Content area
Abstract
This study is a contribution to the understanding of fracture processes in the upper crust. Theoretical, computational and observational approaches are combined to address various questions related to the mechanics of active faults and magma filled cracks.
The results of a theoretical study that aimed to investigate the conditions under which magmatic intrusions are able to-propagate along active faults are presented in Chapter 2. Addressing this question required analysis of the stress field ahead of the propagating intrusion. This is particularly difficult, because the stress field in that region is sensitive to non-elastic deformation and to the response of the pore fluid. In addition to improving the understanding of the mechanical aspects of dike intrusions into faults in the upper crust, the results of this study have implications on the ability to infer paleostresses from “frozen” dike orientations, and the interpretation of seismicity induced by some dikes and hydrofractures.
Several studies of stress transfer between mainshocks and aftershocks found that the correlation between stress change and seismicity rate change disappears below some small stress, typically about 0.1 bar. This result may be interpreted either as a detection limit, or as mechanical threshold for triggering. Because according to the available friction models (both Coulomb and the rate-and-state) there is no physical reason why lower threshold should exist, resolving this issue is important. In Chapter 3 we present the results of a study that aimed to address this question. In this study we reported that lower threshold for earthquake triggering in central California has not been found.
An inherently discrete model of an earthquake fault is presented in Chapter 4. We show that this model produces earthquake catalogs that display many features that are observed in natural seismicity, including foreshocks, aftershocks and close to power law earthquake size distribution. In Chapter 5 we show that this type of computer simulation is useful for the interpretation of aftershock sequences.
A comparison of seismicity rate response to a stress change along the Sargent fault, the Calaveras fault and a section of the San Andreas fault is presented in Chapter 5. This comparison reveals great variability in fault behavior. Comparison between observed and simulated spatio-temporal clustering suggests that these differences may be attributed to differences in fault constitutive properties.
