Evaluation of reduced computational approaches to assessment of tsunami hazard and loss using stochastic source models: Case study for Tofino, British Columbia, Canada, subjected to Cascadia megathrust earthquakes

Abstract

Probabilistic tsunami hazard and risk analyses are important decision support tools in developing tsunami risk reduction strategies and actions for coastal communities. A stochastic source modeling method facilitates the consideration of uncertainties associated with earthquake rupture processes. However, the computational costs are high when inland tsunami inundation and building damage need to be evaluated accurately. To develop a practical solution by keeping the computational requirements at a manageable level, probabilistic tsunami hazard analysis based on low-resolution tsunami simulations but considering a wide range of possible earthquake ruptures can be used to identify smaller sets of stochastic rupture models for target probability levels. These identified stochastic rupture models can be used to obtain the estimates of tsunami building loss by running high-resolution tsunami inundation simulations. A case study is set up for Tofino, British Columbia, Canada, under the potential tsunami threat from the Cascadia megathrust earthquakes to investigate the correlation between the maximum modeled tsunami wave amplitudes at offshore locations and the tsunami building loss. A practical solution is proposed to obtain the tsunami risk estimates based on a limited number of high-resolution tsunami inundation simulations, thus reducing the computational costs for the probabilistic tsunami risk analysis. The effectiveness of the approach is demonstrated by comparing the median value of the tsunami risk estimates from 20 stochastic rupture model simulations that are selected based on probabilistic tsunami hazard analysis for a representative offshore location using the low-resolution tsunami simulations (i.e. 270 m grids) with the full probabilistic tsunami risk analysis of the target building portfolio based on the 1200 high-resolution tsunami simulations (i.e. 5 m grids).