Automatic detection and characterization of seismic tremors in northern Cascadia

• Published in: Geophysical research letters Vol. 34; no. 16; pp. L16313 - n/a
• Main Authors: Kao, Honn, Thompson, Philip J., Rogers, Garry, Dragert, Herb, Spence, George
• Format: Journal Article
• Language: English
• Published: Washington, DC American Geophysical Union 28.08.2007; Blackwell Publishing Ltd
• Subjects: automatic detection; Cascadia; Computational Geophysics; Earth sciences; Earth, ocean, space; ETS tremors; Exact sciences and technology; Geographic Location; Instruments and techniques; North America; Seismic instruments and networks; Seismic monitoring and test-ban treaty verification; Seismology; Cascadia; ETS tremors; automatic detection; algorithms; detection; slip; Scintillation index; moving average; monitoring; waveforms; seismograms
• ISSN: 0094-8276; 1944-8007
• DOI: 10.1029/2007GL030822

We develop an algorithm that automatically detects and characterizes seismic waveforms associated with episodic tremor and slip (ETS) events in northern Cascadia. For each hour‐long seismogram, the process begins with the calculation of moving average and scintillation index functions. A logic chart is designed to sort the observed waveform patterns into a set of classes based on these functions. An ETS event is always manifest by many station‐hours with coherent tremor (CT) patterns. Automatic analysis of seismic waveforms recorded in the past three years gives results consistent with the work done by tedious visual inspection of seismograms. Our algorithm, which is free from human oversight, is ideal for an automatic tremor monitoring system to give immediate notice of an ETS event.