Disputable Non-Double-Couple Mechanisms of Several Strong Earthquakes: Second-Degree Moment Approach

Abstract

The earthquake source mechanism (description of the geometry of the source and its strength in terms of forces equivalent to rupturing of the rock mass) is routinely modeled as a moment tensor (MT) describing the earthquake focus as a point. However, sometimes this approximation is not satisfied, and seismic radiation keeps the directivity due to the finite extent of the source. Inversion into the MT then may yield a biased mechanism. Synthetic study in the Adamova and Silený (2010) demonstrated the appearance of spurious non‐double‐couple components in the mechanism even for a pure double‐couple (DC) source. Their method was designed to reduce the spurious source components by using the second‐degree moments to evaluate their contribution into the records and subtract it from the data. Here we applied the procedure to five moderate to large regional events with large nonshear components. They are mostly located on large tectonic faults where predominantly pure shear slip is expected. We studied one event on the prominent North Anatolian fault, three events in the Pacific area, and one event in Bolivia. In most cases, the non‐DC components essentially were reduced, and the geometry of the mechanism remained largely unchanged. This confirms the hypothesis that part of the non‐DC components in regional MT solutions may be spurious due to the neglect of the source finiteness in the routine procedure of the MT retrieval. In addition, the geometrical and kinematical characteristics of the foci provided by the second‐degree moments (the source ellipsoid and rupture velocity vector) are mostly consistent with the fault geometry, aftershock distribution, and estimates of rupture speed from available previous studies.