Possible Indicator of a Strong Isotropic Earthquake Component: Example of Two Shallow Earthquakes in Greece
Abstract
For routine practice, we need simple tools to reliably identify earthquakes with large isotropic (ISO) components. This study aims to highlight a possible indicator.
Non-double-couple (non-DC) components of moment tensors (MTs) play a key role in our understanding of faulting earthquake processes and/or in identifying explosions. As opposed to DC components of the calculated seismic source model, the non-DC components (compensated linear vector dipole and ISO) are more vulnerable to errors in location, inaccurate velocity modeling, and noise.
Methods for analyzing resolvability of ISO are relatively complicated. We propose a simple procedure to identify an earthquake with a strong ISO component.
Recent MT determinations include space and time grid search of the centroid position, mainly the depth and time. The centroid is identified with a trial source position that maximizes correlation between real and synthetic waveforms.
In synthetic tests with varying ISO percentage, we compare the correlation-depth dependence for two types of MT inversion: full and deviatoric. We show that in the inversion of data with a significant ISO component under the deviatoric assumption (i.e., when ISO is neglected), we might obtain an inaccurate centroid depth.
However, when we make the grid search twice, under the deviatoric-MT and full-MT assumptions, and compare the results, we can obtain an indication of the significant ISO and avoid depth bias. This straightforward method is applied to two shallow earthquakes in Greece (the 27 January 2012 M-w 5.3 Cretan Sea earthquake and the 26 June 2009 M-w 4.9 Santorini earthquake).