Remote sensing of temperature-dependent mosquito and viral traits predicts field surveillance-based disease risk.

Mosquito-borne diseases contribute substantially to the global burden of disease, and are strongly influenced by environmental conditions. Ongoing and rapid environmental change necessitates improved understanding of the response of mosquito-borne diseases to environmental factors like temperature, and novel approaches to mapping and monitoring risk. Recent development of trait-based mechanistic models has improved understanding of the temperature dependence of transmission, but model predictions remain challenging to validate in the field. Using West Nile virus (WNV) as a case study, we illustrate the use of a novel remote sensing-based approach to mapping temperature-dependent mosquito and viral traits at high spatial resolution and across the diurnal cycle. We validate the approach using mosquito and WNV surveillance data controlling for other key factors in the ecology of WNV, finding strong agreement between temperature-dependent traits and field-based metrics of risk. Moreover, we find that WNV infection rate in mosquitos exhibits a unimodal relationship with temperature, peaking at ~24.6-25.2°C, in the middle of the 95% credible interval of optimal temperature for transmission of WNV predicted by trait-based mechanistic models. This study represents one of the highest resolution validations of trait-based model predictions, and illustrates the utility of a novel remote sensing approach to predicting mosquito-borne disease risk.