Data processing techniques for the detection and interpretation of teleseismic signals

Abstract

This paper is a collection of six papers describing recent developments in automated detection and identification of teleseismic earthquakes and explosions in a seismic noise background. The first paper evaluates the assumption that the outputs of seismometer arrays can be added since the signals will reinforce while the noise is cancelled. Signal and noise correlations vs. distance and frequency are presented for an array of 1600 km in extent. The second paper describes a method utilizing orthogonal expansions of the Kautz type in an effort to determine spectral and temporal differences between both types of signals and noise. Theory and measurements indicate that the seismic noise background is largely composed of fundamental and higher mode Rayleigh waves. The third paper describes a thermal equilibrium analogy to estimate the noise energies in each mode to account for the observed depth and frequency behavior. The use of multiple and partial coherence functions for resolving noise backgrounds into their propagation components is described in the fourth paper. Compressional, shear, and surface wave components of signals can be separated from seismic noise backgrounds by recognizing their differing polarization properties, as shown in the fifth paper. The source mechanisms can theoretically be identified from their radiation patterns provided instrument and travel path distortions are removed. A final paper describes this theory and how these various methods of detecting and isolating the signals can be integrated into an automated signal analysis system.