The role of climatic factors in shaping mosquito population dynamics: the case of Culex pipiens in Northwestern Italy

Culex pipiens mosquito is a species widely spread across Europe and represents a competent vector for many human diseases such as West Nile virus (WNV), which has been circulating in many European countries during last years, causing hundreds of human cases. Rigorous surveillance of mosquito density and control programs based on the reduction of the mosquito population represent key components of disease containment and prevention. Therefore, in order to design appropriate control strategies it is crucial to investigate the influence of climatic factors on the dynamics of vector populations during a typical breeding season. To understand which are the main determinants of the high heterogeneity observed in Piedmont region (Northwestern Italy) in Culex pipiens abundance across different seasons, we developed a density dependent mathematical model that takes explicitly into account the mechanisms of diapause characterizing the overwintering of adult mosquitoes (driven by the daylight duration), and the role played by temperature, which strongly affects both developmental and death rates of different life stages of Culex pipiens. The model was calibrated by performing a Bayesian statistical analysis of weekly capture data gathered in our study site from 2000 to 2011. Our analysis suggests that the high heterogeneity characterizing the dynamics of mosquito density among different years is driven by differences in the temporal patterns of temperature and in the larval carrying capacity associated to different breeding seasons. Specifically, high temperatures during early spring may anticipate the onset of the breeding season, while higher temperatures during late spring are associated with longer seasons. On the opposite, high temperatures during the summer can reduce the Culex pipiens abundance by increasing the adult mortality during this period. Our results show that higher density of adult mosquitoes are associated to higher larval carrying capacities which result positively correlated with spring precipitations suggesting that more rain during spring can create more breeding sites. Finally, our simulations show that the initial number of adults does not affect the vector abundance of a specific season, suggesting that two consecutive years might be unrelated and that the inter-seasonal heterogeneity might not be influenced by the adults’ ability of overwintering.