Climate-Informed Nonstationary Multihazard Design Load Characterization for Performance-Based Evaluation of Coastal Structures

Abstract Current structural design guidelines face increasing challenges from interacting natural hazards in coastal regions. Climate-driven shifts in hazard intensity, frequency, and interaction add further complexity. Traditional multihazard design approaches often neglect hazard nonstationarity or treat hazards independently, resulting in inaccurate structural demands. This paper presents a probabilistic framework for simulating future multihazard scenarios under changing climate conditions. The nonstationary trends in hazard intensities and frequencies are characterized by probability distributions and regression models with time- and climate-informed parameters, respectively. Interhazard dependence is captured using multivariate copulas on the detrended hazard intensities. The framework supports two approaches: historical data-driven (suitable for regions with sufficient observations) and hybrid simulation-based (leveraging physics-based models to supplement limited observations). The framework’s application is demonstrated on coastal wind and flood hazards for San Francisco (historical data-driven) and Miami (hybrid simulation-based), producing joint hazard curves (wind speed versus water level at a target return period) under multiple shared socioeconomic pathway scenarios through 2100. The resulting joint hazard curves provide a risk-consistent basis to select wind–flood intensity pairs for adaptive performance-based evaluation of coastal structures under future climate conditions. At the 100-year return period, ASCE 7 wind speeds are lower by about 21% and 17% in 2050 and 2100 for San Francisco, respectively, and by 6% and 9% for Miami, reinforcing the need to assess structural performance over the full joint hazard curve to capture compound loading scenarios and preserve reliability under nonstationary conditions.