Exciting news! We're transitioning to the Statewide California Earthquake Center. Our new website is under construction, but we'll continue using this website for SCEC business in the meantime. We're also archiving the Southern Center site to preserve its rich history. A new and improved platform is coming soon!

Poster #244, Seismology

Testing Three Earthquake Early Warning Algorithms (EPIC, FinDer and PLUM) on Simulated Composite Offshore Earthquakes

Debi Kilb, Colin O'Rourke, Maren Böse, Jennifer R. Andrews, Angela I. Chung, Julian Bunn, & Jeff J. McGuire
Poster Image: 

Poster Presentation

2021 SCEC Annual Meeting, Poster #244, SCEC Contribution #11187 VIEW PDF
We test the performance of three Earthquake Early Warning (EEW) algorithms (EPIC, FinDer, and PLUM) using composite earthquake sequence pairs to better understand the behavior of each algorithm, as applied to offshore earthquake sequences. Our goal is to identify pathways to optimize the ShakeAlert EEW system performance. We build composite event data sets by combining signals from a pair of nearby earthquakes, the 2019 Ridgecrest M7.1 mainshock, and an M5.4 foreshock. We sum earthquake signals recorded at the same station with different time shifts between the earthquake pairs (-60 to 180 s), where negative time shifts create a foreshock sequence and positive time shifts create an aftershoc...k sequence. To simulate offshore earthquake scenarios, we remove stations near the epicenter to create an artificial “coastline” at increasing distances from the source (ranges 25 - 150 km). This tests distances between the shoreline and the current westernmost ShakeAlert alerting boundary, allowing us to assess behavior for potential offshore events such as in the Cascadia subduction zone, Cape Mendocino triple junction, or California continental borderlands. We find that for the scenarios examined the EPIC source-based algorithm is consistently better at identifying the earthquake epicenter(s), while FinDer and PLUM successfully provide information about where large ground motions occur onshore. Importantly, the three algorithms work well in concert to identify the earthquake location(s) and provide additional information about where strong shaking occurs. We conclude that moving to ground-motion-based alert association and aggregation has the potential to obviate some identified challenges such as untangling individual foreshocks and aftershocks and can produce more robust EEW alerts for offshore events.