Water Resources Research Act Program

Details for Project ID UT_2025_Shuai

Advance Baseflow Prediction in Snow-Dominated Watersheds Using Physics-informed Machine Learning

Institute: Utah
USGS Grant Number:
Year Established: 2025 Start Date: 2026-08-01 End Date: 2029-07-31
Total Federal Funds: $308,962 Total Non-Federal Funds: $308,982

Principal Investigators: Pin Shuai

Project Summary: Streams in the mountainous headwaters in the Western United States depend heavily on groundwater, known as baseflow, which provides more than half of total streamflow. Measuring baseflow accurately is essential for managing water, but traditional methods often fail in snow-dominated mountain regions where flow paths are complex and high-water periods can last for months. Machine learning offers new opportunities, yet it is limited by data needs and difficulties in interpreting results. This project will combine physical models with machine learning to improve daily baseflow estimates in the Upper Colorado River Basin, including areas without measurements. We will also use new tools to explain how factors such as soil type and climate influence baseflow. Our goals are to reconstruct past baseflow, separate baseflow into different sources, and project future changes under drought conditions. The results will improve understanding of mountain water systems and support planning for a changing climate.