Three Component Micro and Small Aperture Arrays and Local Networks in NoisyEnvironment. Development of Statistical Data Processing Techniques: Coherent Noise Suppression, Signal Azimuth and Apparent Velocity Estimation, Generic Automated Event Source Loca

摘要

Automatic discrimination of weak explosions and earthquakes at local distances isan important problem for CTBT monitoring in regions where commercial mining operations and quarry blasting generate a large number of seismic recordings on a daily basis. Development of reliable discrimination techniques to solve this problem involve selection of regionally dependent parameters (which have to be automatically extracted from seismograms) and estimation of misclassification probabilities intrinsic for a given region. Discrimination between weak regional or local earthquakes and explosions remains a serious challenge for state-of-the-art CTBT monitoring. In particular, the known fact that discrimination features proved to be efficient in one seismic region are often useless in another region is a serious obstacle for refinement and standardization of source discrimination techniques. Another important unsolved task in seismic discrimination is to estimate the probability of event misclassifications inherent to the seismic region under investigation.