Systematic Model Forecast Errors of Surface Cyclones in the NGM and AVN, January 1990

Abstract

Results from a study examining the performance of the nested grid model (NGM) and the aviation run of the global spectral model (AVN) in predicting surface cyclones during January 1990 revealed that the AVN slightly outperformed the NGM in forecasting cyclone central pressures and placement. Although both models performed better for deepening systems than filling systems, the AVN outperformed the NGM in predicting the characteristics of filling cyclones. Overall, the NGM tended to overdeepen surface cyclones. A large part of the pressure error was due to the model's inability to properly fill cyclones and a tendency to forecast systems to deepen when they were observed to be filling. The AVN tended to underdeepen surface cyclones with the deepening rate errors near 2 mb at 12 h and less than 1 mb by 48 h. The overall pressure errors for deepening cyclones appeared to be linked to a spin-up problem in the AVN and may have also been associated with the AVN cold bias in 1000- to 500-mb thickness forecasts.