Advancing compound coastal flood modeling on aouthern O’ahu, Hawai’i: A hybrid stochastic approach

Abstract

Sea-level rise and changing patterns of storminess associated with climate change are expected to increase the frequency and severity of compound coastal flooding events, posing significant challenges to coastal communities in enhancing preparedness and adaptation strategies. This work presents a hybrid stochastic approach for the probabilistic assessment of compound coastal flooding. The stochastic climate emulator TESLA is employed to generate synthetic time series of oceanographic and hydrologic conditions, incorporating meteorologic-oceanographic drivers such as sea-level anomalies, tropical cyclones, storm surge, tides, wind-waves, and precipitation. These time series are downscaled using a hybrid statistical-numerical framework that integrates surrogate models of high-fidelity hydrodynamic simulators (e.g., Delft3D, SWAN, SWASH, SFINCS). The framework is applied to southern O’ahu, Hawai’i, to simulate flood exposure under present climate as well as various sea-level rise scenarios and climate projections. The framework is designed to support decision-making processes by facilitating the participatory development of dynamic adaptation pathways. By allowing for rapid evaluation of flood risks, the approach provides valuable insights for building resilient adaptation strategies for vulnerable coastal communities.