Ongoing seismogenic processes in the brittle Earth’s crust are substantially driven by different aspects of stress. Thus, characterizing the role of stress during nucleation and rupturing of earthquakes is a crucial factor for understanding the physics of earthquakes. In this thesis I investigate the role of stress fluctuations in seismogenic processes. In particular, I draw inferences from the concept of scale invariance and the analysis of seismicity, induced by the injection of pressurized fluids through boreholes. This hydraulic reservoir stimulation is among others performed in order to develop enhanced geothermal systems (EGS) for sustainable power generation. The investigation of fluid-induced earthquakes is of particular importance because the basic conditions during earthquake nucleation and rupturing at fluid injection sites can be better constrained than for earthquakes on tectonic scale. Observed scale invariance of the physics of earthquakes suggests transferability of results obtained at different scales.
I quantify the perturbation of stress caused by the injection of pressurized fluids. This is done under the assumption that pore pressure diffusion in the fluid saturated pore and fracture space of rocks is the governing triggering mechanism of fluid injection-induced earthquakes. Moreover, the importance of stress changes generated by the occurrence of fluid-induced earthquakes is evaluated by analyzing the waiting times between subsequent seismic events. I show that no signatures of aftershock triggering can be identified in six analyzed seismicity catalogs gathered at EGS sites. Based on this result I demonstrate that the Poisson model can be used to compute the occurrence probability of fluid injection-induced earthquakes. This statistical model is needed in order to assess and mitigate the seismic risk, which still acts as an obstacle for efficient and risk-free use of the geothermal potential of the subsurface for sustainable power generation. The finding that stress changes caused by preceding events are only of second order importance for the seismogenesis of fluid-induced earthquakes underlines the significance of studies assuming pore pressure diffusion to be the triggering mechanism of seismic events. Based on this assumption and the consideration of a nearly critical state of stress in the Earth’s crust, a physically based statistical model describing the seismicity rate of fluid-induced earthquakes during and after injection of fluids is presented. The investigation of seismicity occurring after termination of reservoir stimulation is of particular importance as the physical processes leading to the triggering of post-injection seismic events have not yet been fully understood. In addition, it has been observed that the strongest seismic events tend to occur close before or after the termination of reservoir stimulation. I show that the decay rate of seismicity after reservoir stimulation can be approximated by a modification of Omori’s law, describing the decay of aftershock activity succeeding tectonic main shocks. Moreover, I demonstrate that in the case of fluid injection-induced seismicity the power law exponent of Omori’s law depends on the criticality of stress in rocks.
Furthermore, I investigate the impact of elastic rock heterogeneity on the distribution of stress in the brittle Earth’s crust. The results provide fundamental insights into the nature of seismogenic processes. My findings suggest that the scale invariance of earthquakes originates from scale-invariant fluctuations of stress in rocks. These fluctuations occur naturally because of the universal fractal nature of elastic rock heterogeneity in the Earth’s crust. Scientific evidence for the universal fractal nature of elastic rock heterogeneity is given by measurements along boreholes at various drilling sites in different regions. As a consequence, fault planes and correspondingly magnitudes of earthquakes scale according to a universal power law. This explains the emergence of the Gutenberg-Richter relation characterized by a universal b-value of b = 1 and implies the scale invariance of the magnitude scaling of earthquakes. My findings suggest that the observed stress dependency of the two fundamental power laws of statistical seismology occurs due to characteristic scales of seismogenic processes. Each characteristic scale involved in a process causes a limitation or change of fractal scaling. Moreover, the heterogeneous nature of critical stress changes in rocks, observed in various studies, can be physically explained by the influence of elastic rock heterogeneity. I show that stress changes in the range of a few KPa to a few MPa are capable of triggering brittle failure and associated seismicity in rocks of the Earth’s crust. This result validates the concept of a nearly critical state of stress in the Earth’s crust and suggests that already stress changes just above perturbations caused by tidal forces (approx. 1000 Pa).
Falls Ihr Browser eine Datei nicht öffnen kann, die Datei zuerst herunterladen und dann öffnen.