A STATISTICAL TECHNIQUE FOR VALIDATING VELOCITY MODELS

This study investigates the use of a station influence statistic to id entify velocity model shortcomings in the earthquake hypocenter locati on problem. Two groups of microearthquake events are examined. The fir st is a group of 81 events from the Mount St. Helens region that occur red between November 1987 and September 1991; the second, 110 well-loc ated events from the 1992 Joshua Tree aftershock sequence. We describe a method for validating a postulated earth model. Let lambda denote t he hypocenter estimates that Geiger's method obtains. Systematically r emove each station observation from the location problem, and recomput e the location estimate. Call this estimate lambda((i)) when the ith s tation is removed. For a single event, define a station's influence (S I) as a weighted difference between lambda and lambda((i)). Distributi onal summaries of SI statistics across events are used to identify mod el shortcomings: Given a specified velocity model, SI distributions th at are not homogeneous across stations provide evidence of model inade quacies and/or failures in the weighting scheme. We show that velocity model shortcomings detected using SI statistics for the Mount St. Hel ens sequence under a one-dimensional model appear to correlate with kn own physical anomalies; while SI distributions evaluated under a three -dimensional model are more homogeneous and reflect a modest improveme nt over the one-dimensional model. SI distributions provide evidence o f model failure for the Joshua Tree sequence under a one-dimensional m odel, but no evidence of failure under a three-dimensional model. Fina lly, the weighting scheme's validity is verified for the Joshua Tree s equence, under the three-dimensional model.