Stability and instability of steady states for a branching random walk (Q2241506)

The authors study convergence to equilibrium for a system of branching random walks on the lattice \(\mathbb Z^d\). At time \(0\), there is one particle at each point of \(\mathbb Z^d\). The particles located at \(x\in \mathbb Z^d\) die with rate \(\mu(x)\) and split into two particles at rate \(\beta(x)\). The new-born particles jump independently and their increments are distributed according to a symmetric probability measure on \(\mathbb Z^d\). In the critical case when \(\beta(x) = \mu(x)= \mu\) does not depend on \(x\), such system of branching random walks either converges to a stationary equilibrium state, or locally dies out, as \(t\to\infty\). This has has been shown by the second author and \textit{J. Whitmeyer} [Math. Popul. Stud. 24, No. 3, 147--160 (2017; Zbl 1409.92204)]. The present paper studies the setting in which \(\mu(x) = \mu\) is constant, while \(\beta(x) = \mu + \sum_{i=1}^k \sigma_i \delta_{x_i}(x)\) is constant up to a perturbation concentrated at finitely many points \(x_1,\ldots, x_k \in \mathbb Z^d\), where \(\sigma_1,\ldots, \sigma_k>0\). It is shown that if \(d\geq 3\) and \(\sigma_1 + \ldots + \sigma_k\) is ``not too large'', the branching random field still converges to an equilibrium.