Study of the Potential Earthquake Risk in the Western United States by the LURR Method Based on the Seismic Catalogue, Fault Geometry and Focal Mechanisms

Abstract

Based on the load/unload response ratio (LURR) theory, spatial and temporal variation of Y/Y c (value of LURR/critical value of LURR under 90 % confidence) in the western United States and its adjacent area (31°–44°N, −128° to −112°E) during the period from 1980 to 2011 was studied. The selected study area was zoned into 20 sub-regions, in each of which the fault geometry and the focal mechanisms were very similar such that the stress fields were almost uniform. The loading and unloading periods were determined by calculating perturbations in the Coulomb failure stress in each sub-regions induced by earth tides. Earthquakes occurring in these sub-regions were identified as a loading or unloading type, and the response rate was chosen as the Benioff strain that can be calculated from earthquake magnitude M. With a time window of 1 year, a time moving step of 1 month, a space window of a circle region with a radius of 100 km, and a space moving step of 0.5° latitudinally and longitudinally, snapshots of the evolution of Y/Y c were generated. Scanning results show that obvious Y/Y c anomalies can be detected near the epicenter of all big earthquakes larger than M6.5 in regions with reasonable seismic monitoring abilities. They also show Y/Y c anomalies occurred several years prior to the big earthquakes and the lasting time of the anomaly is from one year to several years. For some LURR anomalous regions, however, no earthquakes occurred. According to the characteristics of LURR anomalies, two regions with a high risk of big earthquakes were detected. One is between the northern region of the Bay Area and the Mendocino triple junction (38°–40°N, −124° to −122°E) and the other is between Lake Tahoe and Mono Lake (37.5°–39.5°N, −120° to −118°E) along the border of California and Nevada.