Enhancing AI-Based Tropical Cyclone Track and Intensity Forecasting via Systematic Bias Correction

2026-03-19

AI Summary

This work addresses the limitations of existing AI-based tropical cyclone forecasting methods, which suffer from coarse-resolution reanalysis data, grid discretization artifacts, and regression loss biases, leading to significant errors in track and intensity prediction—particularly for intense typhoons. To overcome these challenges, we propose BaguanCyclone, a unified framework that introduces a novel probabilistic center refinement mechanism to enhance track accuracy and a region-aware intensity prediction module that leverages high-resolution modeling to capture extreme features in the storm core. By integrating continuous spatial distribution modeling with dynamic sub-grid representations, our approach is trained and validated across six global ocean basins using IBTrACS multi-basin data, consistently outperforming both state-of-the-art numerical models and AI baselines—especially in challenging scenarios such as rapid intensification, sharp recurvature, binary typhoon interactions, and stalled systems.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Tropical cyclones (TCs) pose severe threats to life, infrastructure, and economies in tropical and subtropical regions, underscoring the critical need for accurate and timely forecasts of both track and intensity. Recent advances in AI-based weather forecasting have shown promise in improving TC track forecasts. However, these systems are typically trained on coarse-resolution reanalysis data (e.g., ERA5 at 0.25 degree), which constrains predicted TC positions to a fixed grid and introduces significant discretization errors. Moreover, intensity forecasting remains limited especially for strong TCs by the smoothing effect of coarse meteorological fields and the use of regression losses that bias predictions toward conditional means. To address these limitations, we propose BaguanCyclone, a novel, unified framework that integrates two key innovations: (1) a probabilistic center refinement module that models the continuous spatial distribution of TC centers, enabling finer track precision; and (2) a region-aware intensity forecasting module that leverages high-resolution internal representations within dynamically defined sub-grid zones around the TC core to better capture localized extremes. Evaluated on the global IBTrACS dataset across six major TC basins, our system consistently outperforms both operational numerical weather prediction (NWP) models and most AI-based baselines, delivering a substantial enhancement in forecast accuracy. Remarkably, BaguanCyclone excels in navigating meteorological complexities, consistently delivering accurate forecasts for re-intensification, sweeping arcs, twin cyclones, and meandering events. Our code is available at https://github.com/DAMO-DI-ML/Baguan-cyclone.

Problem

• tropical cyclone
• track forecasting
• intensity forecasting
• systematic bias
• discretization error

Innovation

• probabilistic center refinement
• region-aware intensity forecasting
• sub-grid representation
• tropical cyclone forecasting
• bias correction