Well-Posedness for the Rosenzweig-MacArthur Model with Internal Stochasticity

In this work, we propose a stochastic version of the Rosenzweig-MacArthur model solely driven by internal demographic noise, extending classical Lotka-Volterra-type systems focused on external noise. We give a criterion for the existence and uniqueness of autonomous stochastic differential equations (SDEs) on an open submanifold of $\mathbb{R}^{n}$, and the framework allows for a wider choice of Lyapunov functions. In the meantime, the invariance of open submanifolds, which is a biologically feasible result and has been implicitly incorporated into many biological and ecological models, facilitates the application of analytic tools typically suited to $\mathbb{R}^{d}$ and indicates the persistence of predator and prey populations, thus providing a criterion for determining whether a population will become extinct. We apply the well-posedness criterion to our stochastic Rosenzweig-MacArthur model and show the existence and uniqueness of solutions. Furthermore, the asymptotic estimates of solutions are obtained, indicating the at most exponential growth of the population with internal stochasticity. Some numerical experiments are performed, which illustrate the discrepancy between the deterministic and stochastic models. Overall, this work demonstrates the broad applicability of our results to ecological models with constrained dynamics, offering a foundation for analyzing extinction, persistence, and well-posedness in systems where internal randomness dominates. This paper not only promotes the development of stochastic modeling and stochastic differential equations in theoretical ecology but also proposes a rigorous mathematical methodology for studying the predator-prey system with internal stochasticity.