Using Multi-Station Waveform Coherence to Improve Detection of Microseismicity

Abstract:

The smallest earthquakes go unmeasured except in unusual circumstances. These events are potentially of great importance for understanding and characterizing earthquake physics. For instance the GR relation at small magnitude remains an open question. Micro earthquakes are also potentially of great importance for applications ranging from earthquake precursors to probing the earth’s stress state. As these smallest of earthquakes are the most numerous, their spatial and temporal response to changes in the stress field or variations in the frictional properties of faults may tell us a great deal about the elastic properties of the crust. Here we describe an approach to detect these smallest events, based on waveform coherence.

We describe a method to detect microseismicity that can be applied without dense specialized arrays. Traditional earthquake detection algorithms require the identification of phase arrivals or use a finite number of templates to scan continuous data. Correlating waveforms or waveform envelopes from nearby stations has been used to detect non-traditional seismic emissions like non-volcanic tremor. We modified this method to search for microseismicity. We do not explicitly identify or locate individual events. Instead, we quantify how microseismicity varies in time and space. By applying this method to real data we demonstrate that in addition to detecting all cataloged events, we also detect many smaller events.