Date of Completion
Marina Astitha, Emmanouil Anagnostou, Malaquías Peña
Field of Study
Master of Science
Numerical weather prediction (NWP) models are important tools used by federal agencies and the weather enterprise to inform and warn the public about extreme weather events. Snowstorm forecasts, which can be highly impactful throughout the Northeast United States, face challenges related to microphysical and initial and boundary layer processes that hinder the accuracy of forecasts. In addition, observations of mixed phase precipitation are very sparse spatially and temporally which causes an added challenge to validate predictions. This study presents an evaluation of 38 significant snowstorms and 18 rain/wind events from 2005 to 2020 using three NWP models. Each modeling system has been evaluated using observations of atmospheric variables from the Integrated Surface Dataset for onshore stations, the National Data Buoy Center for offshore stations, the University of Wyoming’s archive for radiosonde stations for the vertical profile, and the Global Historical Climatology Network database for accumulated variables such as liquid water equivalent, snow water equivalent, and snowfall. Results demonstrate that moisture related variables perform poorly at the surface and aloft, wind gust performs poorly at the surface for onshore stations, and the accuracy of snow ratio algorithms primarily depends upon a modeling system’s overall configuration. Furthermore, ice storm prediction is hindered by precipitation type diagnosis, and quantitative precipitation forecasts can be significantly impacted by a modeling system’s domain configuration. Recommendations to improve winter weather prediction includes creating a nested domain with feedback to reduce quantitative precipitation forecast error and applying the European Centre for Medium-Range Forecast’s non-convective wind gust scheme to reduce wind gust bias. For the 18 wind gust events which involved a different model configuration and different planetary boundary layer schemes, wind gust parameterizations involving turbulent kinetic energy performed best.
Walters, Michael, "Evaluation of Winter Weather Prediction during Extreme Snowfall Events and Analysis on Wind Gust Prediction for Non-Convective Rain and Wind Events" (2020). Master's Theses. 1544.