Subordination of trees and the Brownian map (Q1656540)

In the paper under review, the subordination of random rooted compact \(\mathbb{R}\)-trees is discussed. This concept is analogous to subordination of stochastic processes. Subordination for a deterministic rooted compact \(\mathbb{R}\)-tree \(\mathcal{T}\) works as follows. If \(g:\mathcal{T} \to \mathbb{R}_+\) is a nonnegative, nondecreasing function (nondecreasing with respect to the natural genealogical order on \(\mathcal{T}\)) with \(g(0)=0\), one may subordinate \(\mathcal{T}\) by \(g\) by gluing to a single point each connected component of \(\mathcal{T}\) on which \(g\) is constant. This gives a new rooted compact \(\mathbb{R}\)-tree \(\widetilde{\mathcal{T}}\). If \(a \in \mathcal{T}\), then the distance of the root and \(a\) in \(\widetilde{\mathcal{T}}\) is \(g(a)\). The resulting tree \(\widetilde{\mathcal{T}}\) is the subordinate tree of \(\mathcal{T}\) with respect to \(g\). The main interest in the paper is in the case where \(\mathcal{T}=\mathcal{T}_\zeta\) is the Brownian tree coded by a positive Brownian excursion under the Itō excursion measure. The subordination function is the past maximum function of a linear Brownian motion \((Z_a)_{a \in \mathcal{T}_\zeta}\) starting from \(0\) at the root \(\rho\). In this case, the subordinate tree is identified as a stable Lévy tree with index \(3/2\). This result is reminiscent of the classical result that the right-continuous inverse of the maximum process of a standard linear Brownian motion is a stable subordinator of index \(1/2\). As a more precise version, the author also shows that the maximum process of the Brownian snake is a time change of the height process coding the Lévy tree. This allows to recover, in a more precise form, a recent result by \textit{J. Miller} and \textit{S. Sheffield} [``An axiomatic characterization of the Brownian map'', Preprint, \url{arXiv:1506.03806}] identifying the metric net associated with the Brownian map.