Jürgen E. Streit

Fluid pressures at hypocenters of moderate to large earthquakes

Many active faults are expected to develop fluid pressures in excess of hydrostatic pressures below 3 to 7 km depth during interseismic periods. Suprahydrostatic fluid pressures are known to reduce the stresses required for brittle failure. Stress differences that trigger moderate to large earthquakes typically range from 40 to 160 MPa, as indicated by earthquake shear stress drops and paleostress estimates obtained from mylonites. For stresses in this range (40–160 MPa), seismic fault reactivation requires in most cases pore fluid factors (fluid pressure/overburden pressure) higher than for hydrostatic fluid pressures (i.e., >0.37) along misoriented faults with fault angles ≥45°, such as many segments of the San Andreas fault system. For example, at a stress difference of 100 MPa and fault angles of 45°–55°, fault reactivation at 7 to 20 km depth requires hypocentral pore fluid factors of ≈0.8–1 for reverse faults, 0.6–0.9 for strike-slip faults, and <0.8 for normal faults. Pore fluid factors increase with increasing cohesive strength of faults, increasing coefficient of internal friction, and increasing fault angle. Seismic reactivation of cohesive faults at stress differences of 40–160 MPa requires near the base of the seismogenic zone (≈15 km depth) suprahydrostatic fluid pressures at all possible fault angles. Pore fluid factors are ≈0.4–0.9 for normal faults, ≈0.6–1 for strike-slip faults, and ≈0.8–1.05 for thrust or reverse faults. These constraints are potentially useful for the modelling of seismic faulting and earthquake recurrence times.

Mechanics and mechanisms of magmatic underplating: inferences from mafic veins in deep crustal mylonite

Dioritic to gabbro‐dioritic veins with extreme length to width ratios (>1000 : 1) are localized along an amphibolite facies shear zone (the CMB Line) between exposed segments of originally middle and lower continental crust (Strona‐ Ceneri and Ivrea‐Verbano Zones, northern Italy). The geometry of these veins and their mutual cross-cutting relationships with the mylonitic foliation indicate that veining was coeval with noncoaxial flattening of the lower crust in Early Permian time. The veins formed as closely spaced extensional shear fractures and propagated parallel to the originally gently to moderately dipping (30o) mylonitic foliation. Vein opening at high angles (60‐90o) to the inferred1 direction and subparallel to the pre-existing planar fabric requires that melt pressure slightly exceeded the lithostatic pressure and that differential stress was low (10‐20 MPa) in the vicinity of the veins. The interaction of regions of tensile stress concentration at vein tips caused the concordant veins to curve and link up across the mylonitic foliation. Once interconnected, the veins served as conduits for the rapid movement of mafic melt along the shear zone. Thermal modelling constrains the mafic melt in the narrowest, 1 mm wide veins to have crystallized almost instantaneously. Such veins extend no more than a meter from host veins into the country rock, indicating that the minimum rate of vein tip propagation and melt flow was at least several m=s. Maximum crystallization times of only hundreds to thousands of years for even the thickest mafic veins (10‐100 m) in the IVZ are short compared to the 15‐20 Ma duration of Early Permian crustal attenuation and magmatism in the southern Alps. This suggests that veining in the lower crust occurred episodically during extended periods of mylonitic creep. Concordant vein networks within deep crustal shear zones that are inclined (as the CMB Line may have been) can also channel overpressurized mafic melt from deeper sources, e.g. lower crustal magma chambers, into cooler, intermediate crustal rock. This locally widens the depth interval of combined viscous and brittle deformation within the crust and can trigger partial melting of the country rock.

Low frictional strength of upper crustal faults: A model

Low frictional strength of faults is estimated based on existing models for earthquake recurrence, which account for compaction and hydrothermal lithification of fault rocks, as well as for the buildup of high fluid pressures in faults during interseismic periods. In areas of elevated heat flow and for a geothermal gradient of 30°C/km, fault compaction, lithification, and changes in fluid pressure gradient are inferred to occur at shallow crustal depth (3 to 7 km). Calculated shearing resistance of faults at seismogenic depths is compatible with shear stress drops ascribed to moderate and large earthquakes. The high frequency of microearthquakes, as well as the occurrence of some larger mainshock hypocenters at shallow crustal depth along several strands of the San Andreas Fault system are consistent with a predicted minimum of crustal strength at the top of a high fluid pressure regime. On the basis of field evidence from many exhumed faults for lithification of fault gouge and for faulting at near-lithostatic fluid pressures, a modified crustal strength profile is suggested to represent the strength of many active faults which involve hydrothermal fluid flow.

Building geomechanical models for the safe underground storage of carbon dioxide in porous rock

Publisher Summary Geomechanical modeling is an integral part of Australias GEODISC research program to ensure the safe storage of carbon dioxide in subsurface reservoirs. Storage of large volumes of CO2 is envisaged by injecting supercritical and thus pressurized CO2 into deep saline formations or depleted hydrocarbon reservoirs. Geomechanical research is required to ensure that injection-related fluid pressure increases do not cause fracturing or faulting. Induced brittle deformation could create fracture permeability that provides conduits for the escape of CO2-rich fluid from a reservoir or a saline formation. Thus, geomechanical modeling is conducted to determine maximum sustainable fluid pressures that will not induce brittle failure. Geomechanical modeling requires the determination of in situ stresses and knowledge of ambient pore fluid pressures in a potential storage site. The orientation of pre-existing faults is obtained from the interpretation of 2D and 3D seismic data. Brittle failure criteria are applied to estimate the slip tendency of faults and to estimate maximum fluid pressures that can be attained during CO2 injection. Some faults may be assessed to be unfavorably oriented for reactivation because fracturing of intact reservoir rock or seal is more likely to occur. In depleted oil and gas fields, geomechanical modeling needs to incorporate changes of the preproduction stresses that were induced by hydrocarbon production and associated pore pressure depletion. Such induced stress changes influence the maximum sustainable formation pressures and the CO2 storage volume. Hence, geomechanical modeling is applied to provide crucial information for the safe engineering of subsurface CO2 injection and the modeling of storage capacity.

Estimating rates of potential co2 loss from geological storage sites for risk and uncertainty analysis

Publisher Summary This chapter focuses on the rates of fluid flow from a range of scenarios to assist in the risk analysis for CO2 storage. The empirical data may provide guidance in estimating potential leakage rates without requiring detailed assumptions on the geometry and dimension of the flow path. Achievability of such containment can be assessed in risk and uncertainty analyses that consider the occurrence of unlikely events inducing leakage. Envisaged leakage mechanisms are flow driven by buoyancy and overpressure through permeable zones in the top seal of storage reservoirs, along permeable faults and damaged well bores. Seepage rates from natural CO2 accumulations with non-optimum seals, flow rates from natural gas storage, and fluid discharge rates related to earthquakes indicates potential leakage rates for risk and uncertainty analysis. Careful site selection that reduces leakage risk together with the planning for early leak detection and remediation can, in a risk and uncertainly analysis, lead to a high likelihood of project success.

Conditions for earthquake surface rupture along the San Andreas Fault System, California

Earthquake ruptures that nucleate along faults at depth can only propagate to the Earths surface if the shear stresses during faulting exceed the yield strength of the uppermost crust, where fluid pressures are usually hydrostatic. Suprahydrostatic fluid pressures may be required at depth for the seismic reactivation of misoriented faults, such as the San Andreas fault system. Based on fracture criteria for the wall and fault rock, slip events that can cause ground breakage are estimated to occur at seismogenic depth along fault planes that form angles of less than 65° on average with the maximum principal stress direction. Such slip events require minimum shear stresses of about 30 MPa at hypocentral depth. For faults with long interseismic periods, and thus, inferred high cohesive strength, the predicted reactivation angle is ≤55°, suggesting that several segments of the San Andreas fault system in southern California, including the San Bernardino area, parts of the Elsinore fault zone and the San Jacinto fault, provide the most probable hypocentral sites for future large earthquakes that disrupt the surface. To provide further constraints on locations for such earthquakes, we urgently need to investigate frictional properties of fault rocks and the state of stress at depth by employing laboratory experiments, focal mechanism studies, and by drilling into seismically active faults such as the San Andreas fault system.

Conducting comprehensive analyses of potential sites for geological CO2 storage

Publisher Summary This chapter illustrates that geological storage of CO2 may provide a solution to the problem of reducing anthropogenic emissions of greenhouse gases to the atmosphere. To accurately appraise a potential site in terms of its suitability for CO2 storage, a comprehensive workflow analyzing the detailed geological and geophysical characteristics of the site needs to be undertaken. In particular, potential sites need to be evaluated geologically in terms of their injectivity, containment, and capacity. Injectivity can be assessed by a review of the reservoir quality and by a detailed sequence stratigraphic model to estimate the likely flow-unit geometries and connectivity. Containment can be assessed via an analysis of the seal capacity (using MICP), and by assessing the possible migration pathways and trapping mechanisms from the structural geometry and stratigraphic architecture. In addition, integration of hydrodynamic analysis of formation water flow systems and geomechanical studies of fault stability and sustainable pore fluid pressures with the stratigraphic interpretation provide vital confirmation of the containment potential.

Predicting and Monitoring Geomechanical Effects of CO 2 Injection

Predicting and monitoring the geomechanical effects of underground CO 2 injection on stresses and seal integrity of the storage formation are crucial aspects of geological CO 2 storage. An increase in formation fluid pressure in a storage formation due to CO 2 injection decreases the effective stress in the rock. Low effective stresses can lead to fault reactivation or rock failure which could possibly be associated with seal breaching and unwanted CO 2 migration. To avoid seal breaching, the geomechanical stability of faults, reservoir rock, and top seal in potential CO 2 storage sites needs to be assessed. This requires the determination of in situ stresses, fault geometries, and frictional strengths of reservoir and seal rock. Fault stability and maximum sustainable pore fluid pressures can be estimated using methods, such as failure plots, the Fault Analysis Seal Technology (FAST) technique, or TrapTester software. In pressure-depleted reservoirs, in situ stresses and seal integrity need to be determined after depletion to estimate maximum sustainable pore fluid pressures. The detection of micro-seismic events arising from injection-induced shear failure of faults, fractures and intact rock is possible with geophone and accelerometer installations and can be used for real-time adjustment of injection pressures. In the event of injected CO 2 opening and infiltrating extensive fracture networks, this can possibly be detected using multi-component seismic methods and shear-wave splitting analysis.

Predicting, monitoring and controlling geomechanical effects of CO2 injection

Publisher Summary This chapter discusses the effects of pore-fluid pressure change on effective stresses in porous reservoir rock. While there is substantial information on pore pressure/stress coupling during pressure depletion in hydrocarbon fields, little is known about such effects during fluid injection. Thus, in cases where CO2 storage in severely depleted reservoirs is envisaged, geomechanical methods that predict depletion-induced faulting, such as by Streit and Hillis, can be useful for estimating whether fault-seal damage was induced. In cases where Biots α can be determined for reservoir rocks in CO2-storage sites, velocity-effective stress relationships should be established to also measure effective-stress changes with seismic techniques in addition to direct fluid-pressure measurement in monitoring wells. To test for injection-related changes in total horizontal stresses in CO2 storage or demonstration projects, extended leak-off tests and hydraulic-fracturing tests can be conducted. In storage sites with an identifiable risk of injection-related fault reactivation to produce events of noticeable magnitude, geophones, and transducers should be installed for the monitoring of induced MS events. The monitoring can be used to detect accidental over-pressurization of the formation as it is likely to allow for real-time adjustment of injection pressures.