Evaluating the Depiction of Severe Convection in FV3

The Finite Volume Cubed Sphere (FV3) model will replace the GFS in 2019 and eventually become the core around which NOAA’s modelling systems unify. As such, research and development at convection-allowing scales is needed to understand the convective characteristics of storms in the FV3. During the 2018 Spring Forecasting Experiment, fourteen FV3 configurations with convection-allowing nests were run by three agencies: the National Severe Storms Laboratory (NSSL), Center for Analysis and Prediction of Storms (CAPS), and the Geophysical Fluid Dynamics Laboratory (GFDL). These runs provide a rich dataset to help understand the performance of the FV3 in forecasting severe convective weather during the height of the spring severe convection season.

Typical convection-allowing model parameters considered by forecasters when issuing severe convective outlooks will be examined herein, such as updraft helicity (UH), reflectivity, and environmental parameters. UH climatologies and surrogate severe fields help determine appropriate UH thresholds to use as severe weather proxies within FV3. Cases from the five weeks of the 2018 Spring Forecasting Experiment will be tested and supplemented with subjective evaluation results from participants in the 2018 SFE, which showed differences in participant impressions of the different forecasts. FV3 forecasts are evaluated alongside the HRRRv3, to identify strengths and weaknesses of the FV3 compared to current guidance. Evaluation metrics will be displayed using the scorecard framework developed using the Model Evaluation Tools (MET) software package, in addition to traditional graphics.