Abstract
Abstract. The potential of a full-margin rupture along the Cascadia subduction zone poses a significant threat over a populous region of North America. Previous probabilistic tsunami hazard assessment studies produced hazard curves based on simulated predictions of tsunami waves, either at low resolution or at high resolution for a local area or under limited ranges of scenarios or at a high computational cost to generate hundreds of scenarios at high resolution. We use the graphics processing unit (GPU)-accelerated tsunami simulator VOLNA-OP2 with a detailed representation of topographic and bathymetric features. We replace the simulator by a Gaussian process emulator at each output location to overcome the large computational burden. The emulators are statistical approximations of the simulator's behaviour. We train the emulators on a set of input–output pairs and use them to generate approximate output values over a six-dimensional scenario parameter space, e.g. uplift/subsidence ratio and maximum uplift, that represent the seabed deformation. We implement an advanced sequential design algorithm for the optimal selection of only 60 simulations. The low cost of emulation provides for additional flexibility in the shape of the deformation, which we illustrate here considering two families – buried rupture and splay-faulting – of 2000 potential scenarios. This approach allows for the first emulation-accelerated computation of probabilistic tsunami hazard in the region of the city of Victoria, British Columbia.
Highlights
The Cascadia subduction zone is a long subduction zone that expands for more than 1000 km along the Pacific coast of North America, from Vancouver Island in the north to northern California in the south (Fig. 1)
Compared to the hazard values at Seaside, Oregon, as calculated by Park et al (2017) for the 1/1000 probability of exceedance, the expected hazard at Victoria sites is significantly lower. One factor for these discrepancies is the location of the two sites as Seaside is impacted by the tsunami waves from the Cascadia subduction zone directly, whereas Victoria is protected by the Olympic Peninsula and the western side of Vancouver Island; to reach sites in Victoria, the waves have to travel much farther and are attenuated along their path
A sequential design algorithm was employed for the conduction of the computational experiments for earthquake-generated tsunami hazard in the Cascadia subduction zone
Summary
The Cascadia subduction zone is a long subduction zone that expands for more than 1000 km along the Pacific coast of North America, from Vancouver Island in the north to northern California in the south (Fig. 1). Sequential designs adaptively select the set of experiments to optimize the training data for fitting the emulators Such a design can be determined by the efficient mutual information for computer experiments (MICE) algorithm (Beck and Guillas, 2016) that we utilize in this study for probabilistic tsunami hazard prediction in northern Cascadia. Owen et al (2017) demonstrated, by examples involving computer-intensive simulation models, that GP emulation can approximate outputs of nonlinear behaviour with higher accuracy than polynomial regression when considering small- to moderate-sized, space-filling designs The benefit of this approach is the use of a sequential design algorithm in the training to maximize the computational information gain over the multidimensional input space and adaptively select the succeeding set of experiments.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
