Regional Phase Processors
Abstract
The purpose of this study is to evaluate the principle of the technique known as the Smart processor, developed at the SDAC (Smart, 1977) and to determine the propagation direction of regional seismic phases by operating on a comprehensive data set. The Smart single-station maximum-likelihood surface-wave processor and variants thereof, all relying on three-component particle-motion analysis, demonstrated in this study their utility for automatic signal azimuth determination. In this study the Smart processor was found to be that which simply maximized the ratio of energy between components. Thus the P wave azimuth is determined by minimizing the transverse component, the L(g) by maximizing the transverse and the emergence cycle by finding the best rectilinear motion fit to 3 components of data. For the entire test data set, the average azimuthal error was 6.7 degrees for the L(g) signals, and 7.0 degrees for the P arrivals. Combining the estimated L(g) azimuth with that of the P wave, simply by taking their mean, increased the accuracy. The average difference between the mean estimated azimuth and the true geodesic azimuth was 4.9 degrees (7.0 degrees rms). Moreover, the F-statistic computed by these processors serves to separate poor azimuthal estimates from the population: the azimuthal estimates (over 80% for this data set) which passed the arbitrary F threshold set for this study differed from the known geodesic azimuths by an average of 3.9 degrees (4.9 degrees rms). It is the author's experience that such accuracy is comparable to or better than that from a well-designed array of sensors.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 07, 1981
- Accession Number
- ADA105470
Entities
People
- Eugene Smart
- Helene Sproules