New research offers roadmap for improved West Nile forecasting and prevention
February 11, 2026
A new technique for forecasting West Nile virus based on recent weather conditions is the first of its kind to successfully predict caseloads nationwide. This provides a critical foundation for the prevention of West Nile in years to come.
West Nile Virus is both the most common mosquito-borne illness in the United States, and the deadliest. Since the first cases were detected in New York in 1999, West Nile Virus has caused over 30,000 cases of severe illness and nearly 3,000 deaths in the U.S. But there is still no effective national forecast for the virus.
“West Nile has emerged as a major public health concern across much of the United States,” said Ryan Harp, a climate scientist with the University of Minnesota Climate Adaptation Partnership and the paper’s lead author. “Improving our ability to predict the prevalence of West Nile infection will allow for more targeted and effective surveillance, public outreach, and mosquito control.”
Minnesota reported over 120 cases of West Nile — and 10 deaths — in 2025, the most cases since the initial outbreak of West Nile in the state in 2003.
While most West Nile cases are asymptomatic, about 1 in 150 infections lead to a severe, neuroinvasive form of the disease. Ten percent of neuroinvasive disease cases are fatal, and many who survive have enduring disabilities.
In the U.S., West Nile virus is carried by Culex mosquitoes and amplified predominantly by songbirds. The virus is thus intrinsically coupled to environmental conditions, but the relationship with weather variables is complex. For example, precipitation can lead to more areas for Culex mosquitoes to breed, but if rains are too heavy, existing breeding grounds can be washed out.
The model developed by Harp and colleagues takes the novel approach of aggregating county-level data to the regional scale, allowing trends obscured by limited data to emerge. The team found drought and temperature to be the most important weather variables driving variation in West Nile cases. While the model successfully demonstrates that an effective nationwide forecast is possible using historical data, additional work is needed to make real-time forecasts at the local scale.
“This model identifies critical relationships between weather conditions and disease prevalence that reflect varied ecologies across the country and provide the foundation to build more predictive West Nile models in the years to come,” said Michael Johansson, the senior scientist on the project.
"A Regionally Determined Climate-Informed West Nile Virus Forecast Technique" was published January 30 in the journal GeoHealth.
This research was supported by the University of Minnesota College of Food, Agricultural, and Natural Resource Sciences (CFANS), the National Oceanic and Atmospheric Administration (NOAA), the Centers for Disease Control and Prevention (CDC); the University of Colorado Cooperative Institute for Research in Environmental Science (CIRES); the University Corporation for Atmospheric Research (UCAR); and the Network Science Institute and Bouvé College of Health Sciences at Northeastern University.
