Improving Tropical Cyclogenesis Statistical Model Forecasts through the Application of a Neural Network Classifier

A binary neural network classifier is evaluated against linear discriminant analysis within the framework of a statistical model for forecasting tropical cyclogenesis (TCG). A dataset consisting of potential developing cloud clusters which formed during the 1998-2001 Atlantic hurricane seasons is used in conjunction with eight large-scale predictors of TCG. Each predictor value is calculated at analysis time. The model yields a probability forecast for genesis at 6 hour intervals out to 48 hours prior to the event. Results consistently show that the neural network classifier outperforms linear discriminant analysis on all performance measures examined, including probability of detection, false alarm rate, Heidke Skill Score, and forecast reliability.