CHIMP: A 162 Year Dataset of Consistently Reinterpreted Seismic Intensities in California and Implications for Hazard Assessment

Session: Back to the Future: Innovative New Research with Legacy Seismic Data
Type: Oral
Date: 4/28/2020
Time: 08:30 AM
Room: 230 + 235
Description: 

Historical seismic intensity data are useful for myriad reasons, including assessment of the performance of Probabilistic Seismic Hazard Assessments (PSHA) by comparing PSHA predictions with a database of historically observed intensities in the region. To assess the performance of PSHA for California, we have created a long and consistently interpreted intensity record. The California Historical Intensity Mapping Project (CHIMP) dataset centralizes and reinterprets existing intensity information that has been available in disparate and sometimes hard-to-access sources. CHIMP also includes new observations of intensity obtained through archival research, and, in two cases, oral history collection. The CHIMP dataset will continue to grow as additional historical events are analyzed. Version 1 of the database includes intensities for 62 earthquakes with estimated magnitudes from 4.7 to 7.9. The compilation of 162 years of shaking data shows maximum historically observed shaking lower than expected from seismic hazard maps. Past studies have found similar discrepancies in Italy and Japan. Possible explanations for the discrepancy in California include limitations of the CHIMP dataset, such as the non-uniform spatio-temporal sampling due to historical population distribution or event incompleteness. The discrepancy could also be caused by limitations of the maps, for instance if the assumed seismicity rates are too high or if ground motion models over-predict shaking. There is also the possibility that California activity throughout the historical period has been lower than the long-term average, either due to random variability or a stress-shadow due to heightened activity from 1800 - 1918 that released much of the accumulated stress. The most recent earthquake rupture forecast model for California (Field et al., 2014) in fact assumes that long-term state-wide earthquake rates are higher than historic rates.

Presenting Author: Leah Salditch

Authors