Applying Waveform Correlation and Waveform Template Metadata to Mining Blasts to Reduce Analyst Workload

Abstract

Organizations that monitor for underground nuclear explosive tests are interested in techniques that automatically characterize mining blasts to reduce the human analyst effort required to produce high - quality event bulletins. Waveform correlation is effective in finding similar waveforms from repeating seismic events , i ncluding mining blasts . In this study we use waveform template event metadata to seek corroborating detections from multiple stations in the International Monitoring System of the Preparatory Commission for the Comprehensive Nuclear - Test - Ban Treaty Organi zation . We build upon events detected in a prior waveform correlation study of mining blasts in two geographic regions, Wyoming and Scandinavia . Using a set of expert analyst - reviewed waveform correlation event s that were declared to be true positive det ections, we explore criteria for choosing the waveform correlation detections that are most likely to lead to bulletin - worthy events and reduction of analyst effort.