Abstract:The present paper is devoted to the study of the long term dynamics of diffusion processes modelling a single species that experiences both demographic and environmental stochasticity. In our setting, the long term dynamics of the diffusion process in the absence of demographic stochasticity is determined by the sign of $\Lambda_0$, the external Lyapunov exponent, as follows: $\Lambda_0<0$ implies (asymptotic) extinction and $\Lambda_0>0$ implies convergence to a unique positive stationary distribution $\mu_0$. If the system is of size $\frac{1}{\epsilon^{2}}$ for small $\epsilon>0$ (the intensity of demographic stochasticity), demographic effects will make the extinction time finite almost surely. This suggests that to understand the dynamics one should analyze the quasi-stationary distribution (QSD) $\mu_\epsilon$ of the system. The existence and uniqueness of the QSD is well-known under mild assumptions. We look at what happens when the population size is sent to infinity, i.e., when $\epsilon\to 0$. We show that the external Lyapunov exponent still plays a key role: 1) If $\Lambda_0<0$, then $\mu_\epsilon\to \delta_0$, the mean extinction time is of order $|\ln \epsilon|$ and the extinction rate associated with the QSD $\mu_{\epsilon}$ has a lower bound of order $\frac{1}{|\ln\epsilon|}$; 2) If $\Lambda_0>0$, then $\mu_\epsilon\to \mu_0$, the mean extinction time is polynomial in $\frac{1}{\epsilon^{2}}$ and the extinction rate is polynomial in $\epsilon^{2}$. Furthermore, when $\Lambda_0>0$ we are able to show that the system exhibits multiscale dynamics: at first the process quickly approaches the QSD $\mu_\epsilon$ and then, after spending a polynomially long time there, it relaxes to the extinction state. We give sharp asymptotics in $\epsilon$ for the time spent close to $\mu_\epsilon$.

Large-population asymptotics for the maximum of diffusive particles with mean-field interaction in the noises

On the mean-field limit of diffusive games through the master equation: extreme value analysis

Large systems of diffusions interacting through their ranks

From the master equation to mean field game limit theory: Large deviations and concentration of measure

Large Deviations for Small Noise Diffusions Over Long Time

Small mass limit for stochastic interacting particle systems with Lévy noise and linear alignment force

Convergence to the asymptotic large deviation limit

Population dynamics under demographic and environmental stochasticity

Mean field limits of particle-based stochastic reaction-drift-diffusion models

Asymptotic behaviors of multiscale multivalued stochastic systems with small noises

Long-Time Asymptotics of Diffusion in A Random Medium

Long run convergence of discrete-time interacting particle systems of the McKean-Vlasov type

Large deviations for the mean-field limit of Hawkes processes

Large deviations of mean-field interacting particle systems in a fast varying environment

Asymptotics of the Invariant Measure in Mean Field Models with Jumps

The crossover from the Macroscopic Fluctuation Theory to the Kardar-Parisi-Zhang equation controls the large deviations beyond Einstein's diffusion

Superdiffusive central limit theorem for a Brownian particle in a critically-correlated incompressible random drift

Anomalous Fluctuations of Extremes in Many-Particle Diffusion

McKean-Vlasov PDE with Irregular Drift and Applications to Large Deviations for Conservative SPDEs

A Mean Field Game approach to Large Deviations for empirical measures under common noise