Model Configuration versus Driving Model: Influences on Next-Day Regional Convection-Allowing Model Forecasts during a Real-Time Experiment

Abstract As part of NOAA’s Hazardous Weather Testbed Spring Forecasting Experiment (SFE) in 2020, an international collaboration yielded a set real-time convection-allowing model (CAM) forecasts over the contiguous United States which configurations and initial/boundary conditions were varied controlled manner. Three employed, among Finite Volume Cubed-Sphere (FV3), Unified Model (UM), Advanced Research version (WRF-ARW) dynamical cores represented. Two runs produced for each configuration: one driven by Global Forecast System initial boundary conditions, other Met Office’s operational global UM. For 32 cases during SFE2020, these initialized at 0000 UTC integrated 36 h. Objective verification fields relevant to convective forecasting illuminates differences influence configuration versus driving pertinent ongoing problem optimizing spread skill CAM ensembles. The UM WRF tend outperform FV3 precipitation, thermodynamics, simulated radar reflectivity; using with native core also tends produce better aggregate. Reflectivity thermodynamic found cluster more than lead times greater 18 two experiments had notably similar solutions that, despite competitive aggregate skill, large errors diurnal cycle.