Impact of Cascadia Subduction Zone Earthquakes on Coastal Bridges in Washington State

Abstract

The Cascadia Subduction Zone in the Pacific Northwest means is likely to generate a magnitude 9 earthquake in the future. A set of 30 synthetic (physics-based) ground motions were created by researchers from the University of Washington and United States Geological Survey. Using an existing velocity model of the region and assuming a surface shear wave velocity of 600 m/s. This research uses the simulated ground motions as input into regional-scale bridge damage analyses. To account for the variation of near-surface soil conditions across the Pacific Northwest, amplification factors for each site class were generated for 10 locations in the Washington State. To apply the amplification factors to other locations, the medians and dispersions of the amplification factors of all 10 locations for all soil profiles were computed for each class. In the area specifically of interest, Gray’s Harbor and Pacific counties, the most common bridge type is a prestressed concrete girder bridge. Four bridge categories were identified including prestressed simply supported concrete bridges built before 1990, prestressed simply supported bridges built on or after 1990, prestressed continuous concrete bridges built on or after 1990, and movable bridges, which were included because of their unique vulnerability and importance in the region. Fragility curves, which can be used to determine the likely damage state of a bridge given a ground motion intensity, were assigned for each bridge category. The selected fragility curves came from a variety of sources including from HAZUS as developed by FEMA and from research literature. Given the different realizations for ground motions, the variability in amplification factors, and the various fragility curves, a parametric study was completed using at first a single bridge. Then a regional case study was completed for all 483 bridges in Gray’s Harbor and Pacific counties. In the regional case study two different sets of ground motions were used, as well as two site classes and associated amplification factor variability, and three sets of fragility relationships; creating 12 unique cases that included variation in damage predictions due to each component of the analysis. From both of these studies it was concluded that the variation among the possible fragility curves most affect the estimated damage state of the bridges. An important factor in assessing the impact of the ground motions on the region was the number of single span bridges, because these are far less likely to be damaged during an earthquake than multi-span bridges. The bridge inventory in the southern half of Gray’s Harbor and Pacific counties were found to be more vulnerable to the given M9 ground motions than the bridges in the northern half.