Evaluating High-Resolution Model Performance of Convective Initiation During the 2017 Severe Weather Season

Predicting convective initiation in severe weather environments is a challenge. If this challenge is met, then lives and property can be saved. Lead time is critical. Unfortunately, these hi-res models rarely capture complex convective environments with complete accuracy, which leads to errors with respect to initiation times. The goal of this project is to assess the ability of different Convection Allowing Models (CAMs) to correctly predict the timing of convective initiation during severe weather events. We compare CAM model runs (HRRR, parallel version of HRRR), the CONUS nest of the NAM, and operational and parallel versions of the Hi-Res Windows) to radar observations during the 2017 severe weather season as a part of our skill evaluation process. Understanding the biases of CAMs can invoke improvements which lead to better predictions, a safer public and stronger economy.