Multistability and chaos in SEIRS epidemic model with a periodic time-dependent transmission rate

Abstract

In this work, we study the dynamics of a SEIRS epidemic model with a periodic time-dependent transmission rate. Emphasizing the influence of the seasonality frequency on the system dynamics, we analyze the largest Lyapunov exponent along parameter planes finding large chaotic regions. Furthermore, in some ranges there are shrimp-like periodic strutures. We highlight the system multistability, identifying the coexistence of periodic orbits for the same parameter values, with the infections maximum distinguishing by up one order of magnitude, depending only on the initial conditions. In this case, the basins of attraction has self-similarity. Parametric configurations, for which both periodic and non-periodic orbits occur, cover $13.20\%$ of the evaluated range. We also identified the coexistence of periodic and chaotic attractors with different maxima of infectious cases, where the periodic scenario peak reaching approximately $50\%$ higher than the chaotic one.