7
Induced Changes to Fracture Systems

In this chapter a variety of processes associated with induced changes to fracture systems that involve fluid flow are examined. Fractures are sensitive to changes in temperature, pressure, and fluid chemistry. Indeed, slight perturbations can result in significant alterations in fracture properties. The fundamental mechanisms and the tools to measure or predict these changes are described. Many of the same changes that occur in engineering practice also occur in natural systems, as described in Chapter 2.

This topic is extremely complex, and an exhaustive treatment is well beyond the scope of this report. This chapter provides a brief overview of the types and causes of changes to fracture networks in engineering practice, with some explanation of the approaches used to deal with these changes. In this chapter, these changes are organized first according to the type of change and second according to its cause. Four different ways that fracture systems can be altered are discussed here (Figure 7.1). The first is deformation through changes in stresses in the rock mass. This deformation includes changes in the void geometry of fractures, which in turn changes the ability of fractures to conduct fluids. In the extreme, deformation leads to rock failure. Extensional or shear failure can lead to the creation of new fractures.

The second is modification of the fluids contained in fractures. Fluids with significantly different properties can flow into the fractures, or fluids in the fractures can undergo phase changes or alterations of the distribution of phases or components.

The third is the addition of solid material into fractures. This can occur through the introduction of grout or through fluid injections that transport solid materials.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications 7 Induced Changes to Fracture Systems In this chapter a variety of processes associated with induced changes to fracture systems that involve fluid flow are examined. Fractures are sensitive to changes in temperature, pressure, and fluid chemistry. Indeed, slight perturbations can result in significant alterations in fracture properties. The fundamental mechanisms and the tools to measure or predict these changes are described. Many of the same changes that occur in engineering practice also occur in natural systems, as described in Chapter 2. This topic is extremely complex, and an exhaustive treatment is well beyond the scope of this report. This chapter provides a brief overview of the types and causes of changes to fracture networks in engineering practice, with some explanation of the approaches used to deal with these changes. In this chapter, these changes are organized first according to the type of change and second according to its cause. Four different ways that fracture systems can be altered are discussed here (Figure 7.1). The first is deformation through changes in stresses in the rock mass. This deformation includes changes in the void geometry of fractures, which in turn changes the ability of fractures to conduct fluids. In the extreme, deformation leads to rock failure. Extensional or shear failure can lead to the creation of new fractures. The second is modification of the fluids contained in fractures. Fluids with significantly different properties can flow into the fractures, or fluids in the fractures can undergo phase changes or alterations of the distribution of phases or components. The third is the addition of solid material into fractures. This can occur through the introduction of grout or through fluid injections that transport solid materials.

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications FIGURE 7.1 Changes to fracture systems that change hydraulic properties. The fourth is a change in the distribution of solid materials in fractures through chemical reactions. Dissolved minerals can precipitate and seal fractures. Leaching fluids can dissolve fracture fillings or fracture walls. In the extreme, precipitates can completely till fractures and form mineral veins, or the rock can dissolve, leading to the development of karst. Most such changes are caused by natural processes, but induced pressure and temperature alterations can produce similar effects. CHANGES IN FRACTURE VOID GEOMETRY DUE TO CHANGES IN EFFECTIVE STRESS A variety of engineering activities cause fractured rocks to deform or fail. Changes in fluid pressures, the addition or redistribution of loads, or changes in temperature can lead to changes in the state of stress in the rock. The apertures of fractures depend critically on effective stresses acting normal to the fracture planes. The ability of a fracture to conduct fluid is extremely sensitive to the aperture. Four kinds of deformation are discussed below: (1) opening or closing of the fracture owing to changes in fluid pressure, (2) the formation and extension of fractures in response to elevated fluid pressures, (3) deformation or failure of

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications fractures in response to general changes in stress state, and (4) deformation of fractures owing to heating or cooling. Changes in Fracture Aperture Due to Fluid Pressure Changes Fluid pressure in fractures can change during many common engineering activities when fluid is withdrawn from or injected into a fractured rock. For fluid injection or withdrawal in a well, radial symmetry results in large fluid pressure changes near the wellbore. For steady-state Darcy flow into or out of a well with a diameter of a few tens of centimeters, most of the pressure change occurs in the first few meters of rock surrounding the well. The flow gradients alter the stresses by changing the pore pressure and by viscous drag on the pore walls. The drag is analogous to a body force acting in the direction of flow. The field of poroelasticity deals with the relationship between pore fluids and matrix stresses. Perkins and Gonzales (1984) provide an example of the use of poroelastic principles for the analysis of stress changes around an injection well. Any excavation in a rock mass under the groundwater table for tunnels, mines, or foundations tends to drain fluid from the rock. Fluid pressures decrease to near-atmospheric levels in the excavation, which increases the hydraulic gradient. Drainage causes a reduction in water pressure near the excavation. However, this effect can be counteracted by seepage stresses. Reduction of water pressure usually results in a strengthening of the rock mass because of the increase in effective stress. It is possible but rare that this reduction in water pressure can increase the effective stress sufficiently to cause failure and excavation collapse or borehole instability. At the same time, increases in water pressure can produce irreversible changes in fracture network properties, as described in the water well stimulation example illustrated in Appendix 4.C. Effective Stress and Fluid Flow Changes in fracture systems owing to variations in pore pressure can be understood by using the concept of effective stress, as discussed in Chapter 3. Fluid pressure in the pore spaces of rock (including the apertures of fractures) works against externally applied compressive stresses. For fractures the effective stress is the difference between the total stress applied on the fracture face and the pore pressure in the fracture (see Chapter 3). Stress-dependent properties such as frictional strength and fluid permeability are governed by the effective stress. An increase in effective stress will close the fractures and reduce their permeability; a decrease in effective stress will have the opposite effect. A field example of the relationship between flow behavior and effective stress is given in Figure 7.2, for a fractured sandstone gas reservoir at the U.S. Department of Energy's multiwell site in western Colorado (Warpinski, 1991). The matrix permeability of the marine sandstone reservoirs is less that 10-18 m2

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications FIGURE 7.2 Permeability vs. stress as determined in the multiwell site. Modified from Warpinski (1991). (less that one microdarcy; see Chapter 8). The minimum principal stress is normal to the dominant fracture orientation. The fracture network provides an overall in situ permeability that is several orders of magnitude greater than the matrix permeability and a horizontal permeability-anisotropy ratio of about 100:1 (inferred from interference testing with nearby wells). The stress-permeability relationship shown in Figure 7.2 is based on the model presented by Walsh (1981):

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications where k is permeability, ko is the reservoir permeability under in situ conditions, is the normal stress on the fracture, * is the reference stress state, and C is a constant. The parameters C and * were calibrated by using the results from two different stress conditions. The first was for a low effective stress [<6.89 MPa (<1000 psi)], achieved by injecting nitrogen. Under this condition, the effective reservoir permeability was found to be one to two orders of magnitude greater than that for initial reservoir conditions. The second was for high effective stress [>34.5 MPa (>5000 psi)], achieved by producing the well with a drawdown sufficient to stop all production (i.e., for k = 0). During multiple cycles of production, shutin, and injection, the permeability changes in this reservoir were found to be reversible, with no significant changes to the original permeability after reestablishment of the initial reservoir pressure (i.e., after a long shutin period). The impact of an increase in effective stress on reservoir performance is illustrated in Figure 7.3. This figure shows how the steady-state flow rate is related to steady-state drawdown for a rigid system (dashed line) and calculated for the compliant system (solid line) illustrated in Figure 7.2. In the rigid system, flow into the well is proportional to the drawdown. In the compliant system the increase in flow produced by increasing the drawdown, which increases the effective stress, is offset by the consequent decrease in fracture conductivity. For FIGURE 7.3 Change in steady-state flow rate as a function of drawdown based on the permeability reduction relationship given in Figure 7.2.

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications drawdowns somewhat greater than 6.89 MPa (<1000 psi), production can decline dramatically with increasing drawdown. Effective Stress and Anisotropy Changes in pore pressure can affect the degree of permeability anisotropy. For example, in cases with both significant matrix and fracture permeability where there is only one dominant fracture set, an increase in pore pressure will lower the effective stress. This leads to an increase in anisotropy by opening the fractures, thus increasing the permeability parallel to the fracture orientation. The permeability in other directions would be unchanged by comparison because it is controlled by the less stress-sensitive matrix. Where there is more than one fracture set, the nature of anisotropy change would depend on which of the sets were more deformable. In the case of decreasing pore pressures in a low-permeability rock matrix, the closing of fractures could lead to a situation where the rock mass permeability is equal to the matrix permeability by reducing fracture interconnectivity below the critical limit for percolation (see Chapter 6). In a poorly connected fracture network, a decrease in pore pressure could theoretically make the rock more isotropic. As some fractures become relatively impermeable, their connectivity is reduced to near the percolation limit. At the percolation limit, the flow in any direction must utilize fractures of all orientations. The flow observed on a large scale then depends on the same few open fractures, and permeability is consequently isotropic. Determining Effective Stress Determination of the stress field and pore pressure distributions are prerequisites to determining the effective stress. The state of stress is represented by a symmetric second-order tensor, so there will always exist three mutually perpendicular vectors that define the so-called principal stress directions. In general, the state of stress is anisotropic, and three principal values of stress and their directions are needed for a complete description. Generally, when the stress directions are not altered by nearby faults, the principal stresses are horizontal and vertical, with overburden pressure given by where is the rock density, g is the gravitational acceleration, and z is the depth below the surface. For rock with shear strength in an extensional regime, the vertical stress is usually the major principal stress; the horizontal stresses take on a variety of values. For rock with shear strength in a compressional regime, the maximum principal stress can be horizontal, and the least principal stress can be vertical. Near the surface the maximum principal stress also can be

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications horizontal if the terrain is overconsolidated, that is, if some of the original overburden has been removed by erosion. Measurement of the magnitude of the minimum principal stresses and their orientation by hydraulic fracturing is well established in the mining, petroleum, and geotechnical industries. Hydrofracturing is discussed more completely below. A number of other methods use measurements of strain to infer the state of stress. The pore pressure frequently has a hydrostatic gradient, which is given by p = wgz*, where w is is the density of water and z* is the depth below the water table. Under these conditions, the effective vertical stress, v, can easily be estimated [i.e., v = - p = g z - wz*)]. Both bulk deformation and fluid volume change in response to changes in pressure. When pressure changes in fractures, fluid diffuses at a rate that depends on the permeability distribution in the rock. For fractures in impermeable rock, the matrix has no effect on pressure change. Thus, a change in fluid pressure in the fractures will result in a proportional change in the effective stress acting on the fractures. However, for fractures in a permeable matrix, fluid will diffuse into or out of the surrounding rock, and, consequently, the pressure change introduced into the fractures will be damped. Engelder (1993) gives a recent summary of stress regimes in the lithosphere. Hydraulic flow models (Chapter 6) are often used to quantify the fluid pressure distribution in places where there are no measurements. Quasi-continuum models or closed form solutions can be used to calculate fluid pressures. To use these models, it is necessary to relate an equivalent quasi-continuum conductivity to the rock mass, either through single- or double-porosity models (see Chapter 6). Most discrete network models such as those described in Chapter 6 are applicable, but the limited knowledge of fracture geometry and fracture flow characteristics in most engineering problems makes their use unlikely. Further, variability in pressure is much smaller than variability in flow. Hence, less sophisticated models may be adequate. Predicting the Behavior of Stress-Sensitive Flow Modeling of subsurface flow is usually conducted under the assumption that permeability is independent of the state of stress. Some work has been done, however, on flow through so-called pressure-sensitive media. Because the permeability of fractures usually exhibits some dependence on stress, these models should be applicable to fractured rock masses. One of the earliest treatments of this problem was by Raghavan et al. (1972), who developed type curves for flow in a bounded circular reservoir whose permeability and porosity varied with the pore fluid pressure. Wall et al. (1991) present drawdown curves and pressure profiles for slug tests in pressure-sensitive formations. Dvorkin and Nur (1992) present an approximate analytical solution for a one-dimensional filtration front invading a formation whose permeability changes from zero to ko as the pressure

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications changes from p < pc to p > pc, where pc is some critical pressure at which an interconnected network of fractures is formed. In some cases the flow system influences the effective stress enough to significantly change the permeability of the flow system. In these cases it may be necessary to use a coupled stress-flow model to understand the behavior of the fracture flow system. Several coupled models are available (Noorishad et al., 1971, 1982, 1992; Kafritsas, 1987). The hydrogeological simulation models that have been linked to mechanical deformation models originated primarily in geotechnical engineering fields. Coupled models account for the role of deformation in fluid flow and the stability of the rock mass. Flow is linked to deformation through changes in permeability and storage. Deformation is linked to flow through changes in fluid pressure, which results in changes in effective stress. A critical element in these models is representation of the proper hydrological and material constitutive behaviors for the fracture and solid rock. Models based on a discrete representation of the fracture network assign separate values to the fracture and block stiffnesses. Fluid flow is also modeled on a fracture-by-fracture basis (e.g., Asgian, 1989). These hydromechanical models have been used to examine the coupling between fluid pressure responses and deformation for idealized settings. Representation of the fractured material in the mechanical model is more complex than the representations described in Chapter 6 for fluid flow. Just as for fluid flow, if the rock is highly fractured and the fractures have a wide range of orientations, it may be possible to use continuum deterministic or stochastic deformation techniques. For heterogeneous fracturing, additional complexity must be built into the model. For example, rock with a series of parallel fractures may be represented as a laminated material. Methods for coupled modeling of flow and deformation have been evaluated by DECOVALEX (1991). Discrete-element models have been used for coupled fracture flow and deformation problems, such as flow under a dam, rock slope behavior, and flow to or from a well (Noorishad et al., 1971; Kafritsas, 1987). Figure 7.4 provides examples of models of this type. A well is located at the center of a rectangular domain, which contains a much simplified fracture network. The diagram on the left illustrates the computed deformation of the rock mass that occurs with fluid injection. The diagram on the right illustrates the nature of deformation if water is withdrawn from the rock mass through the well. Assessment of Models of Stress-Sensitive Fluid Flow In general, the principles governing the relationship between pressure changes and fluid flow are well understood. However, prediction of this behavior through the use of models is problematic. There are few experiments available at field scales that can be used to test these models. The problems are similar to those encountered in defining fracture flow systems (see Chapter 6). There are difficulties

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications FIGURE 7.4 Deformation computed by a coupled flow and deformation analysis for a well under (a) injection and (b) withdrawal. Computations performed with the Massachusetts Institute of Technology discrete element model. From Kafritsas (1987). in scaling up point measurements to obtain values that apply to field-scale behavior. There are considerable amounts of laboratory data that establish constitutive relationships between effective stress and fracture permeability for single fractures in rock cores (see Chapter 3), but field-based data are few in number. Boundary conditions are often unknown, yet they may control flow and stress behavior. In attempting to fit hydromechanical models to data sets involving single-borehole deformation experiments, it has been observed that there are problems in identifying unique sets of parameters to characterize field responses. Some success has been achieved in duplicating single-fracture experiments (Rutqvist et al., 1992). However, because of the multitude of parameters that determine the nonlinear hydromechanical response of a rock mass containing a network of fractures, site-specific predictive simulation is extremely difficult. The practical significance of stress-sensitive fluid flow in fractured rock needs to be evaluated for a range of field-scale problems. This can be accomplished by using numerical simulation models that account for stress-dependent flow. Creation or Extension of Fractures Due to Increases in Fluid Pressures Hydrofracture is the extensional failure of rock that takes place when fluid pressure increases to the point that effective stresses become negative and exceed the tensile strength of the rock. Hydraulic fracturing occurs intentionally or unintentionally during many activities that provide sources of fluids at high pressures. For fractures to propagate, the fluid must be supplied in sufficient amounts to ''fill" the growing volume of opening fractures. There is an extensive literature available on hydrofracture, and only a brief description is given here.

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications Hydrofracture is used as a stress measurement tool (e.g., Engelder, 1993). In this application a hydrofracture is created in a section of borehole isolated with packers. The pressure in the interval is increased until the borehole ruptures, resulting in the formation of new fractures or the extension of old ones, which causes a sudden decline in pressure. The pressure required to maintain the fracture openings is assumed to be equal to the minimum principal stress. A slightly higher pressure will extend the fracture. (Once it has closed, the pressure required to reopen it can be significantly higher.) For homogeneous isotropic formations, hydrofractures grow in a direction perpendicular to the minimum principal stress. Hydraulic fracturing is also used to increase conductivities around production wells. A decline in permeability can arise from high effective stresses at the borehole, owing to: (1) stress concentrations around the borehole (which depend on the original stress field) or (2) large pressure drops around the borehole owing to radial flow. A newly formed hydrofracture can increase the conductivity near the well and consequently increase the effective wellbore radius by as much as one-half the fracture penetration (Prats, 1961). Hydraulic fracturing is a common practice in wells used for resource recovery. Typical penetrations (i.e., the propped lengths of the hydrofractures) are greater than 200 m, producing an effective well radius of more than 100 m. Fracturing is a very practical means to achieve a large effective wellbore radius at significant depths. Hydrofracture can also occur, unintentionally, during injection of fluids. In these cases, studies can be designed to find safe injection pressures that will not cause fractures to open and extend. Control of fracturing during "flooding" operations has been a long-term concern of the petroleum industry. Flooding includes water injection for "secondary recovery" and chemical injection for "tertiary recovery" (e.g., Lake, 1989). The objective of flooding is to provide an effective sweep of the reservoir with the injected fluids. This is achieved by controlling the extent to which hydraulic fractures propagate from the injection well toward a production well in order to prevent a ''short-circuit" in the flow pattern of the injected fluid. Mud pressure during drilling can either open or create new fractures. This phenomenon can lead to lost circulation (too little drilling fluid returning to the surface) and, in the extreme, to loss of the hole owing to rock failure and stuck tools. The control of hydraulic fracturing is also important for waste disposal wells, and much of the petroleum industry's experience and technology are directly applicable. Over time, injection wells are subjected to the cumulative effects of particulates and other contaminants in the injected fluids, which decrease permeabilities. The maintenance of nominal injection rates requires increasing the injection pressure. Eventually, the elevated injection pressures will cause the extension of fractures that create new injection areas beyond the region of significant permeability impairment. Alternatively, the permeability decrease can be ameliorated by chemical treatment (e.g., Economides and Nolte, 1989).

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications A dam supporting a filled reservoir has large pore and joint water pressure gradients in its abutments and foundations. Such high pressures can reduce the strength of the rock mass and can cause movement of rock blocks, with potentially catastrophic consequences. The pressures can also open fractures and allow significant amounts of fluid to flow around or under the dam. Large pore pressures under the toe or downstream of a dam, in combination with the large shear stresses, can produce foundation failures. Entirely analogous but more serious conditions exist in the dam abutment slopes because the water pressure produces both high pore pressures and high hydraulic thrusts. The 1959 Malpasset Dam failure is an example of such a failure (Londe, 1987). A related problem is that of reservoir-induced earthquakes or similar seismic events. High water loads and pore pressures can induce slippage along fractures. Earthquakes of magnitudes up to 6.5 have occurred, owing to either injection or withdrawal of fluids (Meade, 1991). Studies indicate that the level of induced seismicity depends strongly on preexisting stress conditions in the upper crust. Fracture Initiation and Growth A fracture is initiated when the fluid pressure exceeds the total stress plus the tensile strength of the rock, as discussed earlier. Stress heterogeneity plays a role because it is the total local stress that must be exceeded in order to initiate a fracture. For example, total stress near a well may be elevated above the far-field stress by stress concentrations around the wellbore. If fractures are present, the pressure required to open them is equal to the compressive stresses across the fractures. The stress intensity factor at the tips of the fractures must exceed a critical value in order for the fractures to be extended. The stress intensity factor is a function of the fluid pressure in the crack and the crack geometry, as described in Chapter 2. The fracture growth process is governed by the coupling between fracture deformation, fluid flow, and the mechanical properties and stresses in the matrix. For spatially constant properties and stresses, the pressure required to propagate the fracture decreases as the fracture gets longer, because the greater length of the fracture provides greater leverage on the tip (Economides and Nolte, 1989). Stress variations between different sedimentary layers can enhance lateral fracture extension relative to vertical fracture extension. Such stress variations can produce elongated fractures that will propagate only by increasing pressure (Teufel, 1979). This is a common occurrence for hydraulic fracturing in petroleum reservoirs, as illustrated in Figure 7.5. The figure indicates generally increasing pressure during the injection phase (denoted as "fracture treatment" in the figure). The interval after injection is divided into two time periods. During the first period of "fracture closing," the fluid pressure and fracture aperture decrease as fluid in the fracture infiltrates the surrounding rock (Warpinski, 1985). The period

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications design is explicitly determined and can be compared to the consequences of mitigation or the consequences (i.e., cost) of choosing a design with a reduced risk. Adaptable/Observational Methods As shown in Figure 7.13, a design is chosen on the basis of the most likely expected performance, which can be determined formally or intuitively. Simultaneously, contingency designs are developed to address conditions that deviate from the most likely case. Most importantly, performance criteria and monitoring approaches are included in all designs. The monitored performance is compared with the target performance, and, if deviations are observed, the appropriate contingency design is chosen. Minor adaptations can be made ad hoc. Peck (1969) and Einstein (1978) review many of the applicable methods. Clearly, combinations of the three approaches discussed here are possible, such as determining the distribution of possible performances, and then using the performances to evaluate the benefits of further exploration before deciding on the most likely performance. SUMMARY OF DEFICIENCIES AND RESEARCH NEEDS The preceding discussion of rock-fracture-related problems in civil and mining engineering and engineering geology identified a number of problems: It is not possible to completely characterize the geometric, mechanical, and chemical properties of rock masses. Exploration methods often have inadequate ranges of resolution and penetration and are too expensive. Even with advances in exploration methods, there will always be uncertainties in rock mass characterization. Further formalization of adaptable analysis-design procedures is required, and, particularly, procedures are needed that allow quick adaptation as observations provide new information. Clearly, any progress in knowledge about the geometry, mechanics, and mineralogy of individual fractures and fracture systems is desirable. Most important are relationships between geometry, mechanical behavior, and the creation of new fractures. Knowledge of the long-term behavior of most rock mass changes is unreliable.

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications FIGURE 7.13 Knowledge of design parameters is updated through monitoring and feedback.

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications APPENDIX 7.A NATURAL FRACTURING The basic mechanics of the fracturing process provide insights into natural fracturing deep within the earth's crust. For this case it is appropriate to assume that the formations are massive with continuous field and property gradients. The vertical gradient of stress is proportional to rock density and is larger than the vertical fluid pressure gradients, which means that the effective stress is decreasing in the upward direction. This stress gradient promotes upward propagation of hydrofractures, as illustrated in Figure 7.A1, which shows results from FIGURE 7.A1 Propagation of a vertical hydrofracture. After Clifton (1989).

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications a numerical simulation of a hydraulic fracture with a fluid source centered at the origin. The fracture front progresses, in time, from a near-radial shape to an elongated shape. The rate of penetration and degree of vertical preference are governed by the viscous dissipation of the vertical pressure gradient. The preference for vertical growth from a fluid source of massive horizontal dimension will provide parallel sets of fractures. The interaction of multiple fractures is influenced by the altered stress field around the fractures caused by their deformation. Furthermore, in massive formations the magnitude of the negative effective stress (i.e., pressure) required to propagate a fracture tends to decrease as the fracture dimension increases. The decreasing pressure requirement can be offset by increasing viscosity as the intruding fluid rises and is cooled as it moves away from a heat source. Because they require less pressure in order to propagate, larger fractures will propagate preferentially when all the fractures are connected by a common pressure source. The existing fractures induce a compressive stress in the direction normal to the fracture plane on a scale proportional to the size of the fracture. The scale-dependent stress increase tends to close neighboring fractures, particularly smaller ones with smaller apertures. The coupling of the scale-dependent stress increase and the preference for growth of larger fractures will define the size distribution and spacing between fractures of various sizes. Additional features of natural hydraulic fracture networks can be illuminated by the comprehensive numerical models used in the petroleum industry (Gidley et al., 1989, Chapters 3, 4, and 5).

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications APPENDIX 7.B DRAINAGE METHODS IN CONSTRUCTION Water flow from excavation surfaces may impede construction (operation) or erode fracture fillings or fracture wall rocks. Water flow can be prevented or reduced by applying so-called drainage layers to the excavation walls. These external drainage layers consist of relatively permeable (low cement content) shotcrete or geotextiles (Figure 7.B1a), which are frequently applied in combination with an impermeable seal to form a geomembrane. If drainage occurs from only a few fractures, hoses can be inserted into the fractures to relieve the flow (Figure 7.B1b). Frequently, short holes are drilled into the rock mass to capture the flow before it reaches the surface of the excavation (Figure 7.B1c). This approach may be used in combination with hoses or drainage layers. External drainage involves the capture of water or gas near an excavated or natural surface. When incorporated into permanent structures, this type of drainage must be a carefully designed system of seals and pipes for the removal of water (Figure 7.B1d). Water pressures should not exceed specified limits to prevent overloading of the seal or the permanent supports. However, a reduction in water pressure may result in the precipitation of calcite in fractures if the water is carrying calcite in solution. The purpose of internal drainage is to capture water (gas) at a distance from the natural or excavated surface. This type of drainage can be used to induce favorable stress conditions in a zone near the surface and avoid some of the above-mentioned problems when water drains directly at the surface. Many internal drainage methods make use of boreholes that intercept fractures in the rock mass. Boreholes can be arranged in several geometries as shown in Figure 7.B1e. Possible arrangements include single or multiple rows of vertical or inclined holes to form ''drainage curtains"; drainage curtains are also used in conjunction with dams (Figure 7.B1f). Borehole drainage curtains can be replaced by so-called permeable cutoffs, which are excavated trenches or slots filled with permeable materials (Figure 7.B1 g). The objective of all these methods is to intercept fractures that are under high pore pressures or that are connected to fracture systems under high pore pressures. Depending on the fracture and fracture system geometry, this may require that boreholes be oriented other than in simple vertical or horizontal rows. Figure 7.B1h shows a drainage gallery (tunnel) that can be used alone or in conjunction with drain holes to improve drainage. Internal drainage can also be induced by increasing the permeability of fractured zones in the rock mass. Methods for accomplishing this involve the creation of new fractures by blasting or hydraulic fracturing and increasing fracture conductivity by propping. These methods are discussed earlier in this chapter, under "Creation or Extension of Fractures Owing to Increases in Fluid Pressures."

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications FIGURE 7.B1 Schematic illustration of drainage methods. (a) External drainage layers. (b) Drainage through hoses inserted into fractures. (c) Drainage through short holes drilled into the rock mass. (d) Drainage beneath a permanent liner. (e) Internal drainage through boreholes. (f) Drainage curtain in a dam foundation. (g) Drainage cutoff trench. (h) Drainage tunnel.

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications Fluids from internal drainage can be removed through gravity flow or pumping. Erosion of fracture fillers or fracture wall rock can be minimized by placing filters and suitably graded materials into the drainage facilities. REFERENCES Advani, S. H., T. S. Lee and J. K. Lee. 1990. Three-dimensional modeling of hydraulic fractures in layered media: Part I—finite element formulations. ASME Journal of Energy Research Technology, 112:1–9. Asgian, M. 1989. A numerical model of fluid flow in deformable naturally fractured rock masses. International Journal of Rock Mechanics and Mining Science and Geomechanics Abstracts, 26:(3/4) 317–328. Ashley, D. B., H. H. Einstein, and E. Tse. 1979. Advantages and limitations of adaptable tunnel design—construction methods. Pp. 989–1011 in Proceedings of the 4th Rapid Excavation and Tunneling Conference. Richardson, Tex.: American Institute of Mining, Metallurgical and Petroleum Engineers. Biot, M. A. 1956. Thermoelasticity and irreversible thermodynamics. Journal of Applied Physics, 27:240–253. Boley, B. A., and J. H. Weiner. 1960. Theory of Thermal Stresses. New York: John Wiley & Sons. Cambefort, H. 1964. Injection des Sols. Paris: Editions Eyrolles. Cedergren, H. R. 1977. Seepage, Drainage, and Flownets. New York: John Wiley & Sons. Clifton, R. J. 1989. Three-dimensional fracture-propagation models. In Recent Advances in Hydraulic Fracturing, J. L. Gidley, S. A. Holditch, D. E. Nierode, and R. W. Veatch, eds. Monograph Series 12. Richardson, Tex.: Society of Petroleum Engineers. Clifton, R. J., and J. J. Wang. 1988. Multiple fluids, proppant transport and thermal effects in 3-dimensional simulation of hydraulic fracturing. Paper presented at the 63rd Annual Technical Conference, Oct. 2-5, 1988, Houston, Tex. SPE 18198. Richardson, Tex.: Society of Petroleum Engineers. DECOVALEX. 1991. Bench-mark test 2: multiple fracture model. DECOVALEX Doc 91/104. DECOVALEX Secretariat. Stockholm: Royal Institute of Technology. De Marsily, G. 1987. An overview of coupled processes with emphasis on geohydrology. In Coupled Processes Associated with Nuclear Waste Repositories, C. F. Tsang, ed. New York: Academic Press. Doughty, C., and K. Pruess. 1988. A semianalytical solution for heat-pipe effects near high-level nuclear waste packages buried in partially saturated geological media. International Journal of Heat Mass Transfer, 31:79-90. Dvorkin, J., and A. Nur. 1992. Filtration fronts in pressure compliant reservoirs. Geophysics, 57:1089-1092. Economides, M. J., and K. G. Nolte. 1989. Reservoir Stimulation, Second Edition. Englewood Cliffs, N.J.: Prentice-Hall. Einstein, H. H. 1978. Observational Tunnel Design—Construction Methods in the U.S. Shotcrete for Underground Support III. New York: Engineering Foundations. Einstein, H. H., D. A. Labreche, M. J. Markow, and G. B. Baecher. 1976. Decision analysis applied to rock tunnel exploration. Engineering Geology, 12(2):143–160. Einstein, H. H., G. B. Baecher, and D. Veneziano. 1979. Risk analysis for rock slopes in open pit mines. MIT report 80–20 to U.S. Bureau of Mines, vol. III, Massachusetts Institute of Technology, Department of Civil Engineering, Cambridge, Mass. Engelder, T. 1993. Stress Regimes in the Lithosphere. Princeton, N.J.: Princeton University Press, 457 pp.

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications Fredrick, J. T., and Wong. 1986. Micromechanics of thermally induced cracking in three crustal rocks. Journal of Geophysical Research, 91:12,743–12,764. Geertsma, J., and F. de Klerk. 1969. A rapid method of predicting width and extent of hydraulic induced fractures. Journal of Petroleum Technology, Dec., pp. 1571–1581. Gidley, J. L., S. A. Holditch, P. Nierode, and R. Veach, eds. 1989. Recent Advances in Hydraulic Fracturing, vol. 12. Richardson, Tex.: Society of Petroleum Engineers. Goodman, R. E. 1976. Methods of Geological Engineering. St. Paul, Minn.: West Publishing. Hendron, A. J., E. J. Cording, and A. K. Aiyer. 1971. Analytical and graphical methods for the analysis of slopes in rock masses. Technical Report No. 36, Nuclear Cratering Group. Hoek, E., and J. W. Bray: 1981. Rock Slope Engineering, 3d ed. London: The Institute of Mining and Metallurgy. Hunt, J. R., L. McDowell-Boyer, and N. Sitar. 1987. Colloid migration in porous media: an analysis of mechanisms. Pp. 453–472 in Coupled Processes Associated with Nuclear Waste Repositories, C. F. Tsang, ed. New York: Academic Press. Kafritsas, J. 1987. Coupled flow/deformation analysis with the distinct element method. Ph.D. thesis, Massachusetts Institute of Technology, Cambridge. Khristianovich, S. A., and Y. P. Zheltov. 1955. Formation of vertical fractures by means of highly viscous liquids. Pp. 579–586 in Proceedings of the Fourth World Petroleum Congress, section II/T.O.P., paper 3, Rome. Kohl, T., K. I. Evans, R. J. Hopkirk, and L. Rybach. 1996. Modeling of coupled hydraulic thermal and mechanical processes in the simulation of hot dry rock reservoir behavior. In Fractured and Jointed Rock Masses. Rotterdam: A. A. Balkema. Lai, C. H. 1986. Mathematical models of thermal and chemical transport in geologic media . Ph.D. thesis, University of California, Berkeley, p. 188. Lake, L. W. 1989. Enhanced Oil Recovery. Englewood Cliffs, N.J.: Prentice-Hall. Londe, P. 1987. The Malpasset Dam failure. Engineering Geology, 24(1–4): 295–330. Long, J. C. S., O. Olsson, S. Martel, and J. Black. 1992. Effects of excavation on water inflow to a drift. In Fractured and Jointed Rock Masses. Rotterdam: A. A. Balkema. Majer, E. L., L. R. Myer, J. E. Peterson, Jr., K. Karasaki, J. C. S. Long, S.J. Martel, P. Blumling, and S. Vomvoris. 1990. Join Seismic, Hydrogeological, and Geomechanical Investigations of a Fracture Zone in the Grimsel Rock Laboratory. LBL-27913, Lawrence Berkeley Laboratory, Berkeley, Calif., 173 pp. Martin, C. D., C. C. Davison, and C. T. Kozak. 1990. Characterizing normal stiffness and hydraulic conductivity of a major shear zone in granite. Pp. 549–556 in Rock Joints, Proceedings of the International Symposium on Rock Joints, N. Barton and O. Stephansson, eds. Rotterdam: A. A. Balkema. Meade, R. B. 1991. Reservoirs and earthquakes. Engineering Geology, 30(314):245–262. Meyer, B. R. 1989. Three-dimensional hydraulic fracturing simulation of personal computers: theory and comparison studies. SPE 19329. Pp. 213–230 in Proceedings of the Eastern Regional Meeting of Society of Petroleum Engineers, Morgantown, W. Va. Richardson, Tex.: Society of Petroleum Engineers. Nolte, K. G. 1982. Fracture design considerations based on pressure analysis. SPE 10911. Paper presented at Cotton Valley Symposium of the Society of Petroleum Engineers, Tyler, Tex., May 20. Richardson, Tex.: Society of Petroleum Engineers. Noorishad, J., P. A. Witherspoon, T. L. Brekke, and T. Maini. 1971. Methods for coupled stress and flow analysis of fractured dam foundations and rock slopes. Geotechnical Engineering Publication 71-6, University of California, Berkeley. Noorishad, J., M. S. Ayatollahi, and P. A. Witherspoon. 1982. Coupled stress and fluid flow analysis of fractured rocks. International Journal of Rock Mechanics and Mining Sciences and Geomechanics Abstracts, 19:185–193.

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications Noorishad, J., C. F. Tsang, and P. A. Witherspoon. 1984. A coupled thermal-hydraulic-mechanical finite element model for saturated fractured porous rocks. Journal of Geophysical Research, 89 (B12):10,365–10,373. Noorishad, J., C. F. Tsang, and P. A. Witherspoon. 1992. Theoretical and field studies of coupled hydromechanical behavior of fractured rocks: I—Development and verification of a numerical simulator. International Journal of Rock Mechanics and Mining Sciences and Geomechanics Abstracts, 29(4):401–409. Olson, J. 1990. Fracture mechanics of joints and veins. Ph.D. dissertation, Stanford University, Stanford, Calif., p. 174. Peck, R. B. 1969. Advantages and limitations of the observational method in applied soil mechanics. Geotechnique, 19:171–187. Perkins, T. K., and J. A. Gonzalez. 1984. Changes in earth stresses around a wellbore caused by radially symmetrical pressure and temperature gradients. Society of Petroleum Engineers Journal, Apr., pp. 129–140. Perkins, T. K., and J. A. Gonzalez. 1985. The effect of thermoelastic stresses on injection well fracturing. Society of Petroleum Engineers Journal, Feb., pp. 78–88. Perkins, T. K., and L. R. Kern. 1961. Widths of hydraulic fractures. Journal of Petroleum Technology, Sept., pp. 937–949. Phillips, O. M. 1991. Flow and Reactions in Permeable Rocks. New York: Cambridge University Press, pp. 57–58. Prats, M. 1961. Effect of vertical fractures on reservoir behavior incompressible case. Society of Petroleum Engineers Journal, June, pp. 105–118. Raghavan, R., J. D. T. Scorer, and F. G. Miller. 1972. An investigation by numerical methods of the effect of pressure-dependent rock and fluid properties on well flow tests. Society of Petroleum Engineers Journal, 12:267–275. Rutqvist, J., J. Noorishad, O. Stephansson, and C. F. Tsang. 1992. Theoretical and field studies of coupled hydromechanical behavior of fractured rocks: 2. Field experiment and numerical modeling. International Journal of Rock Mechanics and Mining Sciences and Geomechanics Abstracts, 29 (4):411–419. Settari, A., and M. P. Cleary. 1984. Three-dimensional simulation of hydraulic fracturing. Journal of Petroleum Technology, July, pp. 1177–1190. Stephens, G., and B. Voight. 1982. Hydraulic fracture theory for conditions of thermal stress. International Journal of Rock Mechanics and Mining Science and Geomechanical Abstracts, 19:279–284. Stille, H. G. Gustafson, V. Hakansson, and P. Olsson. 1992. Passage of water bearing fracture zones—experiences from grouting of the section I-1400 m of the tunnel. SKBPR 25-92-19, Swedish Nuclear Fuel and Waste Management Company, Stockholm. Teufel, L., D. W. Rhett, and H. E. Farrell. 1991. Effect of Reservoir Depletion and pore pressure drawdown on in-situ stress and deformation in the Ekofisk Field, North Sea. Proceedings of the 32d U.S. Rock Mechanics Symposium. Rotterdam: A. A. Balkema. Teufel, L. W. 1979. An experimental study of hydraulic fracture propagation in layered rock. Ph.D. thesis, Texas A&M University, College Station. U.S. Bureau of Mines. 1989. In-situ leach mining. P. 107 in Proceedings: Bureau of Mines Technology Transfer Seminars. IC 9216, Pittsburgh, Pa. Wall, J., A. Nur, and J. Dvorkin. 1991. A slug test method in reservoirs with pressure sensitive permeability. Pp. 95–105 in Proceedings of the 1991 Coalbed Methane Symposium, May 13-16, University of Alabama, Tuscaloosa. Walsh, J. B. 1981. Effect of pore pressure and confining stress on fracture permeability. International Journal of Rock Mechanics and Mining Science and Geomechanics Abstracts, 18:429–435. Wan, J., and J. L. Wilson. 1994a. Visualization of the role of the gas-water interface on the fate and transport of colloids in porous media. Water Resources Research, 30:11–23.

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications Wan, J., and J. L. Wilson. 1994b. Colloid transport in unsaturated porous media. Water Resources Research, 30:857–864. Wan, J., J. L. Wilson, and T. L. Kieft. 1994. The effects of the gas-water interface on the fate and transport of microorganisms in unsaturated porous media. Applied and Environmental Microbiology, 60:509–516. Wang, J. S. Y., C. F. Tsang, N. G. W. Cook, and P. A. Witherspoon. 1981. A study of regional temperature and thermohydrologic effects of an underground repository for nuclear wastes in hard rock. Journal of Geophysical Research, 86:3759–3770. Warpinski, N. R. 1985. Measurement of width and pressure in a propagating hydraulic fracture. Society of Petroleum Engineers Journal, Feb., pp. 46–54. Warpinski, N. R. 1991. Hydraulic fracturing in tight, fissured media. Journal of Petroleum Technology, Feb., p. 146. Warpinski, N. R., and L. W. Teufel. 1987. Influence of geological discontinuities on hydraulic fracture propagation. Journal of Petroleum Technology, Feb., pp. 209–220. Warpinski, N. R., Z. A. Moschovidia, C. D. Parker, and I. S. Abou-Sayed. 1994. Comparison study of hydraulic fracturing models—test case: GRI staged field experiment no. 3. SPE Production and Facilities, Feb., pp. 7–160. Williams, D. B., et al. 1989. Impact of inducing fractures at Prudhoe Bay. Journal of Petroleum Technology, Oct., pp. 1096–1101. Wittke, W. 1965. Verfahren zur Standsicherheitsberechnung starrer, auf ebenen Flächen gelagerter Körper und die Anwendung der Ergebnisse auf die Standsicherheitsberechhung von Felsböschungen, Heft 20. Karlsruhe: Veröffentlichungen des Institutes für Bodenmechanik und Grundbauder Technischen Hochschule Fridericiana. NOTE 1   Other aggregates or filters can be fly ash, diatomaceous earth, rock flour, sawdust and gravel. 2   Single-phase injection of silicate solutions containing a gelation retarder is used in several patented processes. Two-phase injection of silicate solutions is the well-known Joosten process, and two-phase injection of silicate solutions followed by an aluminum surface solution is Farnçois process. The reader is referred to Combefort (1964) for detailed descriptions of these processes. 3   Note that the term multiple staging has nothing to do with upward or downward staging.

OCR for page 405
Rock Fractures and Fluid Flow: Contemporary Understanding and Applications This page in the original is blank.