LARGE APERTURE SEISMIC ARRAYS
Abstract
Four topics are discussed in the progress report. The first topic is related to the masking of underground nuclear tests with large earthquakes. Computer simulations, using signals from actual seismic events, suggest that test detonations one magnitude unit smaller than the earthquake can be detected, providing the shot time is some two minutes after the time of occurrence of the earthquake. The second topic deals with the effects of signal-to-noise ratio and local travel time-anomalies on the accuracy of epicenter location using large seismic arrays. Both theoretical analyses and computer simulations indicate standard deviations of angular errors on the order of 0.3 degrees for either a 10 dB signal-to-noise ratio or a peak anomaly on the order of 1/20th of a second. The third topic deals with the application of a previously developed coda-correlation discriminant to DIMUS (hardlimited) seismograms. This discriminant makes use of the average paired correlation coefficient of the 10- second portion of seismic arrivals commencing 3 seconds after P-wave onset for an array of widely separated stations. Results to date indicate a moderate degradation of the discriminant for the DIMUS seismograms compared to the unclipped seismograms. The fourth topic deals with the automatic identification of the pP phase of earthquakes, for events in the 40 to 150 km depth region. Excellent results have been obtained with this automatic scheme, and not only is the pP phase properly identified, but also the sP phase. This combination allows depths to be estimated with considerable confidence.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1967
- Accession Number
- AD0664571