Pikalert: NSF-NCAR's Road Weather Hazard Prediction System

According to the USDOT FHWA Road Weather Management program, more than 1 million weather-related accidents occur every year, with nearly 5,000 deaths and over 400,000 injuries. Accurately diagnosing and forecasting roadway conditions and hazards during winter weather conditions will help state DOTs decide when and where to allocate their plow and road treatment resources. This can promote safety and mobility during winter weather events and ultimately reduce the number of weather-related accidents and deaths.

Meeting the needs mentioned above, NCAR-NSF has developed Pikalert, an advanced winter roadway maintenance and decision support system (MDSS). Pikalert combines a cutting-edge forecasting system (DICast) and a road condition treatment module with a wide array of real-time inputs, including surface observations, radar data, vehicle data, and camera imagery. Leveraging this information, Pikalert produces highly tuned forecasts and hazard assessments out to 72 hours for each road segment and RWIS station within a state’s roadway network. Relevant hazards and alerts include but are not limited to pavement conditions, precipitation, visibility, and wind speed. Pikalert is currently installed and configured for real-time operations for 5 state departments of transportation; Alaska, Colorado, Iowa, Oregon, and Wyoming. During this presentation, I will provide an overview of the current Pikalert system and demonstrate the real-time web display, showcasing many of Pikalert’s primary features. Furthermore, this presentation will highlight recent enhancements to Pikalert, including improved alerting thresholds, various display updates, and the utilization of image recognition for tactical road condition assessment. The presentation will conclude with an outline of ongoing research efforts and future enhancements.