Shemin Ge

Sharp increase in central Oklahoma seismicity since 2008 induced by massive wastewater injection

Wastewater disposal linked to earthquakes The number of earthquakes is increasing in regions with active unconventional oil and gas wells, where water pumped at high pressure breaks open rock containing natural gas, leaving behind wastewater in need of disposing. Keranen et al. show that the steep rise in earthquakes in Oklahoma, USA, is likely caused by fluid migration from wastewater disposal wells. Twenty percent of the earthquakes in the central United States could be attributed to just four of the wells. Injected fluids in high-volume wells triggered earthquakes over 30 km away. Science, this issue p. 448 The recent surge in central U.S. seismicity is likely attributable to injection of wastewater at a small number of wells. Unconventional oil and gas production provides a rapidly growing energy source; however, high-production states in the United States, such as Oklahoma, face sharply rising numbers of earthquakes. Subsurface pressure data required to unequivocally link earthquakes to wastewater injection are rarely accessible. Here we use seismicity and hydrogeological models to show that fluid migration from high-rate disposal wells in Oklahoma is potentially responsible for the largest swarm. Earthquake hypocenters occur within disposal formations and upper basement, between 2- and 5-kilometer depth. The modeled fluid pressure perturbation propagates throughout the same depth range and tracks earthquakes to distances of 35 kilometers, with a triggering threshold of ~0.07 megapascals. Although thousands of disposal wells operate aseismically, four of the highest-rate wells are capable of inducing 20% of 2008 to 2013 central U.S. seismicity.

High-rate injection is associated with the increase in U.S. mid-continent seismicity

Making quakes depends on injection rates Wastewater injection wells induce earthquakes that garner much attention, especially in tectonically inactive regions. Weingarten et al. combined information from public injection-well databases from the eastern and central United States with the best earthquake catalog available over the past 30 years. The rate of fluid injection into a well appeared to be the most likely decisive triggering factor in regions prone to induced earthquakes. Along these lines, Walsh III and Zoback found a clear correlation between areas in Oklahoma where waste saltwater is being injected on a large scale and areas experiencing increased earthquake activity. Science, this issue p. 1336; Sci. Adv. 10.1126/sciadv.1500195 (2015). High injection rates of wastewater into deep wells increase the risk of earthquakes in regions prone to induced seismicity. An unprecedented increase in earthquakes in the U.S. mid-continent began in 2009. Many of these earthquakes have been documented as induced by wastewater injection. We examine the relationship between wastewater injection and U.S. mid-continent seismicity using a newly assembled injection well database for the central and eastern United States. We find that the entire increase in earthquake rate is associated with fluid injection wells. High-rate injection wells (>300,000 barrels per month) are much more likely to be associated with earthquakes than lower-rate wells. At the scale of our study, a well’s cumulative injected volume, monthly wellhead pressure, depth, and proximity to crystalline basement do not strongly correlate with earthquake association. Managing injection rates may be a useful tool to minimize the likelihood of induced earthquakes.

Basin‐scale hydrogeologic modeling

Mathematical modeling of coupled groundwater flow, heat transfer, and chemical mass transport at the sedimentary basin scale has been increasingly used by Earth scientists studying a wide range of geologic processes including the formation of excess pore pressures, infiltration-driven metamorphism, heat flow anomalies, nuclear waste isolation, hydrothermal ore genesis, sediment diagenesis, basin tectonics, and petroleum generation and migration. These models have provided important insights into the rates and pathways of groundwater migration through basins, the relative importance of different driving mechanisms for fluid flow, and the nature of coupling between the hydraulic, thermal, chemical, and stress regimes. The mathematical descriptions of basin transport processes, the analytical and numerical solution methods employed, and the application of modeling to sedimentary basins around the world are the subject of this review paper. The special considerations made to represent coupled transport processes at the basin scale are emphasized. Future modeling efforts will probably utilize three-dimensional descriptions of transport processes, incorporate greater information regarding natural geological heterogeneity, further explore coupled processes, and involve greater field applications.

A governing equation for fluid flow in rough fractures

Fluid flow in a rock fracture bounded by two irregular surfaces is complex even under a laminar flow regime. The major factor causing deviation of predicted fracture flow behavior from the ideal parallel plate theory is the nature of nonparallel and nonsmooth geometry of fracture surfaces. Important questions on the validity of the cubic law and the Reynolds equation for complicated fracture geometries have been studied by many researchers. The general conclusion from these efforts is that the cubic law is valid provided that an appropriate average aperture can be defined. Many average apertures have been proposed, and for some cases, some work better than others. Nonetheless, to date, these efforts have not converged to form a unified definition on the fracture aperture needed in the cubic law, which stimulates the current effort to develop a general governing equation for fracture flow from a fundamental consideration. In this study, a governing equation stemming from the principle of mass conservation and the assumption that the cubic law holds locally is derived for incompressible laminar fluid flow in irregular fractures under steady state conditions. The equation is formulated in both local and global coordinates and explicitly incorporates two vectorial variables of fracture geometry: true aperture and tortuosity. Under the assumption of small variations in both tortuosity and aperture, the governing equation can be reduced to the Reynolds equation. Two examples are provided to show the importance and generality of the new governing equation in both local and global coordinate systems. In a simple fracture with two nonsmooth and nonparallel surfaces, the error in permeability estimation can be induced using the Reynolds equation with the apparent aperture and can reach 10% for a 25° inclination between the fracture surfaces. In a fracture with sinusoidal surfaces, the traditional method can cause significant errors in both permeability and pressure calculation.

Hydromechanical modeling of tectonically driven groundwater flow with application to the Arkoma Foreland Basin

Deep groundwater flow can be driven by several mechanisms in sedimentary basins. In the case of evolving foreland basins, large-scale compression and thrusting could develop abnormally high pressures in the foreland sag that would initiate transient fluid flow. The so-called tectonic “squeegee” effect is thought to have caused basin-wide migration of ore-forming brines and hydrocarbons (Oliver, 1986). Two-dimensional numerical models are developed here to quantify the role of compressional tectonics in driving regional fluid flow in the later stages of thrusting in a foreland basin. Poroelasticity theory coupled with regional groundwater flow form the basic elements of the mathematical model. We use the mathematical model to predict deformation and pressure dissipation in the unfaulted and nonfolded part of a foreland basin in front of a thrust belt as it is subjected to an instantaneous loading event. Sets of numerical experiments show that overpressure zones develop along the leading edge of the thrust belt near the loading front. Stress-induced flow rates of the order of centimeters to meters per year are possible soon after compression of the foreland, and transient flow fields dissipate in about 103 and 104 years. Longer transients can exist in very low permeability strata. Large overpressures may be unable to buildup under conditions of gradual thrusting, as fluid pressures may dissipate too quickly. The general features of tectonically driven flow are also explored through a sensitivity study to consider effects of permeability, fault and stratigraphic heterogeneity, loading magnitude, and variations in rock compressibility. The sensitivity study is based mostly on numerical experiments. As these solutions suffered from stability problems in cases where bulk rock compressibility exceeded 10−9 Pa−1, some simple scaling arguments are used to extend the numerical results for squeezing of soft shale. One basin-specific application to the Ouachita orogen suggests that tectonic squeezing could have caused transient flow systems with relatively large flow velocities in basal Cambro-Ordovician aquifers. The volume of fluid expelled, however, is probably only a small fraction of the total brine volume needed to have formed the huge Mississippi Valley-type lead-zinc ore deposits fringing the northern margin of the Arkoma Basin on the Ozark Uplift.

Coping with earthquakes induced by fluid injection

Hazard may be reduced by managing injection activities Large areas of the United States long considered geologically stable with little or no detected seismicity have recently become seismically active. The increase in earthquake activity began in the mid-continent starting in 2001 (1) and has continued to rise. In 2014, the rate of occurrence of earthquakes with magnitudes (M) of 3 and greater in Oklahoma exceeded that in California (see the figure). This elevated activity includes larger earthquakes, several with M > 5, that have caused significant damage (2, 3). To a large extent, the increasing rate of earthquakes in the mid-continent is due to fluid-injection activities used in modern energy production (1, 4, 5). We explore potential avenues for mitigating effects of induced seismicity. Although the United States is our focus here, Canada, China, the UK, and others confront similar problems associated with oil and gas production, whereas quakes induced by geothermal activities affect Switzerland, Germany, and others.

Effect of Horizontal Heat and Fluid Flow on the Vertical Temperature Distribution in a Semiconfining Layer

By including the constant flow of heat and fluid in the horizontal direction, we develop an analytical solution for the vertical temperature distribution within the semiconfining layer of a typical aquifer system. The solution is an extension of the previous one-dimensional theory by Bredehoeft and Papadopulos [1965]. It provides a quantitative tool for analyzing the uncertainty of the horizontal heat and fluid flow. The analytical results demonstrate that horizontal flow of heat and fluid, if at values much smaller than those of the vertical, has a negligible effect on the vertical temperature distribution but becomes significant when it is comparable to the vertical.

Offshore fresh groundwater reserves as a global phenomenon

The flow of terrestrial groundwater to the sea is an important natural component of the hydrological cycle. This process, however, does not explain the large volumes of low-salinity groundwater that are found below continental shelves. There is mounting evidence for the global occurrence of offshore fresh and brackish groundwater reserves. The potential use of these non-renewable reserves as a freshwater resource provides a clear incentive for future research. But the scope for continental shelf hydrogeology is broader and we envisage that it can contribute to the advancement of other scientific disciplines, in particular sedimentology and marine geochemistry.

Hydrodynamic response to strike‐ and dip‐slip faulting in a half‐space

Field observations have shown strong coupling between earthquake-induced stress-strain fields and subsurface hydrodynamics, reflected by water level change in wells and stream flow fluctuations. Various models have been used in an attempt to interpret the coseismic fluctuations in groundwater level, predict water table rise in the event of an earthquake, and explain stream flow variations. However, a general model integrating earthquake-induced stress-strain fields, coseismic pore pressure generation, and postseismic pore pressure diffusion is still lacking. This paper presents such a general framework with which one can approach the general problem of postseismic pore pressure diflusion in three dimensions. We first use an earthquake strain model to generate the stress-strain field. We then discuss the linkage coupling stress and strain with pore pressure and present an analytical solution of time-dependent pore pressure diffusion. Finally, we use two examples, a strike-slip and a dip-slip fault, to demonstrate the application of the analytical model and the effects of earthquakes on fluid flow. The application to the two fault systems shows that the diffusion time is shorter than conventional estimates, which are based on a diffusivity and a length scale. We find that the diffusion time is predominately a function of the diffusivity of the system, while the length scale influences the magnitude of the initial pore pressure. A diffusion time based on the diffusivity and a length may be misleading because significant localized flow occurs in complex three-dimensional systems. Furthermore, the induced patterns of a pore pressure change resemble the strain field when shear stress effects are neglected but are significantly modified when shear stresses are included in the coupling relation. The theoretical basis of this work is developed assuming a single episode dislocation. However, the methodology and the results can be readily applied to studying pore pressure conditions after multifaulting events by simple superposition.

A theoretical model for thrust‐induced deep groundwater expulsion with application to the Canadian Rocky Mountains

This paper presents a numerical model for simulating deformation and induced fluid flow in fold-and-thrust belts. Unfractured rock strata are modeled as poroelastic media while fault zones are treated as plastic-elastic media. We introduce the slip element technique into a finite element code to accommodate large deformations along faults. Thrust displacements, stress field changes, and the effects of thrust faulting on groundwater flow are investigated by solving the coupled stress and flow equations numerically. The calculation shows that when a thrust sheet is displaced along its fault surface, the displacement-induced stress generates high pore pressure zones near the tectonic stress boundary and beneath low permeability ramps. These overpressures cause transient fluid flow across the thrust belt. Sensitivity studies on the hydrologic properties of the fault zone suggest that hydraulic conductivities within a fault play important roles in initiating slip deformation and in determining the extent of transient disturbances to the flow field. Low permeability can result in rapid pore pressure buildup in the fault, thereby reducing the effective strength of the fault which leads to earlier failures. A low-permeability fault can also impede the movement of flow into the footwall, thereby limiting the tectonic impact on the flow system within the hanging-wall. Application of the model to the McConnell Thrust in the Canadian Rockies indicates that the total volume of fluid flow induced by tectonic compression could have been of the order of 105 to 106 m3 over a time period of tens to hundreds of years accompanied by an average 100 m of thrust movement.