By Manmeet Singh, Western Kentucky University

Abstract 

Artificial intelligence is rapidly transforming Earth system science by enabling faster, higher-resolution, and more adaptive modeling of the atmosphere, ocean, land, and hydrologic systems. In this talk, I will present recent work on AI-driven weather and climate modeling, including high-resolution forecasting, AI-based physical components for Earth system models, and data assimilation using deep learning and diffusion-based methods. I will also discuss how hybrid AI-physics frameworks can improve prediction of precipitation, temperature, soil moisture, and extreme events while retaining physical consistency. Finally, I will highlight emerging opportunities for integrating large language models with Earth system workflows to support scientific discovery, domain-aware reasoning, and decision support in weather, climate, water, and disaster applications.