Improving Tropical Cyclone Forecasting With Video Diffusion Models
This work addresses disaster preparedness by improving cyclone forecasting accuracy and horizon, though it is incremental as it adapts existing video diffusion models to a specific domain.
The paper tackles tropical cyclone forecasting by applying video diffusion models with temporal layers and a two-stage training strategy, resulting in a 19.3% improvement in MAE, 16.2% in PSNR, and 36.1% in SSIM over prior methods, extending the reliable forecasting horizon from 36 to 50 hours.
Tropical cyclone (TC) forecasting is crucial for disaster preparedness and mitigation. While recent deep learning approaches have shown promise, existing methods often treat TC evolution as a series of independent frame-to-frame predictions, limiting their ability to capture long-term dynamics. We present a novel application of video diffusion models for TC forecasting that explicitly models temporal dependencies through additional temporal layers. Our approach enables the model to generate multiple frames simultaneously, better capturing cyclone evolution patterns. We introduce a two-stage training strategy that significantly improves individual-frame quality and performance in low-data regimes. Experimental results show our method outperforms the previous approach of Nath et al. by 19.3% in MAE, 16.2% in PSNR, and 36.1% in SSIM. Most notably, we extend the reliable forecasting horizon from 36 to 50 hours. Through comprehensive evaluation using both traditional metrics and Fréchet Video Distance (FVD), we demonstrate that our approach produces more temporally coherent forecasts while maintaining competitive single-frame quality. Code accessible at https://github.com/Ren-creater/forecast-video-diffmodels.