Updates to the Rapid Intensification Deterministic Ensemble (RIDE) for the Joint Typhoon Warning Center’s Area of Responsibility

Abstract

Rapid intensification (RI) of tropical cyclones poses significant forecasting challenges, especially in data-limited operational environments. The Joint Typhoon Warning Center’s (JTWC) logistic-regression-based Rapid Intensification Deterministic Ensemble (RIDE) model became increasingly constrained by its dependence on having a complete set of input forecast models. The complete input requirement resulted in a limited number of RIDE forecasts during recent seasons. A new model called RIDE update addresses these limitations. RIDE update applies a naïve-Bayes classification approach to estimate RI probabilities using an updated set of tropical cyclone intensity forecast models to generate deterministic forecasts when probabilities exceed 40%. The updated suite of intensity forecast models utilizes models that are more consistently available within JTWC’s area of responsibility. And because of its naïve-Bayes formulation, it operates effectively even with a limited number of input forecasts. Independent evaluation over the 2025 western North Pacific, Indian Ocean, and Southern Hemisphere tropical cyclone seasons demonstrates that RIDE update provides skillful forecasts across multiple RI thresholds, with performance comparable to JTWC’s existing deterministic-to-probabilistic statistical model (DTOP) while offering improved robustness and continuous availability. Since January 2025, RIDE update has replaced the original RIDE in JTWC operations under the familiar RIDE identifier.