Authors: William Currier – NOAA Physical Sciences Laboratory
Title: Does Deep Learning Absorb Ensemble Forecast Post-Processing Benefits for Probabilistic Streamflow Prediction?
Presentation Type: Poster Presentation
Abstract: As part of a FY25 project, Pathways to Improving Flash Flood Forecasting, RTI, the University of Vermont, and the University of Hawaiʻi at Manoa developed graphics that visualize the decision-making process of key stakeholders before and during flash floods. Using data from over 10 meetings with WFOs and EMs in three different mountainous regions (VT, NC, HI), we summarize how forecast lead time, certainty, and severity impact their decisions and actions. With AI-supported thematic analysis, we identified barriers stakeholders face and solutions they need. These findings will help inform strategic next steps to improving flash flood forecasting, grounded in operational realities.