Linking Ensemble Data to Users: AccuWeather’s Snowfall Probability Tool Provides Unique Insight of Forecast Confidence

From the first television weather report to the display of weather information on connected devices today, the communication of weather forecasts to end users has constantly evolved. Users rely on accurate and timely weather forecasts - often, multiple times daily - to make decisions that impact their lives. It is therefore imperative that we as an enterprise provide as much information as possible in a succinct, timely and understandable manner so people can make the best decisions and take necessary actions. With the advent of ensemble prediction systems, new ways to communicate weather forecasts in a probabilistic manner are emerging. This requires new and imaginative ways to display forecast confidence information to a user which is significantly more challenging than a showing deterministic forecast. At AccuWeather, we have addressed this challenge in several ways over the years for business customers for all weather hazards. More recently, AccuWeather launched to consumers a revolutionary method for displaying the likelihood for snow from a given storm on a location-by-location basis. Released in 2017 and refined during the last two winter seasons, AccuWeather’s Snowfall Probability product utilizes unique algorithms to process massive amounts of ensemble NWP and other types of data. The output presents users of AccuWeather’s mobile applications and website with additive information about the confidence in the predicted amount as well as the chance a storm will bring less or more snow than currently forecast. Users can view the percent chance of snowfall in various accumulation ranges for their location, while also monitoring how the percentages change over time. Feedback from the weather community and users who report on the utility of this feature has been extremely positive. The Snowfall Probability product will be discussed along with challenges and opportunities to enhance the way in which we link new and rapidly expanding model datasets to actionable information that benefits society across a variety of weather hazards.