Probabilistic risk assessment of coastal infrastructure systems under climate change

Abstract

The vulnerability of coastal infrastructure systems can be affected by various climate-related hazards, such as tropical cyclones. Over the long term, the impact of hazards on systems may be magnified due to exposure to more frequent and more intense storm events in a changing climate. Therefore, it is necessary to evaluate the potential change in risk of infrastructure systems during the life cycle under climate change. In previous studies, stationary stochastic models (e.g., homogeneous Poisson process) have been widely employed to model hazard occurrence and magnitude in a stationary environment. However, in a changing climate, uncertainties associated with time-varying and nonstationarity characteristics cannot be ignored. It has been widely indicated that non-stationary stochastic processes (e.g., non-homogeneous Poisson process) should be employed. This study aims to develop a probabilistic framework to assess the risk of systems under tropical cyclones hazards in a changing climate. Nonstationary stochastic models are employed for risk assessment to quantify uncertainties in hazard frequency and intensity. Particularly, the impact of these uncertainties on the probabilistic distribution of total risk and the tail risk will be explored. It is identified that the previous studies purely based on the expected risk neglect potential extreme losses significantly. The proposed approach is applied to a highway bridge for illustration. The proposed probabilistic framework can be further applied to risk assessment of infrastructure networks and other hazard scenarios under changing conditions.