A Synergistic Approach: Dynamics-AI Ensemble in Tropical Cyclone Forecasting

Tropical cyclones (TCs) often bring destructive winds, heavy rainfall, high waves, and storm surges, causing severe losses. Accurate TC forecasting is therefore vital for disaster mitigation. However, the chaotic nature of TCs means that small initial errors can rapidly grow, leading to large forecast uncertainties. To address this, ensemble forecasting is employed to generate multiple scenarios, quantify uncertainty, and enhance forecast reliability for disaster prevention. Ensemble forecasting has greatly improved TC predictions at major centers like ECWMF and NECP, etc. , but it requires substantial computational resources due to complex physics-based models. This study would address this challenge by developing an AI-driven optimized ensemble forecast system using Orthogonal Conditional Nonlinear Optimal Perturbations (O-CNOPs). The system bridges the gap between computational efficiency and dynamic consistency in TC forecasting. Unlike conventional ensembles limited by computational costs or AI ensembles constrained by inadequate perturbation methods, O-CNOPs generate dynamically optimized perturbations that capture fast-growing errors of FuXi model while maintaining plausibility. The key innovation lies in producing orthogonal perturbations that respect FuXi’s nonlinear dynamics, yielding structures reflecting dominant dynamical controls and physically interpretable probabilistic forecasts. The study demonstrates generally-superior deterministic and probabilistic skills over the operational Integrated Forecasting System Ensemble Prediction System, establishing a new paradigm combining AI’s computational advantages with rigorous dynamical constraints. Success in TC track forecasting paves the way for reliable ensemble forecasts of other high-impact weather systems, marking a major step toward operational AI-based ensemble forecasting. It is expected that such AI-driven optimized ensemble forecast system can be effectively applied to operational forecasts.