Degree

Doctor of Philosophy (PhD)

Department

Civil and Environmental Engineering

Document Type

Dissertation

Abstract

Numerical modeling has contributed significantly to the understanding of groundwater systems. Many challenges are associated with constructing groundwater models which include an accurate understanding of the geology and aquifer parameters estimation. Traditionally boreholes are a successful way to capture geological features, however, boreholes often have sparse data. Airborne electromagnetic (AEM) data allows for efficient and cost-effective surveying of large areas, providing valuable information about the subsurface electrical resistivity. By bridging the gap between boreholes, AEM data offers a broader view of the aquifer system's structure and heterogeneity. However, interpreting geophysical AEM data has uncertainties. Developing a framework to apply the data fusion concept on boreholes and AEM data is one of the challenges addressed in the first part of this study. The framework introduces depth-dependent resistivity thresholds and unsupervised clustering to classify resistivity into lithofacies, reducing interpretation uncertainties. Applied to the Mississippi River Valley alluvial aquifer (MRVA), the second most-pumped US aquifer, it improves lithofacies models and reduces estimation uncertainties. Employment of AEM resistivity data in groundwater quality, the second part introduces a transformation from resistivity to subsurface quality using a deep learning model and regression formulas that predict chloride concentration using location, AEM resistivity, borehole data, and water quality data. This method gives us a clear picture of the saltwater plumes in 3D for the first time. By mapping the extent and intensity of salinity in the MRVA aquifer, the model provides valuable insights into the sources and controls of groundwater salinity, supporting proactive water resources management. As the final goal of this study, a detailed groundwater model was constructed based on the complex geological model from the data fusion study. This study generates maps to guide risk-based water management decisions and protect aquifer resources by delineating the cones of depression. Moreover, it highlights the significant influence of increased groundwater pumping on groundwater levels and fresh groundwater, evaluating changes in up-coning from deeper saline aquifers through a coupled flow-transport model which enhances our understanding of groundwater flow and storage responses to excessive withdrawals. Ultimately, these findings offer critical insights for effective groundwater management in regions experiencing substantial pumping pressures.

Date

1-29-2025

Committee Chair

Frank Tsai

Download

Available for download on Thursday, May 01, 2025

Share

COinS