Analysis of Model Thermal Profile Forecasts Associated with Winter Mixed Precipitation within the United States Mid-Atlantic Region

Abstract

Winter mixed-precipitation events across the mid-Atlantic region of the United States from 2013–2014 through 2018–2019 were used to analyze common short-term model forecasts of vertical atmospheric thermal structure. Using saturated forecast soundings of the North American Mesoscale (NAM), higher-resolution nested NAM (NAMnest), and the Rapid Refresh models—corresponding with observed warm-nose precipitation events (WNPEs)—several thermal metrics formed the basis of the analysis of observed and forecast soundings, including Bourgouin positive and negative areas. While the three models accurately forecast the general thermal structure well during WNPEs, a warm bias is evident within each. Well forecast are maximum and minimum temperatures within the warm nose and surface-based cold layer, respectively, but the cold layer is commonly too thin for each of the models, and the warm nose is regularly too thick, particularly within NAM and NAMnest forecasts. Forecasts of a cold layer that is too shallow tend to coincide with observations of stronger synoptic-scale upward motion, a deeper cold surface-based layer, and a higher isentropic surface. Forecasts of a warm nose that is too thick tend to coincide with observations of weaker upward motion, a shallower cold surface-based layer, and a lower isentropic surface across the region. Two-thirds of precipitation-type estimates from model soundings agreed with those derived from observed soundings, with the remaining third predominantly representing a warm bias in precipitation type.