Implementation of an Automatic Polarization S‐Wave Picker for Local Earthquake Relocation and Tomography in South‐Central Tibet

Evan Riddle, John Nabelek, Jochen Braunmiller

Research output: Contribution to journalArticlepeer-review

Abstract

<p> We expand on the automatic polarization&hyphen;based <em> S </em> &hyphen;wave&hyphen;arrival detection technique of <a> Cichowicz (1993) </a> and <a> Diehl <em> et al. </em> (2009) </a> to make it more effective for a broader range of distances and crustal structures. We use the characteristic function of <a> Cichowicz (1993) </a> , which combines the polarization attributes of rectilinearity, directivity, and transverse&hyphen;to&hyphen;total&hyphen;energy ratio and, as does <a> Diehl <em> et al. </em> (2009) </a> , modulates it with an amplitude weight function. Compared to Diehl, we increase the weight function exponent from 0.5 to 2, which stabilizes the characteristic function and emphasizes the <em> S </em> wave. We use the weight function by itself to determine if the signal&hyphen;to&hyphen;noise ratio is appropriate for picking and to determine a rough initial <em> S </em> pick. The pick is refined using the characteristic function. The picking algorithm works backward in time along the characteristic function from the initial pick until the characteristic function reaches a minimum. Using initial <em> S </em> picks based on the amplitude weight function, by increasing the weight function exponent and picking backward in time we avoid numerous incorrect early picks and drop traces with poor picking prospects. We apply the new approach to earthquakes in south&hyphen;central Tibet.</p>
Original languageAmerican English
JournalBulletin of the Seismological Society of America
Volume107
DOIs
StatePublished - Oct 1 2017

Disciplines

  • Earth Sciences

Cite this