The Earth's crust is constantly undergoing the tectonic forces that tend to deform it. In the presence of excessive load conditions, rocks can break causing an earthquake, produced by the sudden movement of crustal masses, accompanied by the release of stress and nucleation of elastic waves. Dislocations occur along complex fractures called "faults". The classical seismological models used for the study of earthquakes assume that the rupture takes place over planes with friction under the action of a double-couple of forces orthogonal to each other and oriented to reproduce the wave radiation pattern recorded by seismic networks. Although restrictive, the assumptions of flatness and elastic rebound can explain the most important seismological observables and provide some first-order information about the seismic event, such as its magnitude and the kinematics of the fault. However, the same hypotheses and other simplifications are too rigid for an advanced description of seismicity, which, on the one hand, allow simplistic interpretations of seismological processes to thrive, and on the other, introduce bias in the estimates derived from the source parameters. The development of seismic networks and the tuning of new techniques and technologies used in the characterization of seismicity finally allow us to analyse its properties. In the study "The impact of faulting complexity and type on earthquake rupture dynamics" just published in Communications Earth & Environment, it is shown that the composition of the moment tensor, the matrix describing the average dynamics of the seismic event, changes according to the complexity of the fault that generated it, the geological setting in which it was nucleated, and several other reasons related to the seismic signal processing and analysis techniques. In real nature, during an earthquake, several faults are activated in the crustal destabilized volume. Moreover, they are often quite undulated, and the several branches can be anastomosed. Furthermore, in the shallow cold and brittle layer of the crust, rocks are disgregated into a breccia, but the thickness of this damage zone varies as a function of the tectonic style, being maximum in extensional tectonic settings (see the image below).
Earthquakes are different: the seismic signal in terms of the percentage of double couple in the composition of the tensor moment varies as a function of the structure of the rupture; the latter, in turn, is influenced by the fact that the crustal volumes tend to overlap, move away, or slip parallel to each other. In the first case, the dislocation occurs along a particularly thin and concentrated fracture, while in the other cases it tends to spread and to activate several fault planes.
The higher the double-couple, the greater the concentration of effort on a single and thin plane, while a slip on several planes results in a lower double-couple percentage. Therefore, the work demonstrates that on faults associated with compressive environments, i.e., thrusts, earthquakes have a double-couple larger than strike-slip and normal fault-related events.
Furthermore, it is highlighted that the estimate of the double-couple component is positively correlated to the magnitude of the events in a significant way only for the compressive events and negatively correlated to the scale exponent of the frequency-magnitude distribution of the seismic events (b-value), usually used in forecasting models. The research shows the existence of a dependence between the structure of the seismogenic source, statistical properties of seismicity and stress patterns within the lithosphere, a relationship that current seismological models, designed to study the dynamics of single earthquakes within a certain resolution, are unable to explain. The results of the research show that to understand the relationship between dynamics of single earthquakes, seismicity, and tectonics, it is necessary to make further theoretical and modelling efforts and to systematically continue the analysis of increasingly abundant and reliable data. Specifically, the role of the various energetic contributions to the mobilization of crustal volumes should be better clarified: while thrust-faulting and strike-slip events are mainly driven by elastic relaxation, normal-faulting ones are dominantly fuelled by gravitational energy; it is then desirable to analyse the impact of the complexity of the seismogenic source on the efficiency of the stress dissipation process in the brittle lithosphere and, consequently, on the spatial and temporal evolution of seismicity (as already discussed in Zaccagnino et al., 2022).
This will allow the development of an integrate framework for seismological dynamics and a more advanced and complete understanding of the processes affecting the fragile lithosphere. Furthermore, the study suggests a possible systematic and significant underestimation of the magnitude of earthquakes in the presence of low double-couple values not justified by physical processes. This hypothesis seems to be consistent with the most recent results in the field of 3D modelling of seismic wave propagation (see for example Sawade et al., 2022). Therefore, at least for large events characterized by high non-double-couple components (for instance, the Norcia earthquake 30/10/2016, Mw 6.5 and the Great Kaikoura earthquake, New Zealand, 13/11/2016, Mw 7.8) the need to estimate its magnitude with innovative techniques should be considered, especially due to the impact that an underestimation of the seismic moment can imply on the reliability of the procedures used in the context of seismic forecasting.
Sawade, L., Beller, S., Lei, W. & Tromp, J. Global centroid moment tensor solutions in a heterogeneous earth: the CMT3D catalogue. Geophys. J. Int., 231(3), 1727-1738 (2022).
Zaccagnino, D., Telesca, L., & Doglioni, C. Scaling properties of seismicity and faulting. Earth Planet. Sci. Lett., 584, 117511 (2022).