The Warn-on-Forecast System: Advancements Toward Operations

The Warn-on-Forecast System (WoFS) is a rapidly-updating, storm-scale ensemble prediction system designed to generate probabilistic forecasts of individual thunderstorms and their hazards up to six hours in advance. The suite of probabilistic forecast guidance includes baseline products, such as threshold-determined probabilities, percentiles, and ensemble mean environmental information, as well as machine learning-based probabilities of specific hazards. Since 2017, WoFS has demonstrated its value in supporting decision making and communication of multiple event types at National Weather Service (NWS) national centers and Weather Forecast Offices (WFOs). Due to its success and strong collaboration and coordination between the National Severe Storms Laboratory (NSSL) and NWS, efforts are underway to transition WoFS from NSSL to a demonstration mode within the NWS.

This research-to-knowledge (R2K) effort for WoFS represents a complex exercise. For example, it introduces new data assimilation workflows. Additionally, WoFS will likely require forecasters to provide enhanced services for severe weather watch and warning operations. This project entails the development of extensive documentation for data connections, a cloud-based business model, and an end-to-end concept of operations for integrating WoFS into NWS workflows to potentially contribute to the envisioned NWS ops model and community-centric warnings. The first phase of this transition is marked by a 2025–2026 demonstration period in which NWS will run WoFS routinely throughout the lower 48 states to inform the development of the concept of operations, in addition to providing real-time guidance to NWS forecasters.

This presentation will highlight the strategies for the successful advancement of WoFS into NWS, provide a status update on the project, and outline future WoFS-related goals.