| Earthquakes generated on major continental faults represent one of the deadliest and costliest natural disasters affecting our planet - but one of the hardest to predict. In order to improve seismic hazard evaluation in populated earthquake-prone regions, it is imperative that we understand the underlying physical and chemical processes that control the mechanics of recurrent, unstable slip on faults, i.e. the seismic cycle. Despite significant progress and numerous laboratory-based studies, several major discrepancies exist between fault mechanics theory and observation. In particular, the frictional strength of long-lived seismogenic fault zones appears to be much lower than predicted by models derived from laboratory experiments. However, experiments have not addressed the effects of chemical fluid-rock interactions, pore fluid pressure evolution or permeability anisotropy development over the full seismic cycle. Moreover, most have focused on simulated, homogeneous fault rocks, whereas fault zones often contain a strong internal fabric or foliation, which can strongly influence frictional strength. To address these key issues, I propose to study the evolution of frictional and hydraulic properties of natural and simulated fault rocks under realistic subsurface conditions and using the full range of velocities relevant to the seismic cycle, i.e. to the interseismic and coseimic stages. The results will find direct application in seismic hazard assessment and will elucidate not only the unstable slip behaviour of faults but also their healing/sealing behaviour during interseismic periods. The latter will be highly relevant to fault integrity prediction in oil- and gas exploration and in relation to geological sequestration of CO2 and the safe storage of radioactive waste. |
