Abstract
Introduction/Background : The invasive mosquito Aedes albopictus presents a significant public health challenge due to its potential to spread arboviruses. Understanding and predicting its abundance and seasonality is crucial for effective vector control and public health strategies. Over the past 20 years, various predictive models, both correlative and mechanistic, have been developed. However, a comprehensive comparison of these models to assess their strengths and limitations has been lacking, which is essential for improving predictive accuracy and developing effective intervention strategies.Material and Method : We reviewed and tested eight different modelling approaches to estimate the abundance and seasonality of Ae. albopictus. These included both correlative models, which rely on statistical relationships between observed mosquito populations and environmental variables, and mechanistic models, which simulate the biological processes governing mosquito dynamics. The evaluation criteria focused on the beginning, peak, and end of the seasonal activity of the species. Additionally, the feasibility of integrating multiple models into an ensemble approach to enhance prediction capabilities was evaluated.Preliminary Observation : Initial observations suggest that each modelling approach has distinct strengths and limitations. Correlative models appear to perform well in data-rich environments, while mechanistic models seem to offer valuable insights into the biological processes of mosquito dynamics. The potential of ensemble models to combine predictions from multiple individual models indicates possible improvements in accuracy and reliability, especially in diverse environmental settings.Discussion and Conclusions : Advancements in modelling approaches for Ae. albopictus dynamics represent a significant step forward in vector control and public health strategy development. By comparing and integrating different models, we aimed to improve predictive capabilities, which are crucial for mitigating the growing threat of vector-borne diseases. Future research should focus on refining these models, expanding data collection efforts, and exploring the integration of additional environmental and biological variables to further enhance prediction accuracy and utility in public health applications.