Abstract The permeability structure resulting from high fluid pressure stimulation of a geothermal resource is the most important parameter controlling the feasibility and the viability of enhanced geothermal systems ( EGS ), yet is the most elusive to constrain. Linear diffusion models do a reasonably good job of constraining the front of the stimulated region because of the t 1/2 dependence of the perturbation length, but triggering pressures resulting from such models, and the permeability inferred using the diffusivity parameter, drastically underestimate both permeability and pressure changes. This leads to incorrect interpretations about the nature of the system, including the degree of fluid pressures needed to induce seismicity required to enhance the system. Here, I use a minimalist approach to modeling and show that all of the observations from Basel (Switzerland) fluid injection experiment are well matched by a simple model where the dominant control on the system is a large‐scale change in permeability at the onset of slip. The excellent agreement between observations and these simplest of models indicates that these systems may be less complicated than envisaged, thus offering strategies for more sophisticated future modeling to help constrain and exploit these systems.
<p>Additional Figures for the remaining Profiles. Hypocentral distribution of the 2009 L’Aquila sequence and the 2016 AVN sequence with a close-up view. And calculated fluid-pressure superposed to the recorded seismic data.</p>
Recently reported rain-triggered seismicity from three separate storms occurred exclusively in karst geology. In this paper, I discuss how the hydrogeology of karst controls rain-triggered seismicity by channeling of the watershed after intense rainfall directly into the karst network. Such channeling results in very large increases in hydraulic head, and more importantly, substantially increases the vertical stress acting on the underlying pore-elastic media. Rapid loading upon a pore-elastic media induces seismicity by increasing pore pressure at depth in a manner similar to that observed from reservoir impounding. Using a simple 1-D model of a pore-elastic medium, it is shown that the instantaneous fluid pressure increase at depth is a substantial fraction of the pressure step applied at the boundary, followed by time-dependent pore pressure increases associated with the typical linear diffusion problem. These results have implications for the change in fluid pressure necessary to trigger earthquakes, and leads to the following hypothesis to be tested: Unambiguous rain-triggered seismicity will only occur in karst regions.
Zebra carbonates are characterized by subparallel, rhythmic, mm-scale banding of host rock and vein. Their genesis has been interpreted by different authors as primary sedimentary structure, metasomatic infiltration or mechanical fragmentation followed by deposition of vein minerals. We studied zebra carbonates in the damage zones of normal faults formed during the drainage of an overpressure cell at about 7 km depth, in outcrops on Jebel Shams, Oman Mountains. They show a distinct pattern of mm-scale regularly spaced calcite veins in the dark grey, fine-grained carbonate host rocks, often connected laterally to a wide mode I fracture filled by a single calcite vein several cm thick. Veins in the zebra carbonates are filled with blocky crystals, indicating that the fractures remained open after their formation to allow crystal growth from a supersaturated fluid. Microstructures show no evidence of repeated crack-seal events and we conclude that all the veins in one zebra were formed simultaneously. The very high density of closely-spaced and simultaneously formed fractures indicates that they were formed very rapid loading, producing fracture densities much higher than expected during slow deformation. On the other hand, the highdensity fracture systems formed during explosive fracturing in dry rocks are much less regularly spaced. We hypothesize that the zebra carbonates were formed by rapid loading during faulting in highly overpressured carbonates, in places where coseismic rupture leads to a significant fluid pressure drop in dilatant jogs. The permeability of the matrix carbonate leads to drop in pore-fluid pressure close to the crack walls. This makes the host rock stronger close to the crack wall, so that the next fracture will preferentially propagate into the matrix away from the walls of the existing fracture. This process can lead to a more regularly spaced pattern of veins. Further work on zebra carbonates could provide a new tool to distinguish seismic from aseismic faults in carbonates.
