Exploring the Utility of Earthquake Spectra Collected From Smartphones for Ground-motion Modeling
Description:
The field of ground-motion modeling has been increasingly embracing non-ergodic ground motion models (GMMs) with spatially variable coefficients due to their superior performance in capturing highly spatially variable source, path, and site effects. The estimation of non-ergodic terms in non-ergodic GMMs relies on the availability of as many densely sampled shaking observations as possible. Traditional seismic station data may prove insufficient, creating a need to fill data gaps with unconventional datasets. Acceleration waveforms recorded via the MyShake smartphone app provide one such dataset. MyShake has been delivering ShakeAlert earthquake early warning messages to US West Coast users since October 2019. It also provides triggered waveforms recorded using the onboard accelerometer. Using a database of smartphone waveforms from California, Marcou et al. (2024, BSSA) used residual analysis to show that spatial trends in smartphone peak acceleration amplitudes recorded on smartphones have a significant positive correlation to free-field peak amplitudes, and thus smartphone data has predictive power for free-field ground motion. In this work, we extend their analysis to the spectra recorded by smartphone accelerometers, given that state-of-the-art GMMs now model Fourier Amplitude Spectra (FAS). We uniformly process smartphone FAS to investigate their general characteristics over a wide (0.5 - 25 Hz) frequency band. We compare smartphone FAS to predicted free-field FAS from a modern non-ergodic GMM, as well as to recorded free-field FAS for selected earthquakes. We probe the degree to which the response of the structures in which the smartphones record ground motion affects the recorded spectra and any ability to see site resonances. We finally discuss potential ways forward for ground-motion modeling using the smartphone FAS dataset.
Session: Modern Waveform Processing and Engineering Datasets - Accessibility, Quality Control, and Metadata - I
Type: Oral
Date: 4/17/2025
Presentation Time: 04:45 PM (local time)
Presenting Author: Savvas
Student Presenter: Yes
Invited Presentation:
Poster Number:
Authors
Savvas Marcou Presenting Author Corresponding Author savvas.marcou@berkeley.edu University of California, Berkeley |
Richard Allen rallen@berkeley.edu University of California, Berkeley |
|
|
|
|
|
|
|
Exploring the Utility of Earthquake Spectra Collected From Smartphones for Ground-motion Modeling
Session
Modern Waveform Processing and Engineering Datasets - Accessibility, Quality Control, and Metadata