Date of Award


Document Type

Open Access Master's Thesis

Degree Name

Master of Science in Civil Engineering (MS)

Administrative Home Department

Department of Civil and Environmental Engineering

Advisor 1

Pengfei Xue

Committee Member 1

David Watkins

Committee Member 2

Brian Barkdoll


The Great Lakes of North America are the largest surface freshwater system in the world and many ecosystems, industries, and coastal processes are sensitive to the changes in their water levels. The recent changes in the Great Lakes climate and water levels have particularly highlighted the importance of water level prediction. The water levels of the Great Lakes are primarily governed by the net basin supplies (NBS) of each lake which are the sum of over-lake precipitation and basin runoff minus lake evaporation. Recent studies have utilized Regional Climate Models (RCMs) with a fully coupled one-dimensional (1D) lake model to predict the future NBS, and the Coordinated Great Lakes Regulating and Routing Model (CGLRRM) has been used to predict the future water levels. However, multiple studies have emphasized the need for a three-dimensional (3D) lake model to accurately simulate the Great Lakes water budget. Therefore, in this study, we used the Great Lakes-Atmosphere Regional Model (GLARM) along with the Large Basin Runoff Model (LBRM) and CGLRRM to predict the changes in NBS and water levels by the mid- and late twenty-first century. GLARM is a 3D regional climate modeling system for the Great Lakes region that is fully coupled to a 3D hydrodynamic lake and ice model. This is the first study to use such an advanced model for water level prediction in the Great Lakes. We found that both annual over-lake precipitation and basin runoff are most likely to increase into the future. We also found that annual lake evaporation is most likely to decrease in Lake Superior but increase in all the other lakes. We posit that the decreases in evaporation are due to decreased wind speed over the lakes and decreased difference between saturated and actual specific humidity over the lakes. Our predicted changes in the three components of NBS would lead to mostly increased NBS and water levels in the future. The ensemble average of our predicted water level changes for Lake Superior, Michigan-Huron, and Erie are +0.14 m, +0.37 m, and +0.23 m by the mid-twenty-first century, respectively, and +0.47 m, +1.29 m, and +0.80 m by the late twenty-first century, respectively. However, due to the multiple sources of uncertainties associated with climate modeling and predictions, the water level predictions from this study should not be viewed as exact predictions. These predictions are unique to our model configuration and methodology. Other studies can easily predict different water level changes through the use of different models and methodologies. Therefore, more predictions from advanced modeling systems like GLARM are needed to generate a consensus on future water level changes in the Great Lakes.