APPLICATION OF PATTERN-RECOGNITION TECHNIQUES TO EARTHQUAKE CATALOGS GENERATED BY MODEL OF SEGMENTED FAULT SYSTEMS IN 3-DIMENSIONAL ELASTICSOLIDS

Techniques and seismicity parameters described by Eneva and Ben-Zion [ 1997] are used to examine synthetic earthquake catalogs generated by B en-Zion [1996] for precursory patterns of large model events. Differen t model realizations represent various levels of fault zone disorder. These include models with uniform properties (U), a Parkfield-type asp erity (A), fractal brittle properties (F), and multi-size-scale hetero geneities (M). The seismicity parameters used are based on information contained in typical earthquake catalogs reflecting earthquake distri bution in space, time, and size. The analysis highlights the complexit y of the information content of the synthetic earthquake catalogs. Sim ple repetitive precursory signals have not been found. However, local extrema in the examined parameters are found to have significant assoc iation in time with large events. Thus our techniques and parameters m ay be useful for intermediate-term earthquake prediction, especially w hen parameters are used in combinations. Some analysis results are the same for all model realizations and some depend on the model. Feature s characterizing all catalogs are as follows: (1) Large model events a re statistically predictable on the basis of patterns in the distribut ion of smaller events. (2) For a given parameter, the type of precurso ry extrema (maxima or minima)) is the same for all models. (3) The int erparameter correlation for any parameter pair has the same sign (posi tive or negative) in all models. (4) The large events are neither slip -nor time-predictable based on previous large events. Results that dif fer from model to model include the following: (1) The degree of predi ctability of large events correlates with the degree of regularity in the assumed fault properties, following the order U, F, A, and M. (2) There is no one-to-one correlation between type of precursory extrema (maxima or minima) and type of precursory trends (increase or decrease ); this produces great variations in observable trends for any given p arameter, both from model to model and for different events in the sam e model. (3) The interparameter correlations vary among models, with t he highest correlations in model F. Most discussed patterns are in agr eement with observations from seismically active zones, laboratory mod els, and mining-induced seismicity.