Climate-driven modelling of the spread and control of West Nile virus and Usutu virus in Germany

Abstract

Mosquito-borne flaviviruses remain a significant public health and ecological concern, influenced by climate change, environmental variability, and the range expansion of entomologically relevant mosquito vectors. In recent years, West Nile virus (WNV) and Usutu virus (USUV) have become endemic in parts of Europe, including Germany, posing risks to birds, humans, and other vertebrate hosts. Despite increasing surveillance, important questions remain about the factors driving their spatial spread, seasonal outbreaks, and the success of intervention strategies. This PhD thesis investigates the transmission and spread of mosquito-borne flaviviruses through mechanistic modelling frameworks that integrate mosquito ecology, ecoepidemiological processes, and environmental variables. Initially, a process-based ordinary differential equation (ODE) model was developed to simulate the population dynamics of Culex pipiens s.s./Cx. torrentium, the primary vectors of WNV and USUV in Germany. The model incorporated temperature-, rainfall-, and photoperiod-dependent processes governing mosquito development, mortality, reproduction, and overwintering. Validation against mosquito surveillance data from 106 sampling sites demonstrated good agreement between simulated and observed mosquito phenology, with temperature as the primary driver of mosquito seasonal dynamics. Building on these insights, ODE and partial differential equation (PDE)-based ecoepidemiology models were developed to investigate the transmission dynamics and spatial spread of WNV and USUV. These models incorporated temperature-dependent transmission processes, spatial diffusion to represent local dispersal, and advection to capture directional movement associated with migratory birds. Furthermore, optimal control theory was applied to evaluate the timing, intensity, and cost-effectiveness of intervention strategies, including mosquito control, equid vaccination, and personal protective measures. Results obtained show that temperature is also a key variable influencing seasonal transmission patterns and regional differences in WNV and USUV infection risk. Transmission remains mainly localised, driven by environmental conditions and vector abundance, with eastern Germany and the Upper Rhine Valley identified as consistently high-risk areas for both viruses. Simulations that include migratory bird movements indicate that changes in mosquito feeding behaviour towards migratory hosts may significantly accelerate WNV spread, supporting the hypothesis that migratory birds might facilitate its long-distance dispersal. Conversely, USUV spread appears to be driven mainly by short-range processes involving highly vulnerable resident bird species, especially the Eurasian blackbird (Turdus merula), which is consistent with its relatively slower spatial spread during the early years of endemicity. Optimal control analyses suggest that mosquito control efforts started in early spring and intensified in early May may reduce the August peak in the infectious bird population. Furthermore, a combined approach of mosquito control methods, human PPE adoption, and equid vaccination could be the most effective strategy for both human and equid hosts. These results further indicate that targeting overwintering mosquito populations can lessen the intensity of interventions needed during the transmission season. To aid decision-making, a free and open-source web application has been developed, allowing users to explore the impact of different control strategies under various temperature projections, such as low-emission (SSP1-2.6), intermediate-emission (SSP2-4.5), and high-emission (SSP5-8.5) scenarios. Although WNV and USUV differ in their public health implications, the models highlight their shared ecological transmission pathway, driven by common vector species. This suggests that integrated control strategies can simultaneously reduce the impact of both viruses. The general findings of this thesis advance the understanding of the ecological and climatic factors influencing arbovirus transmission in Germany and provide a quantitative basis for developing climate-informed surveillance and control strategies.