NH017-06
A Source Clustering Approach for Efficient Inundation Modeling and Regional Scale PTHA
A Source Clustering Approach for Efficient Inundation Modeling and Regional Scale PTHA
Wednesday, 9 December 2020: 10:50
Virtual
Abstract:
For coastal regions on the margin of subduction zones, nearfield megathrust earthquakes are the source of the most extreme tsunami hazards and are important to capture properly when computing hazard curves for inundation depth, as a part of a Probabilistic Tsunami Hazard Assessment (PTHA). Typically, great variability in inundation depth at any point is possible due to the extreme variation in extent and pattern of slip over the fault surface.
We present an approach to estimating inundation depth probabilities in this context that consists of two components. The first component uses a Karhunen-Loève expansion to express the probability density function for all possible events, using a method developed in previous work by the authors with parameters that result in a density function that is geophysically reasonable. It is computationally cheap to generate a large N number of samples from this density function. However, to obtain reasonable results for extreme flooding due to rare events, N would have to be so large as to make the necessary tsunami simulations prohibitively expensive. The second component uses importance sampling techniques results to ensure we adequately sample the tails of the distribution and properly re-weight the probability assigned to the resulting realizations. We then further cluster the realizations into a small number of clusters that we believe will give similar inundation patterns in the region of interest, derived from results of computationally cheap low-resolution simulations. Thus only one high-resolution tsunami inundation simulation needs to be computed from a representative realization in each cluster, and weighted by the sum of the weights of all realizations in the cluster. We illustrate the methodology by considering two coastal locations on the Cascadia Subduction Zone margin.
We present an approach to estimating inundation depth probabilities in this context that consists of two components. The first component uses a Karhunen-Loève expansion to express the probability density function for all possible events, using a method developed in previous work by the authors with parameters that result in a density function that is geophysically reasonable. It is computationally cheap to generate a large N number of samples from this density function. However, to obtain reasonable results for extreme flooding due to rare events, N would have to be so large as to make the necessary tsunami simulations prohibitively expensive. The second component uses importance sampling techniques results to ensure we adequately sample the tails of the distribution and properly re-weight the probability assigned to the resulting realizations. We then further cluster the realizations into a small number of clusters that we believe will give similar inundation patterns in the region of interest, derived from results of computationally cheap low-resolution simulations. Thus only one high-resolution tsunami inundation simulation needs to be computed from a representative realization in each cluster, and weighted by the sum of the weights of all realizations in the cluster. We illustrate the methodology by considering two coastal locations on the Cascadia Subduction Zone margin.