Performance of bEPIC Through the 2024 <b>M</b> 7.0 Mendocino Earthquake Sequence
Abstract On 5 December 2024, an M 7.0 earthquake ruptured offshore of Cape Mendocino, California, generating felt shaking along the coast and prompting the issuance of earthquake early warning alerts and a tsunami warning. Contemporaneous to the earthquake sequence, bEPIC, an update to the ShakeAlert early warning algorithm Earthquake Point-source Integrated Code (EPIC), was undergoing development tests in real-time. The observed earthquake sequence provided a rich dataset of earthquakes in which we could test the performance of bEPIC. bEPIC improved epicenter estimates with an average 29.7 km reduction in location error compared to EPIC solutions for matched events. The improvement in location also improved per event magnitude estimates. Although EPIC on average overestimated the maximum magnitude of the earthquakes in the sequence by 0.37 magnitude units, bEPIC more closely aligned with our ground-truth catalog with a much lower overestimate of only 0.05 magnitude units. Despite including more information in the grid-search algorithm, bEPIC did not increase the average time to the first alert. The improvements in the accuracy of the earthquake parameters ultimately improve the accuracy of potential alerting polygons, which are related to the solved location and magnitude. These promising results indicate that improved accuracy with no cost to timeliness can be achieved for offshore earthquakes in northern California through the incorporation of bEPIC in early warning operating procedures.
Authors
- A. Williamson (ORCID: https://orcid.org/0000-0003-1481-725X)
- Angela Lux (ORCID: https://orcid.org/0000-0002-3767-6018)
- R. M. Allen (ORCID: https://orcid.org/0000-0003-4293-9772)
Institutions
- University of California, Berkeley (US)
Publication Details
- Journal
- The Seismic Record
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1785/0320250009
- Primary Topic
- Earthquake Detection and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- U.S. Geological Survey