The Role of Continental Mesoscale Convective Systems in Forecast Busts within Global Weather Prediction Systems

Abstract

Despite significant, steady improvements in the skill of medium-range weather predictionsystems over the past several decades, the accuracy of these forecasts are occasionally very poor.These forecast failures are referred to as “busts” or “dropouts”. The lack of a clear explanationfor bust events limits the development and implementation of strategies designed to reduce theiroccurrence. This study seeks to explore a flow regime where forecast busts occur over Europe inassociation with mesoscale convective systems over North America east of the Rocky Mountains.Our investigation focuses on error growth in the European Centre for Medium-Range WeatherForecasting’s (ECMWF’s) global model during the summer 2015 PECAN (Plains Elevated Convectionat Night) experiment. Observations suggest that a close, but varied interrelationship can occurbetween long-lived, propagating, mesoscale convection systems over the Great Plains and Rossbywave packets. Aloft, the initial error occurs in the ridge of the wave and then propagates downstreamas an amplifying Rossby wave packet producing poor forecasts in middle latitudes and, in somecases, the Arctic. Our results suggest the importance of improving the representation of organizeddeep convection in numerical models, particularly for long-lived mesoscale convective systems thatproduce severe weather and propagate near the jet stream.

Keywords

• rossby wave trains
• meoscale convective systems
• forecast busts
• error growth
• medium-range numerical weather prediction
• convective parameterization
• pecan field campaign