Vertical Fracture Propagation Research Articles

Unraveling the complex deformation pattern at Yellowstone plateau through seismicity and fracture analysis

The intense unrest of the Yellowstone volcanic plateau is manifested through diffuse seismic activity, earthquake swarms, and episodes of complex surficial deformation that have been related to magmatic fluid transfer in the upper crust over the past several decades. While past studies have focused on modeling contemporary geophysical data, there has not been a fully-integrated evaluation of seismicity, fault kinematics, and stress field. Here we analyze a catalog of 10,201 relocated earthquakes recorded between 2010 and 2016 and determine 224 well-constrained double-couple focal mechanisms. The majority of the focal mechanisms (65%) are associated with the 2010 Madison Plateau seismic swarm. The focal mechanisms are predominantly strike-slip with subordinate normal faulting mechanisms. Possible causes of this predominance and of the concurrence of both kinematics are here discussed, in order to unravel the influence of magmatic processes such as past sill intrusions. The earthquake catalog has been analyzed in terms of location, time, and kinematics according to the phases of surficial deformation documented by GPS data in order to identify systematic patterns of deformation and has been compared to the 1988–2009 seismicity. The continuous downwarping of the overburden from 2010 to 2016 was accomodated by structural adjustment of the shallow crust through strike-slip motions on a multitude of scattered small fault planes. Furthermore, the predominance of strike-slip faulting during seismic swarms occurs when the fluid overpressure induces horizontal propagation of vertical fractures with strike-slip motions, followed by horizontal fluid flow.

Experimental Investigation of Hydraulic Fracturing for Multi-type Unconventional Gas Co-exploitation in Ordos Basin

The accumulated gas production and production cycle would be significantly increased by exploiting the unconventional gas resources of coalbed methane and tight gas jointly in coal measure strata (tight gases–coal interbedding reservoir) in Ordos Basin in China. However, whether hydraulic fractures could connect multiple gas zones effectively in the vertical plane is a key problem. To study hydraulic fracture geometries and vertical propagation rules for coal measure strata, six groups of true triaxial experiments in natural outcrops collected from the eastern margin of Ordos Basin in China were carried out. Results show that unlike conventional symmetric fracture geometry in layered media, the principle of fracture vertical propagation in coal measure strata follows the least resistance and most preferential propagation path. The natural fractures and cleats in coal seam make the overall fracture morphology asymmetric. A moderate stress difference between vertical stress and maximum horizontal stress (4 MPa in this paper) is the key for connection of multiple layers as well as SRV enhancement. Developed natural fractures and bedding planes in interlayer could alter fracture extension path and impede hydraulic fracture crossing layer interface. In the process of hydraulic fracture propagating from interlayer to layer interface and then to the coal seam, the fluid pressure in hydraulic fractures is characterized by an apparent fluctuation of decreasing first and increasing later. Furthermore, the greater the elastic modulus of the interlayer, the greater the pump pressure fluctuations.

Investigations into Mining-Induced Stress–Fracture–Seepage Field Coupling in a Complex Hydrogeology Environment: A Case Study in the Bulianta Colliery

The effect of mining-induced multi-physics coupling, characterized by stress–fracture–seepage field coupling (SFCC), is critical to safe mining and environmental protection. We investigated the spatiotemporal characteristics of the SFCC at the #31401 panel of the Bulianta Colliery. Changes in overburden movement, groundwater level, and groundwater inrush were monitored by means of borehole and working face observations, water leakage, and real-time video imaging. The results indicate that the distribution of mining-induced fractures was dominated by the key stratum, distributed as a “ladder shape” 30–75 m above the coal seam. The fracture penetrated into the lower key stratum and developed into the confined aquifer with a height of 140.5–154.0 m, while being truncated by the higher key stratum. Horizontal and vertical fractures developed sequentially in the confined aquifer-disturbed zone. Initiation, propagation, and interconnection of the seepage channel occurred in the fracture field, mediated by the redistributed stress. An abnormal discharge of 120–300 m3/h along the direction of mining commenced at the transition of the lower key stratum to the higher key stratum, and concentrated below the Bulian Gully. A fracture height model was developed based on the vertical fracture propagation, and mining parameters were modified, including reducing the mining height and increasing the advance rate and support resistance of the longwall mining face to effectively reduce the SFCC effect.

Observing the effects of transition zone properties on fracture vertical propagation behavior for coal measure strata

The transition zone is a typical interface with enough thickness and high heterogeneity between the sandstone and coal and has a great effect on the fracture propagation in coal measure strata. In the study, the true tri-axial experiments were conducted with the combination of sandstone, coal and limestone outcrops. The effect of cementing strength and thickness of the transition zone was mainly studied. Three types of hydraulic fracture initiation positions were used to simulate three fracturing situations in the field. The results showed that the effect of the transition zone on the fracture vertical extension could be concluded: the fracture was easy to cross or slip with low cementing strength and turned inside the transition zone with high cementing strength; the fracture tended to branch or be blocked by the thick transition zone. The horizontal flow through face cleats and natural fracture activation in coal contributed to a more developed fracture network. The characters of pumping curve pressure “increasing – fluctuation – dropping”, and the “fluctuation pressure more than initiation pressure” indicated the interface crossing. The results of experiments agreed with the previous numerical simulation and helped to optimize the fracturing design in layered coal strata.

Effect of interface property on hydraulic fracture vertical propagation behavior in layered formation based on discrete element modeling

Effect of interface property on hydraulic fracture vertical propagation behavior in layered formation based on discrete element modeling

Microseismic evidence for horizontal hydraulic fractures in the Marcellus Shale, southeastern West Virginia

An integrated analysis of multistage hydraulic fracture stimulation done in three horizontal wells in the Marcellus Shale in southeastern West Virginia revealed that nearly half (14 out of 31) of the examined stages initiated horizontal hydraulic fractures. The study was performed in an area characterized by a transitional strike-slip/reverse faulting stress state where the maximum horizontal principal stress (SHmax) is oriented N55°E. The stages that produced horizontal hydraulic fractures were all within the organic-rich Lower Marcellus Shale. Two lines of evidence indicate horizontal hydraulic fracture propagation. The measured least principal stress of those stages was of similar magnitude to the vertical stress (SV), indicating that the vertical stress and microseismic events are limited to a narrow horizontal layer and do not propagate vertically out of the Lower Marcellus Shale. Both lines of argument indicate that the vertical stress is the least principal stress, perhaps due to viscoplastic stress relaxation in the clay- and kerogen-rich Lower Marcellus Shale. In the 17 stages where perforations were placed in the stiffer Onondaga and Cherry Valley formations, the measured least principal stresses were less than the magnitude of SV, indicating that Shmin was the least principal stress. The microseismic data indicate vertical hydraulic fracture propagation, principally upward outside the Marcellus Shale. Significant gas was produced from the two wells with horizontal fractures in the organic-rich Lower Marcellus Shale, perhaps because twice as much proppant was used as in stages characterized by vertical fracture propagation.

Laboratory investigation of shale rock to identify fracture propagation in vertical direction to bedding

Affected by beddings and natural fractures, fracture geometry in the vertical plane is complex in shale formation, which differs from a simple fracture in homogeneous sandstone reservoirs. However, the propagation mechanism of a hydraulic fracture in the vertical plane has not been well understood. In this paper, a true tri-axial pressure machine was deployed for shale horizontal well fracturing simulation experiments of shale outcrops. The effects of multiple factors on hydraulic fracture vertical propagation were studied. The results revealed that hydraulic fracture initiation and propagation displayed four basic patterns in the vertical plane of laminated shale formation. A hydraulic fracture would cross the beddings under the high vertical stress difference between a vertical stress and horizontal minimum stress of 12 MPa, while a hydraulic fracture propagates along the beddings under a low vertical stress difference of 3 MPa. Four kinds of fracture geometry, including a single main fracture, a nonplanar fracture, a complex fracture, and a complex fracture network, were observed due to the combined effects of flow rate and viscosity. Due to the influence of binding strength (or cementing strength) on the fracture communication effects between a hydraulic fracture and the beddings, the opening region of the beddings takes the shape of an ellipse.

Integrated simulation of vertical fracture propagation induced by water injection and its borehole electromagnetic responses in shale gas systems

We analyze fracture propagation induced by hydraulic fracturing with water injection and examine their detectability with crosswell electromagnetic (EM) geophysical methods. For rigorous 3D coupled flow-geomechanical modeling, we employ a numerical method that can model failure by tensile and shear stresses, dynamic nonlinear permeability, dual continuum approach, and thermo-poro-mechanical effects. From numerical simulation, we find that the fracture propagation is not the same as propagation of the water front, because fracturing is governed by geomechanics whereas water saturation is determined by multiphase flow. At early times, the water front is almost identical to the fracture tip, suggesting that the fracture is mostly filled with the injected water. However, at late times, movement of the water front is retarded compared to fracture propagation, yielding a significant gap between the water front and the fracture top, filled with reservoir gas. During fracture propagation, the coupled flow-geomechanical models are transformed via a rock-physics model into electrical conductivity models. We employ a full 3D finite-element EM geophysical simulator to evaluate the sensitivity of the crosswell EM method to fracture propagation. It is shown that anomalous distribution of electrical conductivity is closely related to the injected water saturation, but not closely related to newly created unsaturated fractures. Our numerical modeling experiments demonstrate that the crosswell EM method can be highly sensitive to electrical conductivity changes that directly indicate the migration pathways of the injected fluid. Accordingly, the EM method can serve as an effective monitoring tool for monitoring the injected fluids during hydraulic fracturing operations.

Analysis of hydraulic fracture initiation and vertical propagation behavior in laminated shale formation

The extent of hydraulic fracture vertical propagation extent is significant in evaluating simulated reservoir volume for laminated shale reservoirs. Given that it is affected by the discontinuities (beddings, natural fractures, and other factors), fracture geometry is complex in the vertical plane and is different from a simple fracture in a homogeneous sandstone reservoir. However, the propagation mechanism of hydraulic fracture in the vertical plane has not been well understood. To clarify this mechanism, several groups of large-scale tri-axial tests were deployed in this study to investigate the fracture initiation and vertical propagation behavior in laminated shale formation. The influences of multiple factors on fracture vertical propagation were studied. The results showed that hydraulic fracture initiation and propagation displayed five basic patterns in the vertical plane of laminated shale formation. The ultimate fracture geometries could be classified into four categories: simple fracture, fishbone-like fracture, fishbone-like fracture with fissure opening, and multilateral fishbone-like fracture network. Furthermore, the favorable geo-stress conditions for forming the complex fracture network were as follows: vertical stress difference close to 6MPa and vertical stress difference coefficient from 0.2 to 0.5. In addition, when q·μ-value (the product of injection rate and fracturing fluid viscosity) was roughly 3×10−9, a complex fracture geometry of fishbone-like fracture with bedding opening was formed; however, extremely small or extremely large values were both harmful. Variable injection rate fracturing with low viscosity fracturing fluid of 3mPa·s was proved to be an effective treatment to improve the connectivity of induced hydraulic fracture with the discontinuities. Moreover, because of the influence of cementing strength on fracture communication effects between hydraulic fracture and the beddings, the overall propagation region generally displayed an ellipse in shape with beddings opening asymmetrically along two wings of the main hydraulic fracture.

Numerical analysis of fracture propagation during hydraulic fracturing operations in shale gas systems

We perform numerical studies on vertical fracture propagation induced by tensile hydraulic fracturing for shale gas reservoirs. From the numerical simulation, we find that tensile fracturing occurs discontinuously in time, which generates saw-toothed responses of pressure, the fracture aperture, and displacement, and that fracture propagation is sensitive to factors such as initial condition of saturation, a type of the injection fluid, heterogeneity, tensile strength, elastic moduli, and permeability models. Gas injection induces faster fracturing in shale gas reservoirs than water injection, for the same mass injection, because of high mobility of gas. However, water injection to highly water-saturated formations can contribute to fast pressurization and high mobility of water, resulting in large fracturing. For moderate initial water saturation, complex physical responses within the fracture result from strong nonlinear permeability and multiphase flow with gravity.Pressure diffusion and pressurization within the fracture are also affected by permeability. High intrinsic and high relative permeabilities result in fast fluid movement of injected fluid, followed by fast fracturing. High Young׳s modulus and high Poisson׳s ratio do not seem favorable to fracture propagation, although they are not significantly sensitive. For heterogeneity, a geological layer of high strength between near surface and above the shale gas reservoirs can prevent vertical fracture propagation, changing the direction of fracturing horizontally.

Experiments of vertical fracture propagation based on the digital speckle technology

Abstract The vertical crack propagation behavior for sandstone/mudstone layered specimen made by the similar material was analyzed. The stress and strain field was obtained using Digital Speckle Correlation Method to analyze the propagation characteristics of vertical cracks. The test shows that the higher the strength of the interface layer, the more susceptible the crack to deflect from the side of the low-strength than from the side of the high-strength of the material. It is easy for the interface to unglue when the interfacial adhesion strength is weak, which results in the dislocation propagation after the crack passing the interface whether crack extending from the side of high-strength material or the side of low-strength material. The longer the expansion path is before the crack passing the interface, the more likely it deflects after passing the interface. Similarly, the shorter the initial prefabricated crack is, the more susceptible to deflect after the crack passing the interface. The main factor of the deflection is the difference in the properties of the composite materials, resulting in a larger shear strain of the interfacial layer. The shear strain can induce type II shear fracture, so it is the main factor leading to crack deflection. The bigger the shear deformation is, the more intensive the crack deflection is.

Development of the T+M coupled flow–geomechanical simulator to describe fracture propagation and coupled flow–thermal–geomechanical processes in tight/shale gas systems

We developed a hydraulic fracturing simulator by coupling a flow simulator to a geomechanics code, namely T+M simulator. Modeling of the vertical fracture development involves continuous updating of the boundary conditions and of the data connectivity, based on the finite element method for geomechanics. The T+M simulator can model the initial fracture development during the hydraulic fracturing operations, after which the domain description changes from single continuum to double or multiple continua in order to rigorously model both flow and geomechanics for fracture–rock matrix systems. The T+H simulator provides two-way coupling between fluid-heat flow and geomechanics, accounting for thermo-poro-mechanics, treats nonlinear permeability and geomechanical moduli explicitly, and dynamically tracks changes in the fracture(s) and in the pore volume. We also fully account for leak-off in all directions during hydraulic fracturing.We first test the T+M simulator, matching numerical solutions with the analytical solutions for poromechanical effects, static fractures, and fracture propagations. Then, from numerical simulation of various cases of the planar fracture propagation, shear failure can limit the vertical fracture propagation of tensile failure, because of leak-off into the reservoirs. Slow injection causes more leak-off, compared with fast injection, when the same amount of fluid is injected. Changes in initial total stress and contributions of shear effective stress to tensile failure can also affect formation of the fractured areas, and the geomechanical responses are still well-posed.

Distributed fracturing affecting isolated carbonate platforms, the Latemar Platform Natural Laboratory (Dolomites, North Italy)

Isolated carbonate platforms are highly heterogeneous bodies and are typically composed of laterally juxtaposed first order domains with different sedimentological composition and organization, i.e. a well-stratified platform interior, a massive margin and a slope with steeply dipping and poorly organized layers. Once subjected to overburden and tectonics forces, directions and magnitudes of principal stresses inside such heterogeneous bodies are expected to deviate significantly from the regional trends potentially resulting in non-obvious fracture patterns. In addition, characteristics such as fracture height and spacing distances are likely to change moving from one domain to the other as well as through time.The Middle Triassic Latemar platform (Dolomites, N Italy) is affected by ubiquitous fractures, typically joints. Orientation and height of more than 1500 fractures from 33 outcrops have been measured and their spatial characteristics were assessed with dedicated software. Fractures are organized in two directional sets, NNW–SSE and ENE–WSW. The orientation of each set is constant throughout the platform, showing that the trajectories of causative stresses were not influenced by the platform architecture. Fracture spacings and heights of each set change moving from one domain to the other. In the well-bedded platform interior, fractures are generally <2 m tall and have spacing distances <1 m controlled by the thickness of the hosting fracture unit and its dominant lithology. Only <50% of the fractures terminates at ±2 cm from the strata boundaries with grainstones–grainstones contacts being the most effective in preventing vertical fracture propagation. In the platform margin, fractures are up to 10 m tall and spacing distances are typically <5 m. In the massive, poorly organized slope fractures are ca. 10 m tall and have spacings of 5–10 m.There are also significant differences in the characteristics of fractures of the two sets. In all platform domains, fractures of NNW–SSE set are less tall and more narrowly spaced than those of the ENE–WSW set. In the platform interior fractures of the first set have a higher tendency to be confined to fracture units than those of the second set. These differences are associated with diagenesis-driven changes in bulk mechanical properties of the platform during its subsidence to maximum burial depth.

Techbits: Integrated Water-Management Strategies Explored in Scotland (January 2010)

Techbits Edinburgh was the host of an SPE Applied Technology Workshop (ATW) dedicated to addressing the growing global challenges of effective water management in the oil field. Approximately 50 attendees representing production technologists and reservoir, processing, and petroleum engineers attended the ATW titled “Integrated Water Management: From Planning to Operations” and reviewed the latest technologies and strategies for maximizing value through improved water management at the field-planning and operational stages of an asset’s life cycle. Field-Planning Stage Michael Hannan of the UK’s Department of Energy and Climate Change (DECC) first discussed the regulator’s role in reservoir preparation. He stressed that a regulator must ensure that an operator’s field plans include processes for maximizing economic recovery factors and for complying with environmental laws. The DECC expects operators to have a thorough understanding of voidage and sweep in the reservoir and design wells and facilities such that recovery can be maximized. The operator’s plans should also be flexible enough to handle any unforeseen challenges associated with water handling. On the regulatory front for marine discharge of water, Hannan encouraged operators to keep abreast of future risk-based discharge limits being proposed by OSPAR—the north-east Atlantic convention that sets limits for marine discharges. In his presentation titled “Planning Water Management,” Owen Vaughan of Apache Corporation said that despite the importance of planning in effective water management, conflict may arise within an organization because planning has different meanings for different people, depending on their background and experience. He also discussed the impact of data quality on the planning process, saying that for greenfield developments the quantity of data may be insufficient to allow a comprehensive water-injection plan to be developed. Conversely, for brownfields there may be such a volume of data that much of its quality may be questionable. A discussion followed amongst the attendees on the pros and cons of produced water reinjection (PWRI) vs. water disposal. Participants were asked to take a stance on one of the methods with no middle ground and then list the relative benefits of their method. Among the benefits that the proponents of water disposal listed were the abundant availability of affordable technology to allow for disposal at sea and the lower capital and operating expenditures involved. Reinjection proponents cited the positive environ-mental image of injecting water back into the reservoir from which it came, as well as avoiding the need to chemically and mechanically clean the water for discharge. Another reinjection debate followed on whether to inject water above or below the fracture gradient. Those supporting injection above the fracture gradient pointed to the benefits of an easier-to-manage injectivity index, lower capital expenses owing to smaller pumping equipment, and design advances allowing for better sweep control. Those arguing for injection in the matrix highlighted the benefits of a better injection profile, a diminished risk of vertical fracture propagation, and lower injection pressures resulting in better vertical sweeps.

Reply

We thank Dr. Batu for his comments on our paper (Eaton et al. 2007) on three-dimensional ground water flow in the Maquoketa Aquitard. We agree that the classic analytical methods for leaky aquifer analysis, developed by Neuman and Witherspoon (see reference citations given by Batu) and others, are still relevant. However, indurated and fractured rock aquitards like the one we discussed present additional complexities for flow and hydrogeologic characterization. The objectives of our paper were to highlight these complexities and present a conceptual model that provides a better understanding of flow in these systems, which are very different from the unlithified, less heterogeneous aquifer-aquitard systems for which the classic analytical solutions were developed. We welcome the opportunity to provide additional discussion of our work. Dr. Batu discusses mainly the vertical flow assumption in aquitards, analytical solutions, aspects of the MODFLOW numerical code, and steady-state vs. transient conditions of the hydraulic head data. Notably absent from his comment is one of the main emphases of our work, which is that layered sedimentary bedrock is typically fractured. Preferential flow occurs along bedding-plane fractures; the mechanical stratigraphy inhibits propagation of vertical fractures. Although we refer to the classic aquitard framework to understand the bulk properties of the Maquoketa Formation, we also emphasize the role of bedding-plane fractures in the three-dimensional flow system (Figures 2 and 10 in Eaton et al. [2007]) and present an innovative application of thin model layers in our MODFLOW model (Table 1 in Eaton et al. [2007]). The presence of bedding plane fractures is supported by ground water geochemical data (Figure 7 in Eaton et al. [2007]). In response to Dr. Batu’s specific points, the tendency toward aquitard vertical flow due to hydraulic conductivity contrasts in layered systems is elementary both for analytical and numerical solutions, but our work adds to this framework the effect of the bedding plane fractures we describe. Of course, we recognize that analytical solutions for leaky aquifers are valid for both upward and downward flow through the aquitard and that both upward and downward flow occurs under field conditions (Figure 5 in Eaton et al. [2007]). We again refer interested readers to Eaton and Bradbury (2003), in which there is additional discussion of monotonic trends in hydraulic head profiles under transient conditions. We also recognize that the analytical solutions for leaky aquifers assume horizontal flow in the aquifers. We point out again that the main focus of our work is to present a conceptual model for three-dimensional flow in an aquitard. The possibility of three-dimensional flow in aquitards was recently recognized and analyzed analytically by Sepulveda (2008). Finally, we point out the distinction between steady-state conditions, which have apparently not been reached in the Maquoketa Aquitard, and equilibrated long-term values for the purpose of field data collection. As we stated, the hydraulic head data we used in our analysis of the Maquoketa Aquitard were collected over a period of years, not months.

Hydraulic fracturing and wellbore completion of coalbed methane wells in the Powder River Basin, Wyoming: Implications for water and gas production

Excessive water production (more than 7000 bbl/month per well) from many coalbed methane (CBM) wells in the Powder River Basin of Wyoming is also associated with significant delays in the time it takes for gas production to begin. Analysis of about 550 water-enhancement activities carried out during well completion demonstrates that such activities result in hydraulic fracturing of the coal. Water-enhancement activities, as the operators in the basin call this procedure, consists of pumping 60 bbl of water/min into the coal seam during approximately 15 min. This is done to clean the wellbore and to enhance CBM production. Hydraulic fracturing is of concern because vertical hydraulic fracture growth could extend into adjacent formations and potentially result in excess CBM water production and inefficient depressurization of coals. Analysis of the pressure-time records of the water-enhancement tests enabled us to determine the magnitude of the least principal stress (S3) in the coal seams of 372 wells. These data reveal that because S3 switches between the minimum horizontal stress and the overburden at different locations, both vertical and horizontal hydraulic fracture growth is inferred to occur in the basin, depending on the exact location and coal layer. Relatively low water production is observed for wells with inferred horizontal fractures, whereas all of the wells associated with excessive water production are characterized by inferred vertical hydraulic fractures. The reason wells with exceptionally high water production show delays in gas production appears to be inefficient depressurization of the coal caused by water production from the formations outside the coal. To minimize CBM water production, we recommend that in areas of known vertical fracture propagation, the injection rate during the water-enhancement tests should be reduced to prevent the propagation of induced fractures into adjacent water-bearing formations. In areas where S3 is unknown, a minifrac should be done to determine the magnitude of S3 (to know whether fracture propagation will be vertical or horizontal), so the water-enhancement activities at the time of well completion are done to minimize water production and optimize gas production.

Effective Hydraulic Fracturing Of The Lower Amaranth Formation In Southern Manitoba

Abstract The Spearfish, or more properly, Lower Amaranth Formation in southwest Manitoba is a low permeability sandstone reservoir containing low GOR light oil. Wells drilled into this formation are generally not capable of commercial production until after a hydraulic fracturing stimulation. Factors such as an underlying water zone, along with an unconfined producing interval, have resulted in numerous hydraulically fractured Lower Amaranth wells producing water rather than-oil, limiting potential field development. Following the unsuccessful stimulation of several wells in the South Pierson field where hydraulic fractures propagated into the underlying water zone, a comprehensive re-evaluation and detailed design effort was implemented to minimize the potential for water production. The project incorporated geological studies; reservoir rock elastic properties determination and correlation with wireline logging techniques; determination of horizontal in situ stresses; reservoir performance testing; in situ fracture fluid performance evaluations; and two- and three-dimensional fracture geometry modelling techniques. It has resulted in new treatment designs incorporating fracture diversion anti-net pressure control techniques. This paper discusses the general stratigraphy of the Lower Amaranth zone and the completion requirements. Presented are the design concepts, data analyses and results of a number of Lower Amaranth Formation stimulations. Optimal treatment considerations and guidelines are provided to minimize the probability of water production following hydraulic fracture stimulation. Introduction Hydraulic fracturing has long been the most economic technique for the purpose of stimulating low permeability reservoirs and providing for commercial production of oil and gas. The success of fracture stimulation treatments depends primarily on achieving large contact area with the production interval and providing a large permeability contrast and high flow capacity within the fracture. Ideally, the hydraulic fracture created should extend laterally within the zone of interest, however, it is well known that substantial vertical fracture propagation may also occur, significantly impacting the success of the treatment. Complicating factors such as underlying water zones or overlying gas sections can be easily penetrated and subsequently reduce or eliminate any sought after oil production. BenNaceur and Touboul(12) considered the problem of a vertical fracture intersecting a formation with the underlying water zone. Several factors have been shown to influence the production of such a system, including the fracture extent and its distance from the water transition zone, its dimensionless conductivity, the vertical permeability or absence thereof, the relative permeability curves, and finally the mobility ratio between oil and water. It was shown that in addition to the benefit of increasing productivity, hydraulic fracturing resulted in a significant delay of the water breakthrough, even with values of vertical permeability being one-tenth the horizontal one. The major concern in the current study was to avoid any fracture propagation through an underlying low-permeability interval downward into a porous dolomite section that would have resulted in water breakthrough using the fracture as a flow path. Additional motivation for control of fracture height growth was to increase the effectiveness of the treatment by decreasing the extension into non-producing intervals and to reduce the over-all cost of the treatments.

Analysis of Coalbed Methane Stimulations in the Warrior Basin, Alabama

Summary An objective of the U.S. DOE's in-house coalbed methane R&amp;D effort has been to improve coalbed hydraulic fracturing techniques by improving the understanding of stimulation treatments. Current efforts focus on examining data from coal mine excavations to observe stimulation results, conducting Nolte-Smith analyses to characterize fracture behavior, and exercising DOE/METC's stimulation modeling codes. Test Wells DHM-6 and DHM-7, two research degasification wells located in the Oak Grove field, AL, were analyzed in this project. METC's hydraulic fracture models predicted Well DHM-6 coalbed wing lengths 3 to 5% greater than the excavated wing length. This modeling was based on observation of fluid penetration into the shale roof rock and on guidelines derived from the Nolte-Smith method, which indicated vertical propagation into the bounding strata. Results to date indicate that the Nolte-Smith method may provide adequate interpretation of coalbed fracture behavior when the effects of variable injection rates, natural fracture fluid interaction, and the interface effects of coalbed and bounding formations are considered. Further evaluation of hydraulic fracture models that predict the effects of geologic discontinuities on coalbed fracture propagation is recommended. Additionally, a future mine excavation to expose the coalbed roof rock and to assess the extent of vertical fracture propagation is suggested.

Changes in the tooth enamel of early Paleocene mammals allowing increased diet diversity.

Hunter-Schreger bands (HSB) are seen in teeth that are composed of crossed sets of enamel prisms. They are present in the teeth of man and many other mammals, but absent in most insectivores and multituberculates. It has been suggested that the presence of HSB makes the tooth enamel less likely to split and is associated with chewing ability. We have traced the occurrence of HSB back to the arctonic condylarths of the early Palaeocene (Puercan) age; this must be close to the first appearance of the bands in placental mammals. Our data indicate that the teeth of almost all large mammals since the early Palaeocene have contained these bands, in an orientation that is optimal for limiting the propagation of vertical fractures. The appearance of the bands is associated with the differentiation of herbivores and carnivores from insectivores and our data indicate that their development was critical to the diversification of mammals because it allowed the use of new types of foods.

Three-Dimensional Simulation of Hydraulic Fracturing

Summary A comprehensive three-dimensional (3D) simulator of hydraulic fracturing has been developed. The model formulation couples 3D, two-phase flow in the reservoir with a 3D fracture model in a vertical plane, proppant transport, and heat transfer. The numerical implementation is in finite differences that use several integrated grid systems. The model is computationally efficient and greatly increases the realism of modeling compared with two-dimensional (2D) simulators. The new features important for applications include treatment of proppant transport and closure process, simulation of fracture growth after shut-in, and in particular the vertical fracture propagation, which gives the model the capability to predict fracturing pressure increase or decrease. predict fracturing pressure increase or decrease. Examples show the sensitivity of the model to design parameters and properties of confining strata, and give parameters and properties of confining strata, and give comparisons between simulations that use different models of fracture geometry, Introduction Modeling of hydraulic fracturing has progressed recently, spurred by the increased importance of well stimulation and other fracturing applications in the field. The historical background and survey of the field is included in Refs. 1 and 2. Developments in computation of fracture geometry are discussed in Ref. 3, which presents a comprehensive 3D model of fracture geometry that is practically oriented and efficient. practically oriented and efficient. The approach taken is the same as in the previous 2D model described in Refs. 1 and 2--i.e., simultaneous modeling of the fracturing process, reservoir fluid flow and heat transfer, and proppant transport. The 2D model includes much of the physics necessary for realistic simulation of fracturing treatments, especially in low-permeability reservoirs, as discussed in Ref. 2. However, it has a weakness (common with other existing industrial models) of treating the fracture geometry as 2D--i.e., with separately determined fracture height. The model described here represents a major advance because it determines simultaneous lateral and vertical fracture evolution coupled with leakoff and subsequent production from a 3D reservoir. A new concept of a production from a 3D reservoir. A new concept of a pseudo-3D hydraulic (P3DH) fracture geometry has pseudo-3D hydraulic (P3DH) fracture geometry has been developed and integrated within the overall model. The resulting scheme for calculation of fracture geometry is described in detail in Ref, 3. It is computationally very fast and, when combined with efficient integration of the components of the overall model, results in a practical 3D simulation tool. This paper describes the formulation and integration of the overall model and deals in detail with some of its components that have not been described before. The treatment of the physics of filtration, fracturing fluid rheology, reservoir flow, etc. are not repeated here. Although the model can treat both vertical and horizontal fractures, we deal here only with fractures extending in a plane perpendicular to the reservoir bed. An example of an application including horizontal (pancake-type) fracture is found in Ref. 4. The physical and numerical formulation of the model is presented in the first two sections with the exception of the proppant transport and fracture-closure model. These parts of the system are coupled only weakly and for parts of the system are coupled only weakly and for clarity are dealt with in a separate section. The capabilities of the model are illustrated by examples of fracturing treatment simulation using both 2- and 3D fracture propagation. propagation. Model Formulation The model formulation consists of a set of equations describing (1) the 3D, two-phase flow in the reservoir, (2) two-phase non-Newtonian flow in the fracture, (3) heat transfer in the fracture and in the reservoir, including overburden losses, (4) lateral and vertical propagation of the fracture (resulting in a general fracture propagation of the fracture (resulting in a general fracture shape), (5) proppant transport, and (6) wellbore hydraulics. Additional relations describe the PVT and rheology behavior of reservoir and fracturing fluids, the physics of filtration on fracture walls, interaction of solid physics of filtration on fracture walls, interaction of solid stress and fluid flow, etc. The general physical configuration of the model is shown in Fig. 1. The direction of the lateral propagation of the fracture can be specified by giving the path of the leading edge--which can be independent of the bedding plane of the reservoir--and its height. Given this constraint, the fracture can then propagate arbitrarily through the reservoir and the adjoining propagate arbitrarily through the reservoir and the adjoining strata, as dictated by their elastic properties and confining stress. As a special case, the geometry model can be used without fracture height growth with generalized Perkins-Kern (PK) or Christianovich-Geertsma-DeKlerk (CGD) Perkins-Kern (PK) or Christianovich-Geertsma-DeKlerk (CGD) type geometries (Fig. 2). Fluid Flow and Heat Transfer. The fractured well is assumed to be at the origin of the x-y, plane and the fracture propagates along the x-z plane. JPT P. 1177