The validity of the continuum limit in the gravitational \(N\)-body problem (Q1766261)

Numerical simulations of self-consistent \(N\)-body problems are desirable since they are often the most accurate numerical representation of real physical systems. Here two variants of gravitational \(N\)-body problem are investigated: 1. A frozen \(N\)-body problem, i.e. a system of immovable \(N\) bodies that creates a discrete field; 2. A full self-consistent \(N\)-body problem, i.e. \(N\) bodies can move and interact with each other. The author concentrates on the first formulation and studies the motion of test particles in the frozen \(N\)-body environment. The experimental setup is straightforward. The frozen \(N\)-body problem is associated with a discrete density, which generates a discrete potential. It is feasible to approximate this discrete density with a smooth continuous density, which then generates a smooth continuous potential. The main question in the paper: if one compares a discrete frozen \(N\)-body problem with its corresponding smooth environment, what is the minimum number of particles \(N\) providing that two systems are characterized by approximately the same properties in point-wise and statistical level? It is noted that, in spite of many numerical investigations, there is no an answer to this question until now. Some new numerical simulations are provided in the article, and the results obtained are discussed.