Integrating Physical Computation, AI, and Big Observations to Advance Hydrology and Earth Sciences

Abstract: The rapid growth of observational data presents unprecedented opportunities to enhance both the predictability and mechanistic understanding of Earth systems. However, fully harnessing big Earth data needs computational frameworks that bridge the gap between physics-based models and machine learning. In this talk, I will first demonstrate how AI methods can significantly improve the prediction of environmental systems. Despite their predictive accuracy, machine learning models often lack physical interpretability, limiting their ability for scientific inquiry. To address this, I will introduce the developed hybrid, differentiable modeling framework that unifies physical models with machine learning in an end-to-end trainable system. This framework autonomously learns from large observations while maintaining physical clarity. The machine learning components can be seamlessly embedded into physical backbones to assimilate multi-source data, support automatic parameterization, and represent uncertain processes. I will showcase applications of this framework in simulating and understanding the terrestrial water cycle and its interactions with ecosystems at continental and global scales. This talk will highlight how differentiable modeling not only improves the modeling ability in both data-rich and data-scarce scenarios, but also provides a systematic pathway to enhancing model structures, deciphering uncertain physical relations, and facilitating knowledge discovery in Earth system sciences.

Bio: Dapeng Feng is a postdoctoral scholar in Earth System Science at Stanford University. He was a postdoctoral fellow at Stanford’s Institute for Human-Centered Artificial Intelligence. He earned his Ph.D. in civil engineering at The Pennsylvania State University, where he developed the differentiable hydrologic modeling framework as a generic way to unify physical models and machine learning. He earned his Master’s degree in water resources from Peking University and his Bachelor’s degree in hydraulic engineering from Wuhan University. His current research focuses on studying the terrestrial water cycle and its interactions with plant and climate systems by integrating geoscientific models, machine learning, and large Earth observations. His research has been published in Nature Reviews Earth & Environment, Nature Communications, Geophysical Research Letters, Water Resources Research, Geoscientific Model Development, etc.

*this seminar will be held in person and online*

Topic: IACS/SoMAS Candidate: Dapeng Feng

Time: Mar 27, 2025 02:30 PM Eastern Time (US and Canada)

Join Zoom Meeting

https://stonybrook.zoom.us/j/98736659326?pwd=Wyx9QKtDnKBHF6tUw5lNzVmL3Rri3M.1

Meeting ID: 987 3665 9326

Passcode: 779358