Quick Fault-Plane Identification by a Geometrical Method: Application to the MW 6.2 Leonidio Earthquake, 6 January 2008, Greece

Abstract

Focal mechanisms of earthquakes provide two nodal planes. A foolproof method of identifying which one is the fault plane is the “seismologist's dream.” This is true because knowledge of causative faults is of key importance for seismotectonic studies. For example, intermediate-depth earthquakes, such as the one studied in this paper, rarely have known fault planes, but the correct interpretation of such a fault plane would help constrain regional geodynamic models of subducted plates and stress fields. It is equally important to identify active crustal blind faults, knowledge of which may improve earthquake hazard assessment. The fault plane can sometimes be well “mapped” (constrained) by the spatial distribution of numerous early aftershocks. However, this technique has serious limitations. One of them is the fact that in sparsely instrumented regions, accurate location of weak aftershocks is impossible. Moreover, some events lack numerous aftershocks at the mainshock fault plane altogether (this is typical of intermediate-depth earthquakes). The apparently straightforward case, where an earthquake occurs at or close to a geologically well-known fault, also benefits from an independent check, because the “known” fault may have a complex tectonic structure at depth. The most challenging task is the quick identification of the earthquake fault plane. If made in near real time, it might play a vital role in the fast simulation of strong ground motions (shake maps) for post-event emergency services. If “quick'” means a few hours or a few days after the event, the identification might still greatly contribute to assessing increased spatial probabilities of aftershocks based on the Coulomb stress-loading of neighboring faults due to the mainshock rupture (McCloskey et al. 2005). Existing methods to identify the fault plane from seismograms are based mainly on finite-extent source models: distributed-slip models are generated for both nodal planes and the one that better fits …