Date of Award


Document Type


Degree Name

Doctor of Philosophy in Civil Engineering (PhD)

College, School or Department Name

Department of Civil and Environmental Engineering


David W Watkins


Veronica W Griffis


This dissertation addresses water resources decision making in the Great Lakes Basin by developing a multi-model framework for climate change impact assessment, including integrated climate and hydrologic modeling. Physically based watershed models, using soil moisture accounting and temperature index (degree-day) snowmelt algorithms, are developed, calibrated and validated to simulate baseflow, snowmelt, and surface runoff under historic conditions. Comparison with an existing model of the Great Lakes basin, the NOAA Large Basin Runoff Model (LBRM), showed improvements resulting from the increased spatial resolution and use of a more process-based snow algorithm in the Hydrologic Engineering Center's Hydrologic Modeling System (HECHMS). As an alternative to the physically based hydrologic models, and particularly appealing for ungauged basins or locations where record lengths are short, regional regression models are developed to directly predict selected streamflow quantiles, using physical basin characteristics as well as meteorological variables output by general circulation models (GCMs). Hydrologic responses are evaluated based on different combinations of hydro-climatic modeling approaches, when driven using GCM outputs. The model results, presented in a probabilistic context of multi-model predictions, provide insights to potential model weaknesses, including comparatively low runoff predictions from hydrologic models using temperature proxy potential evapotranspiration (PET) approaches and limited accuracy of regional regression models for small, groundwater-dominated watersheds. Additional insights are gained by replacing the temperature-proxy PET method with an approach that maintains a consistent energy budget between the climate and hydrologic models.

Hydrologic projections for the Great Lakes watersheds under future climates are evaluated using the model with a consistent energy budget, and differences in responses are explained by differences in watershed characteristics, aridity index, and the future climate projections. It is proposed that these hydrologic projections inform adaptive water resources decision making through a multi-stage decision model, and applications to water withdrawal permitting and BMP implementation are described. The framework developed herein demonstrates an integrated analysis of climate change impact assessment and will potentially be useful for researchers, water managers, and regulators as an aid to decision making and policy implementation.