Award Date
5-15-2026
Degree Type
Thesis
Degree Name
Master of Science (MS)
Department
Geoscience
First Committee Member
Hannes Bauser
Second Committee Member
Michael Nicholl
Third Committee Member
Deena Hannoun
Fourth Committee Member
Monika Neda
Number of Pages
74
Abstract
Elevated water temperatures can pose a significant threat to dam infrastructure, potentially damaging turbines, overheating internal components, and forcing generator shutdowns. This study uses a backcasting framework to evaluate how future scenarios may result in elevated water temperatures. Backcasting defines undesirable outcomes and works backward to identify the conditions that lead to them. The developed backcasting framework integrates 3D physics-based simulations with a Long-Short Term Memory (LSTM)surrogate model and SHapely Additive exPlanations (SHAP) interpretation. The combination captures temporal water temperature dynamics and quantifies the contributions of different drivers to elevated water temperature releases. As proof of concept, these methods are applied to intake structures at Hoover Dam that supply water to power-generating units. The results show that increased water temperatures going through the intakes are mainly driven by low lake elevation, high air temperature, high solar radiation, high inflow temperature, and high Schmidt Stability Index. Together, the identified drivers suggest that a combination of external climate forcing, upstream river conditions, and lake mixing dynamics drive elevated water temperatures when water is withdrawn from a reservoir. This study serves as a case study for Lake Mead and Hoover Dam, illustrating how the methods can be applied to assess water quality around critical infrastructure under changing environmental conditions.
Keywords
AEM3D; Drought; Management; Modeling; Reservoir; Water quality
Disciplines
Applied Mathematics | Hydrology | Water Resource Management
File Format
File Size
2400 KB
Degree Grantor
University of Nevada, Las Vegas
Language
English
Repository Citation
Ledres, Eunice, "Harnessing Backcasting to Identify Drivers of Critical Warming at Hoover Dam Using Hydrodynamic and Machine Learning Models" (2026). UNLV Theses, Dissertations, Professional Papers, and Capstones. 5566.
https://oasis.library.unlv.edu/thesesdissertations/5566
Rights
IN COPYRIGHT. For more information about this rights statement, please visit http://rightsstatements.org/vocab/InC/1.0/