Revealing the Potential of Learnable Perturbation Ensemble Forecast Model for Tropical Cyclone Prediction

TL;DR

This paper introduces FuXi-ENS, a novel AI-based ensemble forecasting model using learnable perturbations, which improves tropical cyclone prediction accuracy and better captures atmospheric dynamics compared to traditional methods.

Contribution

The paper presents a new learnable perturbation scheme for ensemble forecasting, demonstrating its effectiveness in tropical cyclone prediction over traditional ECMWF-ENS models.

Findings

FuXi-ENS outperforms ECMWF-ENS in predicting physical variables and track forecasts.

FuXi-ENS better captures large-scale circulation and moisture distribution around TCs.

The model reduces ensemble spread and improves forecast accuracy.

Abstract

Tropical cyclones (TCs) are highly destructive and inherently uncertain weather systems. Ensemble forecasting helps quantify these uncertainties, yet traditional systems are constrained by high computational costs and limited capability to fully represent atmospheric nonlinearity. FuXi-ENS introduces a learnable perturbation scheme for ensemble generation, representing a novel AI-based forecasting paradigm. Here, we systematically compare FuXi-ENS with ECMWF-ENS using all 90 global TCs in 2018, examining their performance in TC-related physical variables, track and intensity forecasts, and the associated dynamical and thermodynamical fields. FuXi-ENS demonstrates clear advantages in predicting TC-related physical variables, and achieves more accurate track forecasts with reduced ensemble spread, though it still underestimates intensity relative to observations. Further dynamical and thermodynamical analyses reveal that FuXi-ENS better captures large-scale circulation, with moisture turbulent energy more tightly concentrated around the TC warm core, whereas ECMWF-ENS exhibits a more dispersed distribution. These findings highlight the potential of learnable perturbations to improve TC forecasting skill and provide valuable insights for advancing AI-based ensemble prediction of extreme weather events that have significant societal impacts.