Investigating the Performance of Earthquake Early Warning Algorithms on the Cascadia Subduction Zone

Date: 4/25/2019

Time: 06:00 PM

Room: Grand Ballroom

Earthquake early warning (EEW) algorithms use the first few seconds of an ongoing earthquake to rapidly predict when and where strong ground shaking is expected. The usefulness of an EEW alert depends on the accuracy of ground motion predictions and whether the alert provides enough warning to take mitigating actions. The low level of background seismicity at the Cascadia subduction zone (CSZ) makes it difficult to test EEW algorithms. However, we can assess the expected performance of point-source and finite-source EEW algorithms for events on the CSZ with a global dataset of local recordings of subduction zone earthquakes and a suite of thirty 3-D ground motion simulations of M9 Cascadia megathrust rupture scenarios. By re-arranging seismic stations to mimic the station density and source-to-site distances in the Pacific Northwest, we test the accuracy and timeliness of EEW algorithms in Cascadia and estimate expected warning times.

Presenting Author: Mika Thompson

Authors