Probabilistic Emulation of a Global Climate Model with Spherical DYffusion
This work addresses the problem of efficient, data-driven climate simulations for climate scientists and policymakers, representing a significant advance but building on existing frameworks.
The authors tackled the challenge of extending data-driven deep learning to climate modeling by developing the first conditional generative model for accurate and physically consistent global climate ensemble simulations, achieving near gold-standard performance and stable 100-year simulations at 6-hourly timesteps with low computational overhead.
Data-driven deep learning models are transforming global weather forecasting. It is an open question if this success can extend to climate modeling, where the complexity of the data and long inference rollouts pose significant challenges. Here, we present the first conditional generative model that produces accurate and physically consistent global climate ensemble simulations by emulating a coarse version of the United States' primary operational global forecast model, FV3GFS. Our model integrates the dynamics-informed diffusion framework (DYffusion) with the Spherical Fourier Neural Operator (SFNO) architecture, enabling stable 100-year simulations at 6-hourly timesteps while maintaining low computational overhead compared to single-step deterministic baselines. The model achieves near gold-standard performance for climate model emulation, outperforming existing approaches and demonstrating promising ensemble skill. This work represents a significant advance towards efficient, data-driven climate simulations that can enhance our understanding of the climate system and inform adaptation strategies.