This paper presents evidence and the existence of seasonality in current existing COVID-19 datasets for three
different countries namely Zimbabwe, South Africa, and Botswana. Therefore, we modified the SVIR model
through factoring in the seasonality effect by incorporating moving averages and signal processing techniques
to the disease transmission rate. The simulation results strongly established the existence of seasonality in
COVID-19 dynamics with a correlation of 0.746 between models with seasonality effect at 0.001 significance
level. Finally, the model was used to predict the magnitude and occurrence of the fourth wave.