Date of Award
12-1-2025
Degree Name
Master of Science
Department
Civil Engineering
First Advisor
Kalra, Ajay
Abstract
Flooding is among the most destructive natural disasters globally, and it inflicts severe damage on both natural environments and human-made structures. The frequency of floods has been increasing due to unplanned urbanization, climate change, and changes in land use. Flood susceptibility maps help identify at-risk areas, supporting informed decisions in disaster preparedness, risk management, and mitigation. This study aims to generate a flood susceptibility map for two regions: Davidson County of Tennessee using an integrated geographic information system (GIS) and analytical hierarchical process (AHP), and the Briar Creek watershed of Mecklenburg County, North Carolina using an integrated GIS and machine learning (ML) algorithms. For GIS integrated AHP approach, ten flood causative factors are employed to generate flood-prone zones. AHP, a form of weighted multi-criteria decision analysis, is applied to assess the relative impact weights of these flood causative factors. Subsequently, these factors are integrated into ArcGIS Pro (Version 3.3) to create a flood susceptibility map for the study area using a weighted overlay approach. The resulting flood susceptibility map classified the county into five susceptibility zones: very low (17.48%), low (41.89%), moderate (37.53%), high (2.93%), and very high (0.17%). The FEMA flood hazard map of Davidson County is used to validate the flood susceptibility map created from this approach. For GIS integrated ML approach, three machine learning algorithms —bagging (random forest), extreme gradient boosting (XGBoost), and logistic regression—were used to develop a flood susceptibility model that incorporates topographical, hydrological, and meteorological variables. Key predictors included slope, aspect, curvature, flow velocity, flow concentration, discharge, and rainfall. A flood inventory of 750 data points was compiled from historic flood records. The dataset was divided into training (70%) and testing (30%) subsets, and model performance was evaluated using accuracy metrics, confusion matrices, and classification reports. The results indicate that logistic regression outperformed both XGBoost and bagging in terms of predictive accuracy. According to the logistic regression model, the study area was classified into five flood risk zones: 5.55% as very high risk, 8.66% as high risk, 12.04% as moderate risk, 21.56% as low risk, and 52.20% as very low risk. The FEMA flood hazard map is used to validate the flood susceptibility map created from this approach The resulting flood susceptibility maps constitute a valuable tool for emergency preparedness, infrastructure planning, and sustainable land-use management. By identifying and categorizing areas based on varying levels of flood vulnerability, these maps enable local authorities, planners, and policymakers to prioritize mitigation measures, optimize resource allocation, and enhance community resilience against future flood events.
Access
This thesis is Open Access and may be downloaded by anyone.