c)\) must be distinct.) The author next considers the Ruelle zeta function, in which the product (*) runs over primitive closed geodesics of a connected Riemann surface equipped with a hyperbolic metric giving it finite volume. In this case, the prime geodesic theorem shows that the limit-gap must be zero, unless there are repeated lengths. However for a dense subset of the Teichmüller space one has no repeated lengths. In the case of PSL\(_2(\mathbb{Z})\), it is shown that the limit-gap is 1; and for congruence subgroups it is always finite.