8-10-Day Probabilistic Forecasts: What Do Users Need to Understand and Use Longer-Range Uncertainty Information in Decision-Making?

As forecast models and skill in communicating uncertainty advances, probabilistic information is becoming a more common part of forecasts issued to stakeholders and public audiences. The National Weather Service (NWS) Weather Prediction Center (WPC) currently issues precipitation and other forecasts in the 4-7 day time period. To meet rising demand for uncertainty information, WPC is developing user-tested products in the 8-10 day timeframe, and partnering with a research team to understand: what decisions are users making in the 8-10 day timeframe, what sort of uncertainty information is important to core users during this time, and how they need the information presented.

To address these questions, a research team from East Carolina University and Nurture Nature Center, working with WPC, conducted a study of users’ needs for probabilistic forecast information in the 8-10 day timeframe. Using surveys, interviews and focus groups, the research team created and iteratively tested 8-10 day probabilistic weather products and gathered feedback from stakeholders and end users about their needs for probabilistic forecast information in 8-10 day range. The team sought to understand the kinds of decisions core users make, how they use uncertainty information in those decisions, and how web-based products can best deliver information efficiently and clearly.

The presentation will provide insights into how stakeholders and core users want to integrate probabilistic forecast information, particularly in the 8-10 day time frame, into decision-making. Insights into how product design and delivery affect the use of products will also be discussed.