Lessons Learned from Codeveloping the New Operational Subseasonal Forecast for Laypeople on Yr

Abstract Subseasonal forecasting bridges the gap between short-term weather forecasts and seasonal outlooks, covering lead times of 2–6 weeks. While these forecasts are often operationally presented as anomalies over broad areas due to decreasing skill with lead time, this approach might be less easy to comprehend for laypeople. Typical users of weather apps are used to basing their decisions on highly localized forecasts and are often unfamiliar with interpreting anomalies. In this paper, we explore the development and implementation of a localized subseasonal (3-week) forecast, presenting actual weather variables with their uncertainties, integrated into the existing Yr weather service. Steered by coproduction with existing Yr users, we prioritized creating a forecast that emphasizes usability over maximizing forecast skill. Through this process, we gained valuable insights into ways to present a subseasonal forecast that users find useful and actionable. Our user feedback suggests that, despite the inevitable decline in skill by week 3, localized forecasts presenting actual weather variable values still provide significant value, as long as forecast uncertainty is clearly communicated. Additionally, we discuss how the subseasonal forecast was implemented into the Yr weather service, with a primary focus on obtaining consistency with existing forecast products. Ensuring coherence across products is essential, as user feedback indicates that most users base their decisions on combining the information of the medium-range (10 days) and subseasonal forecasts. Significance Statement This paper shares the lessons we learned from codeveloping a 3-week weather forecast with users from the general public and integrating it into the Yr weather service. Forecasts for longer time periods are more uncertain, and the way this information is presented can further influence how easily people can interpret and use it. To make the forecast useful, it was important to ensure that users could easily grasp both the weather information and the uncertainty involved. The main outcome of this work is a way to present the 3-week forecast that users find practical and easy to interpret, while also clearly showing the uncertainty, helping them make informed decisions.