Incorporating Smartphone Seismic Data for Improved Earthquake Early Warning Performance
Session: Earthquake Early Warning Live in California! Current Status and Challenges II
Type: Oral
Date: 4/23/2021
Presentation Time: 03:00 PM Pacific
Description:
MyShake is a citizen science smartphone application which uses an algorithm to filter anthropogenic noise and detect earthquake onsets in data recorded by the phones’ onboard accelerometer. ShakeAlert is the US Earthquake Early Warning (EEW) system built on a dispersed network of permanent seismic stations that feed realtime data to a detection algorithm. Each of these systems is independently capable of rapidly detecting earthquakes and generating alerts of impending shaking. The goal of our research is to assess the potential of MyShake to augment the speed and accuracy of the ShakeAlert system in the US, with a focus on the state of California where alerts are publicly available. As MyShake has been providing alerts in California since October 2019, MyShake now has a substantial user base in the state, roughly 500,000 users (compared to the ~1000 ShakeAlert stations), with the densest clusters in urban areas.
We compare the detection capabilities of MyShake and ShakeAlert in combination and independently, using real data collected by both systems in earthquakes since October 2019. On an event-by-event basis, we assess whether incorporating MyShake data in the ShakeAlert system could have sped up detection and/or improved initial estimates of location and size. Preliminary results find that MyShake triggers can arrive at a server a full second earlier than the first ShakeAlert triggers. Both when added to ShakeAlert or when assessed independently, MyShake can speed event creation by up to half a second without compromising the quality of the location estimate, as compared to ShakeAlert operating alone. We also explore the efficacy of different algorithm approaches to incorporating MyShake data to best improve ShakeAlert performance.
Presenting Author: Sarina Patel
Student Presenter: Yes
Authors
Sarina Patel Presenting Author Corresponding Author sarina.patel@berkeley.edu University of California, Berkeley |
Angela Chung aichung@berkeley.edu University of California, Berkeley |
Qingkai Kong kongqk@berkeley.edu University of California, Berkeley |
Jenn Strauss jastrauss@berkeley.edu University of California, Berkeley |
Richard Allen rallen@berkeley.edu University of California, Berkeley |
|
|
|
|
Incorporating Smartphone Seismic Data for Improved Earthquake Early Warning Performance
Category
Earthquake Early Warning Live in California! Current Status and Challenges