S002-0019
The 2020 M7.8 Alaska Peninsula Interface Earthquake Ground Motions and Alaska-Specific Ground Motion Models

Monday, 7 December 2020
Poster
Chris H Cramer, University of Memphis, CERI, Memphis, TN, United States and Anuradha Mahanama, Center for Earthquake Research and Information, Memphis, TN, United States
Abstract:
Appropriate ground motion models (GMMs) are needed to properly estimate seismic hazard in a region. The July 22, 2020 M7.8 Alaska Peninsula subduction interface earthquake provided significant ground motion observations to help evaluate Alaska subduction GMMs. Current GMMs tend to bracket the M7.8 observations with considerable scatter in the GMM curves with distance. However, the Atkinson and Boore (2003) interface model tends to over predict observed ground motions, the Abrahamson et al. (2016) interface model, with its fore-arc and back-arc alternatives, tends to bracket the observations (with fore-arc to the high side and back-arc to the low side), and the Zhao et al. (2006) interface model tends to under predict observations. More importantly, the two Next Generation Attenuation (NGA) Subduction interface GMMs tend to under predict the M7.8 observations, with the Kuehn et al. (2020) Vs30=760m/s model coming the closest at the lower edge of the observations. Few Alaska interface subduction earthquake observations exist at distances less 100 km, and the M7.8 earthquake only contributes one more. Recent M7 in-slab subduction earthquakes in south-central Alaska have a tendency for their ground motion observations to fall above current Alaska GMMs, particularly in fore-arc regions. For deep in-slab earthquakes below 80 km in depth in south-central Alaska, there is one back-arc region with unusually low ground motions to the north and west of Cook Inlet. Although mainshocks tend to have ground motions higher than expected, aftershocks appear to have ground motions consistent with current Alaska GMMs, particularly Abrahamson et al. (2016). This suggests that subduction mainshocks have consistently higher stress parameters (stress drops) than their aftershocks. A preliminary check on earthquake stress parameters shows this to be the case, with mainshocks having Brune values above 20 MPa and aftershocks having values in the 1 – 10 MPa range. The M7.8 event had a value of ~25 MPa. Higher mainshock stress parameters and subducting slab geometry in fore-arc regions likely contribute to observations being higher than current GMMs for Alaska.