Combining Analyst and Waveform-Correlation-Based Arrival Time Measurements in the Bayesloc Multiple-Event Location Algorithm.

摘要

We add relative arrival-time measurements that are derived from waveform correlation to the Bayesloc multiple-event location algorithm. Bayesloc is a formulation of joint probability over event locations, travel time corrections phase labels, and arrival-time measurement errors. The Bayesloc formulation is hierarchical with distinct statistical models for each component of the multiple-event system, including prior constraints for any of the parameters. Bayes' Theorem allows calculation of the joint probability for hypothesized configurations of Bayesloc parameters which facilitates using the Markov-Chain Monte Carlo (MCMC) method to draw samples from the joint probability function. The marginal posteriori distribution for each parameter or covariance between parameters is inferred from MCMC samples. Correlation- based picks are integrated into the Bayesloc formulation by including a new category of arrival time measurement that is derived from correlation of empirical waveforms. Because relative picks are derived from correlation between two waveforms and absolute-time picks are made by analysis of a single waveform -- typically an analyst, error processes for relative and absolute arrival time measurements are independent. Relative pick precision is formulated as a function of correlation coefficient and the time-bandwidth product of the correlated waveforms, and absolute arrival times precision -- as described in previous work -- is formulated as a function phase type, the station, and the individual event.