Adaptive Wavelet Thresholding and Grouping Approach for Repetitive Seismic Swarm Detection: Application to the Noto Peninsula Earthquake Swarm
Description:
Repetitive seismic swarms serve as good indicators of fluid movement and aseismic deformation in the crust and may alert the imminence of devastating earthquakes. Before the occurrence of the 2024 Mw7.5 Noto Peninsula, Japan earthquake, intense seismic swarms involved with an upward fluid migration are continuously reported in the northeastern Noto Peninsula since December, 2020. However, most of earthquakes during the ~3-year swarm activity have small magnitudes and are hardly detectable. In this study, we develop an adaptive wavelet approach to identify earthquake swarms buried in noisy data recordings. By characterizing seismic waveforms in the frequency and time domain using wavelets, a short-term/long-term average magnitude (STA/LTA) algorithm is firstly performed to adaptively divide the whole trace into different wave segments. Noise is removed though soft-thresholding approach and then the signals having compatible characteristics after normalization are identified as repeating earthquake swarms and classified into different groups based on scalogram similarity. We apply the technique for continuous time series data using Hi-net stations in the Noto Peninsula. The result for a daily recording of September 28, 2023 show that tens of small-magnitude (<=Mb 4) earthquake swarms missing in the catalog can be clearly identified with improved SNR. The wavelet method provides a computationally efficient way for real-time, rapid detection of repetitive earthquake swarms, which can further improve existing earthquake catalogs, enabling a detailed investigation of swarm activity and implications of fluid migration.
Session: The 2024 Magnitude 7.5 Earthquake and the Associated Earthquake Swarm Beneath the Noto Peninsula, Central Japan [Poster Session]
Type: Poster
Date: 5/3/2024
Presentation Time: 08:00 AM (local time)
Presenting Author: Jia
Student Presenter: No
Invited Presentation:
Authors
Jia Zhang Presenting Author Corresponding Author jzhang@cuhkri.org.cn Chinese University of Hong Kong |
Hongfeng Yang hyang@cuhk.edu.hk Chinese University of Hong Kong |
|
|
|
|
|
|
|
Adaptive Wavelet Thresholding and Grouping Approach for Repetitive Seismic Swarm Detection: Application to the Noto Peninsula Earthquake Swarm
Category
The 2024 Magnitude 7.5 Earthquake and the Associated Earthquake Swarm Beneath the Noto Peninsula, Central Japan