A Data-Driven Regional Model for Skillful Medium-Range Typhoon Prediction

TL;DR

This paper introduces HITS, a regional AI system trained on high-resolution data that combines dynamical constraints and perceptual training to significantly improve medium-range typhoon forecasts in the Asia-Pacific region.

Contribution

The paper presents a novel hybrid AI framework that integrates large-scale dynamical constraints with regional autoregressive models and perceptual training for improved typhoon prediction.

Findings

HITS reduces typhoon intensity errors by up to 47.8% at 72 hours.

HITS improves precipitation structure representation.

HITS produces a near-unbiased wind-pressure relationship.

Abstract

Accurate prediction of tropical cyclones remains a major challenge for both numerical weather prediction and emerging artificial intelligence weather prediction systems. While recent global AI models have demonstrated strong skill in large-scale circulation prediction, they often struggle to represent the mesoscale structures critical for tropical cyclone intensity and precipitation. Here we develop the Hybrid Intelligent Typhoon System (HITS), a regional AI forecasting framework for medium-range typhoon prediction over the Asia-Pacific region, trained on a newly constructed 9-km high-resolution typhoon reanalysis dataset. The model combines regional autoregressive prediction with large scale dynamical constraints from the state-of-the-art ECMWF Artificial Intelligence Forecasting System (AIFS), allowing it to remain dynamically consistent with the evolving large-scale circulation while resolving mesoscale structures. HITS is further extended with a structure-aware perceptual training strategy (HITS-LPIPS) that improves the representation of convective and typhoon rainband structures. Experiments show that the hybrid framework substantially improves precipitation structure and typhoon intensity forecasts compared with both purely autoregressive regional AI models and standalone AI downscaling approaches. In particular, HITS-LPIPS reduces intensity errors by up to 47.8% relative to AIFS at a 72 hour lead time and produces a near-unbiased wind-pressure relationship for simulated typhoons. These results demonstrate that dynamically constrained regional AI systems provide a promising pathway for improving medium-range typhoon prediction.