Towards more actionable climate modeling—Optimizing land parameters in km-scale climate models

Assistant Professor 

Department of Earth Sciences

University at Buffalo

Over the past decades, many regions in the Arctic have experienced unprecedented warming, leading to noticeable shifts in hydrologic regimes, such as significant temperature increases, changes in precipitation patterns, permafrost degradation, glacier melt and alterations in seasonal snowpack. These changes directly impact local communities, with Indigenous communities disproportionately affected. Given the ongoing trajectory of climate change, the impacts on water-related sectors in cold regions are expected to intensify, underscoring the urgency of quantifying these effects for effective adaptation and mitigation.

In this seminar, I will present a study on robust climate assessments of hydrologic cycles in Alaska and Yukon through advanced modeling techniques and knowledge co-production. I utilized emerging techniques, such as machine learning optimization, to calibrate parameters for high-resolution numerical models, thereby improving both hydroclimate and terrestrial hydrologic simulations.

Additionally, I actively incorporated inputs from Indigenous communities through an interactive knowledge co-production process to inform key modeling decisions. The resulting dataset is more tailored to community needs and gains greater trust from end-users. This high-resolution dataset also enables climate scientists, social scientists, and Alaskan and Yukon communities to assess localized climate impacts with exceptional temporal and spatial granularity.

Dr. Yifan Cheng is a computational hydrologist and currently an Assistant Professor in the Department of Earth Sciences. Before joining UB, he was a Project Scientist at the National Center for Atmospheric Research in Boulder Colorado. He received his PhD from the Department of Civil and Environmental Engineering at the University of Washington, Seattle in 2020.

His research focuses on leveraging innovative modeling techniques, interdisciplinary collaboration, and knowledge co-production towards more convergent and actionable hydrologic sciences, with a strong focus on cold regions and water-ecosystem nexus.