Hybrid Models

Three Ways AI Is Rewiring Climate Prediction: Downscaling, Hybrids, and Emulators

Artificial intelligence is transforming climate prediction through three key methods: downscaling coarse global model outputs to local detail, integrating AI into physics-based hybrid models, and using AI as fast emulators for complex simulations. Each approach addresses different bottlenecks in climate science, offering enhanced resolution, physical consistency, or computational speed, but also comes with specific limitations and challenges.