High Frequency 3-Component Waveform Inversion for Source and Structural Parameters

摘要

We present the results of three studies to develop and verify techniques to classify weak seismic events. (1) The method and results of full waveform inversion for both detailed source parameters and structure parameters are described. Input data were seismograms from industrial explosions in Eastern Kazakhstan recorded by the NRDC seismic network in 1987. Very good fits were produced between the synthetic seismograms and the observed data on all three components simultaneously and for P-wave, Rayleigh wave, and Love wave. We interpreted some of the inverted source parameters as characteristic of several different types of industrial surface mining operations. (2) The same technique was used to determine detailed source and structure parameters using an event that is highly relevant to nuclear monitoring. We determined that a salt mine collapse near Solikamsk, the Ural Mountains on 5 January 1995 was most likely a mine collapse instead of an underground explosions. (3) This study was carried out jointly by the Seismology Group of the University of Colorado and the Russian team from the Int'l Institute of Earthquake Prediction Theory and Math. Geophysics. We developed a new technique to identify a seismic event based on simultaneous inversion of surface wave amplitude spectra and signs of first motions of body wave. We applied this technique to several events near the Chinese test site at Lop Nor and demonstrated significant differences in source parameters characterizing explosions and natural earthquakes in this region.