IMPROVED MULTIPLE EVENT LOCATION METHODS FOR GROUND TRUTH COLLECTION, Annual Report

The goal of this project is to develop a new methodology for seismic multiple-event locations which can improve the quality of ground truth event databases. The improvements sought include more complete and rigorous quantification of location uncertainty, better quality control on secondary phase picks associated with ground-truth events, and a greater quantity of events that meet specified ground-truth criteria. To achieve these improvements we are pursuing several enhancements to multiple-event location methodology applicable to earthquake clusters, such as the combined use of local, regional and teleseismic arrival-time data in a unified analysis, and application of a rigorous framework for uncertainty analysis that incorporates more realistic, and generally less restrictive, assumptions about travel-time model errors attributable to unknown velocity variations in the Earth. This report describes the mathematical framework and numerical algorithms we developed in the first year of the project and shows results of preliminary tests we performed with data from Nevada Test Site explosions.