Emergency Managers Perspective of Probabilistic Information during the 2024 Watch-to-Warning Experiment

Despite the increasing development of probabilistic forecast guidance, which has the potential to enable forecasters to communicate important weather information to a wide range of end-users, the period between severe weather watches and warnings often lacks sufficient forecast information. This presents a scientific and societal challenge that is currently a focal point for severe weather research. To address this, the 2023 Hazardous Weather Testbed (HWT) conducted the Watch-to-Warning Experiment (W2WE) to explore how forecasters could communicate and utilize tools to generate rapidly-updating, probabilistic severe weather products between watches and warnings. In this experiment, 3 WFO forecasters and 1-2 SPC forecasters per week worked together to issue 2-hour "amoeba" watches, short-term mesoscale discussions, probabilistic hazard information (PHI) plumes, warnings, and public-facing graphics. They utilized tools such as the PHI Tool, the Warn-on-Forecast System (WoFS), AWIPS, and a new machine learning product called WoFS-PHI. The 2023 W2WE showed the potential value of watch-to-warning products and information for the end-users from the SPC and WFOs. However, the products produced in that experiment were not tested by end-users. The 2024 W2WE aimed to address the potential of end-user engagement and use of probabilistic information. During the last case of each week of the 2024 experiment, the forecasters produced products for and communicated with real-world emergency managers (EMs). Each EM was assigned a personalized scenario with details relevant to their jurisdiction. At the end of the activity, the EMs completed a survey on the usability of probabilistic information in their operations and the most helpful products/information for their decision-making. We will highlight the activities and survey results from the EMs’ perspective during the 2024 W2WE gaining insight into how EMs could use rapidly-updating, probabilistic forecast products.