Intelligent Shanghai Typhoon Model (ISTM): A generative probabilistic emulator for typhoon hybrid modeling

TL;DR

The paper introduces ISTM, a probabilistic generative model that improves typhoon intensity prediction by downscaling coarse AI weather forecasts to high-resolution data, outperforming existing models.

Contribution

It presents a novel two-stage UNet-Diffusion framework for high-resolution typhoon forecasting, integrating AI and physics-based models for enhanced accuracy.

Findings

Outperforms ERA5 and baseline UNet in structure and intensity prediction.

Effectively maps AI forecasts to high-resolution typhoon data.

Enables fast, physically consistent downscaling of typhoon forecasts.

Abstract

To address the systematic underestimation of typhoon intensity in artificial intelligence weather prediction (AIWP) models, we propose the Intelligent Shanghai Typhoon Model (ISTM): a unified regional-to-typhoon generative probabilistic forecasting system based on a two-stage UNet-Diffusion framework. ISTM learns a downscaling mapping from 4 years of 25 km ERA5 reanalysis to a 9 km high resolution typhoon reanalysis dataset, enabling the generation of kilometer-scale near-surface variables and maximum radar reflectivity from coarse resolution fields. The evaluation results show that the two-stage UNet-Diffusion model significantly outperforms both ERA5 and the baseline UNet regression in capturing the structure and intensity of surface winds and precipitation. After fine-tuning, ISTM can effectively map AIFS forecasts, an advanced AIWP model, to high-resolution forecasts from AI-physics hybrid Shanghai Typhoon Model, substantially enhancing typhoon intensity predictions while preserving track accuracy. This positions ISTM as an efficient AI emulator of hybrid modeling system, achieving fast and physically consistent downscaling. The proposed framework establishes a unified pathway for the co-evolution of AIWP and physics-based numerical models, advancing next-generation typhoon forecasting capabilities.