IDENTIFICATION OF LOW-FREQUENCY EARTHQUAKES ON NON-VOLCANIC TREMOR VALIDATED WITH THE SUBSPACE DETECTOR METHOD

It has been proposed that non-volcanic tremor (NVT) episodes may be comprised of a sequence of identifiable, repeating low-frequency earthquakes (LFEs). In previous work, Shelly et al. identified 188 LFEs in a one-hour episode of tremor recorded at borehole stations belonging to the Japan Meteorological Agency (JMA). We seek to build upon this work by searching for additional LFEs within the data stream. Using Shelly's catalog of events, we build a set of 'master' event waveforms through a correlation and stacking method, which will be used in the subspace detection.

The subspace detection method takes a matrix of master event waveforms and, using SVD, builds a set of basis vectors that, in some linear combination, can reproduce the master waveforms. Operating on the assumption that any waveform coming from the same source region and having a similar source mechanism can likewise be characterized as some linear combination of our basis vectors, we explore the utility of the method to search for additional LFEs which fall within the subspace spanned by our basis vectors.