Identifying seismic fault geometry from focal mechanisms based on fault instability ratio during a stress inversion

Unveiling the geometry and kinematics of subsurface seismic faults is important for earthquake hazard assessment. Earthquake focal mechanisms can provide such fundamental aspects of faulting; however, they often require additional information to distinguish faults and auxiliary planes. We attempt to identify faults using a stress inversion technique, in which fault planes in individual focal mechanisms are selected based on the fault instability parameter. This stress inversion algorithm developed by Vavryčuk (2014) selects a higher instability nodal plane as the fault and finds 70-80% faults correctly to derive reliable stress results. Our tests using synthetic and simulated focal mechanism data with faults known beforehand show that faults are correctly identified especially when the instability of the selected fault plane is significantly higher than that of the auxiliary plane, which can be quantitatively expressed by the instability ratio of the fault plane to that of the auxiliary plane being higher than ~1.3. This constraint can improve further the ability to identify subsurface seismic faults. We apply this technique to the case of geothermal-induced earthquakes that occurred in Pohang, South Korea during 2016-2017. A total of 53 well-constrained and well-located focal mechanism data are inverted to derive a stress condition, during which faults are identified as higher instability nodal planes. These earthquakes occurred in spatially distinct portions of the region associated with water injection through two respective boreholes (PX-1 and PX-2). For the PX-2-related earthquakes, 70% of identified faults are well aligned in their locations and orientations with a large-scale fault, indicating that these earthquakes occurred on the patches of this fault. This fault is responsible for the 2017 Mw 5.5 main earthquake. Fault planes whose instability ratio is higher than ~1.3 are all consistent with this plane. There is more variation in identified fault orientations in PX-1 earthquakes. However, a few fault planes with high instability ratios are generally subparallel to one another. The locations and orientations of these high instability ratio planes are well aligned with a large-scale fault, which is different from, but subparallel to the PX-2 fault. This study demonstrates the possibility of identifying and imaging subsurface seismic faults only using faulting mechanics without other additional information.