Theses and Dissertations from UMD

Permanent URI for this communityhttp://hdl.handle.net/1903/2

New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM

More information is available at Theses and Dissertations at University of Maryland Libraries.

Browse

Filters

20241

Settings

Author: search.filters.author.Steeple, Jennifer LynneSubject: search.filters.subject.groundwater

Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    MODELING GROUNDWATER FLUCTUATIONS IN THE COASTAL PLAIN OF MARYLAND: AN ANN POWERED STRATEGY
    (2024) Steeple, Jennifer Lynne; Negahban-Azar, Masoud; Shirmohammadi, Adel; Environmental Science and Technology; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Groundwater management in the face of climate change presents a critical challenge with far-reaching implications for water resource sustainability. This study evaluates the effectiveness of Artificial Neural Networks (ANNs) as predictive tools for estimating current groundwater levels and forecasting future groundwater levels in the Aquia aquifer in the Coastal Plain ofMaryland. The groundwater levels of the Aquia aquifer have declined under the pressures of land use change, increases in agricultural irrigation, and population growth. We tested, trained, and employed eight county-level artificial neural network (ANNs) models to predict and project Aquia aquifer groundwater levels for the near (2030-2050) and far (2050-2100) future under two socio-economic pathways (SSP245 and SSP585). The models exhibited significant predictive performance during testing (R²= 0.82-0.99). Minimum temperature and population were the most influential variables across all county-based models. When used to forecast groundwater level under two climate scenarios, the models predicted declining groundwater levels over time in Calvert, Caroline, Queen Anne’s, and Kent counties, aligning with regional trends in the Aquia aquifer. Conversely, Anne Arundel, Charles, St. Mary’s, and Talbot counties exhibited projected increases in groundwater levels, likely influenced by correlations with the variable irrigated farm acreage, underscoring the importance of considering nonlinear relationships and interactions among variables in groundwater modeling. The study highlights the ability of ANNs to accurately predict county-scale groundwater levels, even with limited data, indicating their potential utility for informing decision-making processes regarding water resource management and climate change adaptation strategies. This study also assessed the usability of multiple methods to fill in the missing data and concluded that using the repeated groundwater level data still resulted in powerful ANN models capable of both predicting and forecasting ground water levels in the Coastal Plain of Maryland.