NWS Probabilistic Precipitation Forecasting: Lessons Learned and Pathways Forward (330)

Jeff Waldstreicher, NOAA/NWS Eastern Region HQ, Bohemia, NY

David Radell, NOAA/NWS Eastern Region HQ



Five years ago NWS Eastern Region initiated a pilot project at WFOs Sterling and Taunton to enhance the communication of uncertainty associated with winter storms. Now a nationwide effort, the NWS Probabilistic Snowfall experiment was expanded to 82 forecast offices for the 2017-2018 winter season. In addition, probabilistic ice accumulations were produced for the first time last winter for internal evaluation. As a parallel effort, experimental probabilistic QPF is being produced at 7 test offices starting in the spring of 2018. All three probabilistic precipitation suites leverage information generated from the Weather Prediction Center's (WPC) and employ similar methodologies to produce forecast probabilities. Common output products and web display have been designed for all three precipitation types to facilitate consistency in messaging as well as user interpretation and comprehension. Much has been learned during the previous 5 winters of the probabilistic snowfall forecast experiment about the communication and utility of snowfall uncertainty information. User surveys, feedback from core partners, and interviews and focus groups conducted by Eastern Research Group (ERG) has yielded considerable insights on user needs for winter weather information, and how uncertainty information is understood and potentially applied. This has resulted in annual adjustments to the probabilistic products produced, their associated web pages displays, and best practices for incorporating this information within IDSS briefings and social media posts. This presentation will discuss the evolution of the probabilistic snowfall information, particularly the changes made for the 2017-2018 winter and the associated user feedback on these changes. While considerable improvements have been made, many challenges remain in communicating precipitation uncertainty information, including the subtle but important distinction between forecast uncertainty and forecaster confidence. These challenges include: ensemble skill in adequately capturing the atmospheric uncertainty and simulating mesoscale phenomena considerations of audience diversity in formulating messaging and presentation media (e.g., automated static web page graphics, vs. forecaster annotated graphics in published presentation sides and social media posts vs. live interactions such as webinar presentations and on-site IDSS forecaster presentations). Improved forecaster tools and training to interact with ensembles to more effectively understand and formulate uncertainty messaging are also needed.