The performance of coastal infrastructure is threatened by natural hazards such as hurricanes and floods. The intensity and frequency of many of these climate related hazards are expected to be influenced by climate change, which will add uncertainty to the performance of coastal infrastructure. Furthermore, coastal regions are experiencing rapid population growth, which is expected to continue in the future as well. Therefore, in view of multiple hazards, uncertainty due to climate change, and increasing coastal population, comprehensive performance assessment of regional portfolio of coastal infrastructure is essential for managing the existing infrastructure and ensuring adequate performance after extreme events such as hurricanes and earthquakes. Therefore, this study focuses on the development of a methodology and supporting tools that can be used to facilitate comprehensive multi-hazard performance assessment of regional portfolios of costal infrastructure. Particularly, this study focuses on above ground storage tanks (ASTs) and bridges since they are crucial components of the energy and transportation infrastructure, respectively, and they have been observed to fail in past events with severe economic and environmental impact.
First, in order to describe and effectively communicate the effects of multiple hazards on the performance and design decisions of infrastructure systems, taxonomy for multi-hazard combination is developed. For understanding the effects of different hazards on the performance and design of structures and classifying hazards according to the taxonomy this study develops a dual layer metamodel based fragility assessment methodology. The proposed method harnesses statistical learning techniques to enable efficient response and reliability assessment of structures with a broad range of design details when subjected to hazardous conditions. Using the dual layer fragility assessment methodology, fragility functions are developed for most common failure modes of ASTs: due to strong winds and storm surge. For bridges, the fragility assessment framework is used to develop fragility function for most common causes of bridge failure such as: scour, scour and vehicular loads, barge impact, barge impact and scour, hurricanes, earthquakes, and vehicle loads and earthquakes.
This study also develops frameworks for ASTs and bridges to facilitate resilience assessment, i.e. their post event functionality and recovery time, which is essential for comprehensive performance assessment of ASTs and bridges. For ASTs subjected to storm surge and strong winds, the estimates of repair costs, repair time, and estimates of potential spill volumes due to tank failures are developed. Similarly, for bridges, a methodology is developed to assess the entire distribution of repair costs considering uncertainties in damage to bridge components, repair costs, and repair actions. Additionally, empirical seismic damage data is used to develop decision trees that can determine the traffic restrictions and their duration for given the damage state of bridge components such as columns, bearings, and abutments. Finally, in order to develop structure specific performance targets from regional level performance targets this study develops a simple heuristic methodology which determines the performance targets for individual structures commensurate to their performance.
All the fragility and resilience assessment frameworks are applied to individual structures and regional portfolio of structures which has provided several valuable insights in to the performance of ASTs and bridges. The application of the fragility functions for ASTs for tanks in the Houston Ship Channel shows that tanks are more vulnerable to storm surges than strong winds. Additionally, application of the fragility functions show that mitigation measure such as anchoring tanks can lead to contradictory effects on the performance of ASTs for different failure modes such as flotation and buckling. For bridges, application of the fragility functions highlights that earthquakes and hurricanes can have competing effects on selection of column height. Furthermore, estimation of seismic repair costs highlights the multi-modal nature of the distribution of repair costs. The use of decision trees to determine traffic restrictions and their duration highlights the influence of damage to different bridge components on traffic restrictions. Finally, the application of the heuristic methodology to determine structure specific performance targets for all the ASTs in the Ship Channel shows that anchoring of tanks and simple procedural measures can significantly reduce the spill volumes. Similarly, application of the heuristic methodology for a portfolio of bridges in the Charleston region shows the framework can be used to obtain bridge specific performance targets which can significantly improve the performance of the entire bridge network.