Abstract
BACKGROUND: Mosquitoes are vectors of diseases globally, making development of models that better explain mosquito abundance imperative. Mosquito population dynamics are particularly sensitive to local weather conditions, and mosquito-borne disease outbreaks can be spatially concentrated. There is a need for improved modeling studies to address whether spatial variation in disease outbreaks is driven by spatial variation in weather conditions, especially in dry and hot environments. In the present study, we built a climate-driven model of mosquito population dynamics and compared whether predictions of mosquito abundance at the county scale were improved by accounting for subcounty weather variation. METHODS: Using a 5-year time series of weekly Culex quinquefasciatus abundance data collected for each zip code in Maricopa County, USA, we assessed how local variation in weather could explain and predict mosquito population dynamics. We built a mechanistic model of mosquito population dynamics influenced by daily maximum temperature and 30-day accumulated precipitation. We grouped zip codes on the basis of similar patterns of temperature and precipitation using functional clustering. We compared two approaches: one using county-level average weather and another using data from the identified weather clusters. We used Markov chain Monte Carlo simulations to fit the mechanistic model using averaged weather data in each cluster, then compared the model fit with observed data between the county-level model and the model on the basis of weather-based clusters. RESULTS: Simple, weather-forced modeling accurately estimated detailed Cx. quinquefasciatus abundance trajectories throughout the 5-year period. Modeling mosquito abundances in the subcounty spatial clusters demonstrated that the same effects of temperature and precipitation on population growth rates could explain small-scale changes in mosquito abundances. However, when we aggregated the subcounty model fits to the county-scale, the resulting fits were more precise but sometimes overly confident, leading to lower overall accuracy and predictive performance. CONCLUSIONS: Our study demonstrated the importance of collecting fine-scale mosquito abundance data to improve our understanding and the predictability of mosquito population dynamics. The strong performance of both the cluster-based and county-level models illustrated the value of spatially sensitive modeling in this application. We anticipate that such modeling efforts will aid in using weather forecasts to predict increases in mosquito populations, thereby aiding in efforts to control the spread of infectious disease.