Category: Watershed

394712 - A hierarchical Bayesian model for streamflow regionalization in the Great Lakes basin

Wednesday, Jun 6
8:30 AM – 5:30 PM

The water balance of the Great Lakes is notoriously difficult to characterize. The surface area of the lakes accounts for 32% of the total drainage area of the basin, and therefore on-lake precipitation and evaporation, which are very difficult to measure, contribute substantially to the total water balance. In addition, a substantial fraction of the contributing river network is ungaged, leading to large uncertainties in runoff contributions. These uncertainties compound, making it difficult to determine the cause of anomalously low and high water levels, with large implications for management. In this work, we develop a new approach to reduce the uncertainty in runoff estimation across the Great Lakes Basin. We adapt the commonly used drainage area ratio method (ARM) into a probabilistic ARM framework that allows watershed conditions, time-varying climate, and spatial proximity to inform how monthly flow is scaled from multiple gaged catchments to an ungaged site. The proposed model is developed and calibrated in a hierarchical Bayesian framework to pool regional information across sites and enhance regionalization skill. The proposed model is tested for 224 watersheds in the Great Lakes basin in a cross-validation framework, and is compared to the standard ARM approach currently used in water balance calculations for the lakes. We will also discuss a broader vision to integrate this approach into a formal Bayesian framework to characterize the uncertainty in all water balance components for the Great Lakes.

Co-Authors: Scott Steinschneider, Ithaca – Cornell University

Kuk-Hyun Ahn

Postdoctoral Fellow
Department of Biological and Environmental Engineering, Cornell University
Amherst, New York

Kuk-Hyun Ahn

Postdoctoral Researcher, Department of Biological and Environmental Engineering, Cornell University, NY.

Kuk-Hyun Ahn joined the graduate program in Civil Engineering at Purdue University in 2011 and received Doctor of Philosophy degree in December, 2014. Kuk-Hyun Ahn is now working as Postdoctoral Researcher for Department of Biological and Environmental Engineering, Cornell University, NY