Date of Award

8-16-2024

Document Type

Open Access Dissertation

Department

Civil and Environmental Engineering

First Advisor

Erfan Goharian

Abstract

Flood risks in coastal and inland areas have historically been underestimated and underpredicted, resulting in increased flood damage and loss of human life. Flood risk under climate change and sea level rise will disproportionately impact low-wealth and marginalized communities, as they are often located in areas with less flood protection and have limited resources to adapt to the impacts of flooding. The systematic workflow of the life cycle of FRM is frequently disengaged in the current state of flood hazard assessment with system vulnerability assessment. This research aims to improve the life cycle of FRM by incorporating flood observation, monitoring, modeling, and system vulnerability assessment into a FRM framework for risk-informed decision making of FRM.

Firstly, the research is done developing an integrated multidimensional hydrologic-hydraulic-hydrodynamic (MH3) process-simulation aiming to perform enhanced compound flood modeling in coastal areas. A tight-coupling procedure is applied to represent nonlinear and complex compound flooding processes in coastal urban watersheds. This procedure interconnected multidimensional hydraulics (pipe and channel), hydrodynamics (2D overland flow), and distributed hydrologic models at fine scale physical modeling using a flexible triple mesh configuration including a basin-node-link configuration. The modeling framework has been built for a complex drainage network consisting of tidal creeks, tidal channels, underground sewer networks, and detention ponds. Additionally, a Gaussian process-based Bayesian optimization algorithm is used to optimize model structural uncertainty in real-time flood modeling. This framework optimizes uncertainty in real-time flood modeling by minimizing error propagation caused by uncertain parameters. The MH3 model improves urban flood simulation by about 15% to 33% using LiDAR Digital Surface Model and fine-scale meshing process in physical representation. Further, Gaussian Process based uncertainty optimization improves real-time flood forecasting skills by 6-11%. Therefore, MH3 and the uncertainty optimization process can enhance flood forecasting skills significantly.

Finally, a data-driven coastal vulnerability assessment is engaged in a socio-environmental system using an integrated coastal vulnerability index (CVI) to identify the resilience gaps of a socio-environmental system. This framework for a coupled socio-environmental system simultaneously combines information from biophysical and socio-economic factors. Moreover, the explanatory power of index-based coastal vulnerability approaches is evaluated using two proxy data sets associated with post-disaster outcomes: the historical flood inventory and the cost of fatalities data. Six different vulnerability assessment approach using deterministic and probabilistic factor aggregation helps to identify multi-scale vulnerability information using CVI ranging from 30m grid level to county scale. Therefore, it can be envisaged that the developed enhanced compound flood model structures, modeling tools and vulnerability framework can provide valuable decision supporting information to vulnerable populations in coastal and inland areas that are exposed to flood risk. Moreover, flood risk managers, who are accountable for controlling flood risk, can make informed decisions based on predicted risk for FRM.

Rights

© 2024, Ahad Hasan Tanim

Download

Share

COinS