Hierarchical equilibria of branching populations (Q1767538)

This paper is devoted to the study of spatial branching models with a hierarchical geographical structure: \(N\) islands (blocks of radius \(l\)) per archipelago (block of radius \(l+1\)) and \(N\) blocks of radius \(l\) per block of radius \(l+1\) for \(l\geq 2\). The migration process then is modelled by so called hierarchical random walks: at a certain rate depending on \(l\), an individual jumps to a randomly chosen island in distance \(l\). This ultrametric structure makes the model particularly suitable for a thorough analysis of equilibrium states and cluster formulation. It turns out that in the hierarchical mean field limit (with order of \(N\) individuals per island and \(N\to \infty\)) there is a separation of time scales in which the population densities in the blocks of different radii evolve. In the case of one-level branching, the natural time scale for a block with radius \(l\) is \(N^l\) [see \textit{D. A. Dawson} and \textit{A. Greven}, Electron. J. Probab. 1, Paper No. 14 (1996; Zbl 0890.60093)]. In the case of two-level branching, the results of this work show that the natural time scale is \(N^{l/2}\). On this time scale, the population density in a block of radius \(l\) performs as \(N\to \infty\) a process whose fluctuations are governed by the branching and whose drift is given by a flow of emigration to and immigration from the surrounding block. For a sequence of nested blocks, this leads to a hierarchy of branching equilibria whose structure is described clearly here. For the case of two-level branching, the authors prove that the population densities in nested blocks converge as \(N\to \infty\) to this hierarchy. In the hierarchical model, the migration process sustaining an equilibrium is usually at the border to recurrence for one-level branching, and at the border to weak transience for two-level branching. It is well known that Euclidean random walks are transient if and only if the dimension is bigger than two, and strongly transient if and only if the dimension is bigger than four. In this sense, hierarchical one-level branching equilibria (see \textit{D. A. Dawson} and \textit{A.Greven}, loc. cit.) correspond to a situation ``near dimension two'', and the hierarchical two-level branching equilibria studied in this paper correspond to a situation ``near dimension four''. Therefore, this work completes the picture of the equilibria of hierarchical branching populations. We remark that dimension four is of considerable interest because it serves as a critical dimension not only for the two-level branching systems studied in this work but also for a number of phenomena in statistical physics including the large scale fluctuations of ferromagnetic models at the critical temperature. The authors also give a description of the genealogy of the equilibrium population in the mean-field limit using a cascade of subordinators.