Correlation of microearthquake body-wave spectral amplitudes

For applications such as stress tensor inversion and the calculation of composite focal mechanisms, it is of great importance to be able to identify earthquakes with similar focal mechanisms. Inspired by the notion that earthquake focal mechanisms can be calculated by using spectral amplitudes and P-wave polarities, we propose a focal-mechanism assessment scheme based on the correlation of body-wave spectral amplitudes. The spectral amplitudes used are the low-frequency spectral amplitude asymptotes of the direct P, SV, and SH waves. We calculate correlation coefficients using all common spectral amplitudes for pairs of events. The correlation coefficients are used as the basis for grouping the events, groups that will contain closely located events with very similar focal mechanisms. We discuss three applications of the spectral amplitude correlation and grouping technique (SAG) by using microearthquake data from the South Iceland Lowland network in the Olfus area. First, using 26 very well-correlated events, we calculate a composite focal mechanism that has one nodal plane that agrees well with the events' common fault plane from high-accuracy relative relocation. Second, the SAG scheme is used to address the problem of assessing the redundancy of focal mechanisms for stress tensor inversion, enabling a more realistic calculation of the confidence limits and reducing inversion computing time. Finally, by correlating and grouping a sequence of 2943 events before a magnitude M-L 5.1 earthquake, we observe significant temporal changes in the grouping pattern before the large earthquake. This suggests a change from seismicity dominated by clusters of events with very similar focal mechanisms to a regime dominated by solitary, nonsimilar events.