MULTI-SCALE ASSESSMENT OF HURRICANE-GENERATED COASTAL AND INLAND FLOODING USING A NUMERICAL AND OBSERVATIONAL FRAMEWORK

Abstract

Tropical cyclones - also called hurricanes and typhoons - result in major damage and fatalities around the world almost every year due to the massive coastal and inland flooding. Integrating coastal and inland regional flooding models as well as assessing the impacts of climate change helps better understand the overall risk of a hurricane at the present and in the future. Coastal flooding causes damages to the coastal structures. While regional models are not capable of accurately predicting the damage to the coastal structures, the high-fidelity computational fluid dynamics (CFD) models can be implemented to simulate the flow properties around the coastal structures. These CFD models should be forced by regional models due to their high computational cost. This dissertation tries to address some knowledge gaps in the literature with regard to coastal flood risk using both numerical and experimental techniques. In particular, compound coastal-inland flooding, climate change and sea level rise, and multi-scale numerical modeling have been focused. In the first manuscript of this dissertation, the impacts of a number of factors (reservoirs, historical textile mill dams, and bridges) on the severity of a record-braking flood within a watershed in the New England, Pawtuxet River, were assessed. These factors are currently omitted within the risk assessments tools such as flood insurance rate maps. Further, to better understand possible future risks in a warmer climate, an extreme flood event under a synthetic wet hurricane was simulated. It was shown that this hurricane can generate a flooding equivalent to a 500-yr event in this watershed. Further research is necessary to develop the integrated coastal and inland flood models that are forced by a single atmospheric model. In the second manuscript, the impact of sea level rise (SLR) to estimate the water elevation in the regional storm surge models was studied using two methods. The linear method superimposes the SLR linearly to the estimated water elevations while the nonlinear method uses a new model with updated data to include the impact of SLR. A simplified theoretical formulation, a number of idealized cases, and two real case studies were assessed to compare the linear and nonlinear methods. In general, based on the results of the idealized and real studies, a discrepancy of up to 10 % between the linear and nonlinear approaches is expected in estimation of maximum water elevation. Further research is necessary to investigate the effect of SLR and climate change