Modelling climate-driven spatiotemporal transmission dynamics of Aedes-borne arboviruses

The distribution of Aedes spp. mosquitoes and the diseases they vector is heavily influenced by climate. A large body of work has previously characterised the role of temperature on various mosquito life-history traits. However, given these are largely derived from laboratory-studies, there is ongoing uncertainty surrounding whether these can be directly translated into field-settings. Here, a metapopulation model of Aedes-borne arbovirus transmission is presented that incorporates temperature-dependent parameters. To account for additional climate variables, existing maps of dengue force of infection are utilised to calibrate location-specific mosquito carrying capacities, enabling a background level of environmental suitability to be accounted for. The development process of the model is initially described in detail. Key model features are highlighted, such as stability criteria and spatial structure. A thorough sensitivity analysis of previous proposed thermal responses of mosquito life history traits is also conducted. The model is applied to both endemic and epidemic contexts, with particular focus on dengue and Zika in Colombia. Through these applications, it is demonstrated that the temperature dependency of the model can mechanistically explain the increased risk of Aedes-borne arbovirus outbreaks that are associated with El Nino-driven temperature patterns. A novel statistical approach is also developed to aid model fitting in highly stochastic systems, which has the potential to be applied in various contexts aside from infectious disease modelling.