Date of Award

5-2015

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Legacy Department

Civil Engineering

Advisor

Weichiang Pang

Committee Member

Firat Y Testik

Committee Member

Nadarajah Ravichandran

Committee Member

Hsein Juang

Abstract

This research work focused on developing and evaluating the methodologies proposed for hazard quantification and loss estimation of the combined hurricane wind and flood. The overall research was divided into three studies. A methodology to quantify the joint occurrence probability of hurricane wind and flood was first introduced in Study I. A procedure to select ensembles of hazard-consistent full-track hurricane scenarios (i.e. with similar return periods) was proposed next in Study II. Finally in Study III, a new hurricane mitigation framework was developed using the agent-based modeling (ABM) technology to evaluate the effectiveness of incentive programs under both hurricane wind and flood hazards. In Study I, the peak wind speeds and peak flood elevations were calculated for candidate hurricanes in a database of 50,000 years of simulated hurricane events. A wind field model and a boundary layer model were used to compute the surface wind speeds. The flood elevations were calculated using a storm surge model. The joint mean recurrence interval (MRI) contour plots and the joint hazard maps were introduced to quantify the joint hurricane wind and flood hazards for a case study area (Charleston Peninsula, South Carolina). The joint hazard maps contain paired wind speed and flood elevation contour lines with constant joint occurrence probabilities (or MRIs). It was found that the hurricane events with different wind speed and flood elevation combinations can have the same joint MRI or return period. Using the method developed in Study I to quantify the joint MRI for hurricane wind and flood hazards, a new methodology was then proposed for the selections of the hazard-consistent hurricane scenarios (i.e. with similar joint MRIs). The selected hurricane scenarios were imported into the HAZUS-MH (Hazards U.S. - Multi-Hazards) program for combined hurricane wind and flood loss estimations. We found out that the joint occurrence probability of hurricane wind and flood should be considered in scenario-based hurricane loss estimations, because hurricane scenarios selected based on only wind speeds or only flood elevations may overestimate the combined losses. To evaluate the effectiveness of different incentive levels on mitigating regional hurricane hazards, a new hurricane mitigation framework was developed using agent-based modeling approach in Study III. This framework models each individual building and householder (agent), as well as the change in building inventory due to constructions and demolitions. The decisions of agents whether to implement certain retrofit measures are influenced by incentive and the hazards experienced by the agents. An ABM loss assessment case study was performed for Miami-Dade County, Florida using property tax reduction (PTR) as incentive for retrofitting buildings to mitigate hurricane hazards. By varying the maximum reduction of property tax, it was found that the PTR between 50% and 75% was an optimal level in terms of minimizing the overall societal cost for Miami-Dade County. The ABM loss assessment methodology developed in this study can be used by coastal jurisdictions for structure design and hazard mitigation, or insurance companies that need to determine the optimal incentive-based insurance premiums.

Share

COinS