Developing Multiple-Frequency Discriminants for Use with Regional Coda-Amplitude Measurements
Abstract
This paper investigates the feasibility of employing local to near-regional coda-wave amplitude measurements, in multiple, narrow bands, for the purpose of discriminating small seismic events. The motivation comes from previous studies that have shown that regional, single-station coda-magnitude estimates are more stable and accurate than any direct phase measure to date (e.g., Mayeda et al., 2003). Typically, the source amplitude estimates derived from the coda have interstation variances on the order of 0.07 log amplitude units; hence, the method is excellent for regions with sparse station coverage. The method has been thoroughly tested over large geographic regions spanning both local and regional distances for the purpose of magnitude estimation. In terms of discrimination, only a preliminary study using coda waves was performed on Nevada Test Site (NTS) explosions and earthquakes in the study by Walter et al., (1995). Since this time, however, no additional studies have been conducted, in spite of the coda's successful application for moment magnitude (MW) estimates in the broad areas of Eurasia and the Middle East. Due to the nature of the scattered energy comprising the coda, path and azimuthal source-radiation effects are averaged over, making coda amplitudes insensitive to local structure, in sharp contrast to direct regional phases such as Pn, Pg, and Lg. Calibrated with respect to seismic moment (M0) or MW, coda-derived source spectra provide a means to obtain moment estimates from seismograms of smaller or more distant events that cannot be analyzed with conventional waveform or spectral source-inversion techniques because of signal-to-noise limitations. Tying coda magnitude to Mw also provides a physical measure of event size that is unbiased and therefore transportable.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2007
- Accession Number
- ADA519380
Entities
People
- Arben Pitarka
- Kevin Mayeda
- Nancy R. Collins
Organizations
- University of California, Berkeley