Modelled Stress perturbation at Cooper basin fault tips

Ehtesham Karatela,SPE, Department of civil, mining and environmental engineering-University of Adelaide

Abstract: Tight reservoirs such as shale gas, shale oil, coal seam gas (CSG) and tight sands require fracture stimulation to produce at economic rates. Experience shows that production rates from fracture stimulated wells can be highly variable even between closely spaced wells in the same field. Possible explanations for production variation include local changes in stress and changes in natural fracture orientation and density. (Reynolds, Mildren et al. 2006) documents stress magnitudes at different depths in the Cooper Basin, Australia. This study utilizes stress data from (Reynolds, Mildren et al. 2006) to create number of stress simulation models using Poly 3D. Stresses are usually considered on a regional scale and the influence of local factors is usually oversighted. This study determines the impact of variations in maximum horizontal stress (SHmax) magnitude and explains the possible fracture propagation near faults. Strike-slip stress regime prevails at the selected depth interval. A thorough study using different lithologies, σH azimuth and fault size is carried out. Stress concentration at the fault tips increases with fault size. However, the intensity of fluctuation appears to diminish after fault size of 1500 meters. These models help in understanding the orientation of fractures during hydraulic fracturing and help to recognize stress barriers that may limit the drainage area from an unconventional reservoir. Introduction During last few decades the need of energy is more intensified. As major conventional reservoirs of the world are at the verge of depletion the oil and gas industry is considering various ways to exploit tight reservoirs such as shale gas, shale oil, tight sands and coal seam gas (CSG). But unlike the most conventional reservoirs these reservoirs need to be hydraulically fractured. The process is carried out by pumping specially engineered fluids at high pressure into the reservoir interval to be treated, causing a fracture to prop open (Holditch 2007). All fracture simulation treatments are not always successful, potentially due to number of reasons explained by (King 2010). (Baihly, Malpani et al. 2010), observed production variability in a number of shale gas plays, across North America. One of the possible reasons in the actual production and predicted production from wells is the stress around well bore which can control the initiation and development of fracture (King 2010). The notable development in determining global stress magnitude and orientation has led to use this vital information in exploration and production process. The world stress map project has prepared an extensive database of stresses around the world, using a number of conventional methods such as borehole breakouts and minifrac test (Tingay, Müller et al. 2005). However, variation in stress magnitude due to local geologic structures in sedimentary basins is poorly understood (Tingay, Müller et al. 2005). Cooper basin is the most prolific onshore basin in Australia. Due to intense exploration ample amount of stress data sets in the form of borehole breakouts, drilling induced tensile fractures (DITF), overcoring measurements and earthquake focal mechanism is available. The in situ stress field plays an important part in not only determining the orientation of new fracture but also define the fractures that may be more vulnerable to flow in a naturally fractured reservoir (Reynolds, Mildren et al. 2004). Therefore, this study attempts to simulate stress behaviour around faults with varying fault size, lithology and Maximum horizontal stress (σ H) direction. The

with major part lying in Queensland and other portion lying in South Australia.simulations are made using Schlumberger’s stress simulation package Poly 3D which is based on Boundary element method (BEM).9 mmstb) of recoverable oil have been found in the Cooper basin (Gravestock. Cooper basin The Cooper Basin is Australia’s most proficient onshore basin Fig 1. Prior to the formation of Cooper basin. Mildren et al. Figure 1. 2006) . 1998) Extensive drilling in the Cooper basin has provided substantial amount of stress data. 1997). plan view. 229 x 109 m3 (8.9*106 kL (43. maximum horizontal stress magnitude can be loosely confined regionally (Reynolds. Due to variability of σ hmin and σv estimates. 2006). Hibburt et al. Since 1963. It is northeast-southwest trending basin located in the central Australia. a number of orogenies resulted in intense deformation in the region causing crustal shortening over south-eastern Australia and eastern central Australia. Vertical stress magnitude (σv) is equivalent to the overburden of the rock at a particular depth. Regional map showing location of Cooper basin from (Gravestock. Well Dullingari North-8 from (Reynolds. Mildren et al. it is true for rocks at greater depths (Zoback 2007). 1998). The results are presented in two ways.2 tcf) of recoverable gas and 6. Minifrac test provide for the magnitude of minimum horizontal stress (σ hmin) indicating magnitude approximately equal to the magnitude of σv. 2006) has been used as the basis of stress input. leading the stresses to be transmitted in the basin area (Apak. Hibburt et al. The software use average values of Young’s modu lus (YM) and Poisson’s ratio (PR). which shows colour contours of σH and a vertical view of fault expressing the perturbation of stress across fault tips. Stress magnitudes have been calculated by (Reynolds. Stuart et al. Mildren et al.

Map representing orientation of maximum horizontal stress in the Cooper basin. Mildren et al. Therefore. The average σHmax orientation from all wells determined by borehole breakouts and Drilling induced tensile fractures (DITF) is 101˚ (Reynolds. For a successful fracture treatment. understanding these stress perturbations is very important. 2005). Geologic and geomorphologic features also affect trend of σHmax. fractures may grow tens of meters and perhaps encounter a natural fracture. . Geomechanical modelling of such geologic structures can reveal important information helpful to develop an oil and gas field more competitively. 2005) North-south oriented σHmax in the Amadeus basin. Therefore. Stress data from Patchawarra trough indicate southeast-northwest orientation. 2005). Mildren et al. localized orientation of σHmax direction in different parts of the basin is affected by regional in situ stress. The fracture stimulation treatment may cause shear failure in the natural fracture if the natural fracture is critically stressed. σHmax direction at Gidealpa-Merrimelia Innamincka (GMI) rigde is west-northwest which changes to east-west at Nappamerri trough (Figure 2) Figure 2. 2005) used datasets from 61 wells to interpret the orientation of maximum horizontal stress. 1998).Stress orientation and magnitude (Reynolds. Meyer et al. Mildren et al. (Reynolds. Mildren et al. (Hillis. located north of Cooper basin and the existing eastwest orientation of σHmax shows the systematic clockwise rotation of σHmax orientation is part of regional rotation across the Australian continent (Reynolds. Methodology Tight reservoirs need to be hydraulically fractured to flow at economic rate. Faliure of these structures may disrupt the fracture pattern which limits the conduit system of fracture pattern. It uses wells ranked A to C quality determined by World Stress Map (WSM) project ranking scheme.

Thomas 1993). we can model rock volume as an infinite or semi-infinite elastic mass (Lorig and Brady 1982. lithology and σ Hmax azimuth. Modelling Approach Boundary element method (BEM) BEM is a numerical method used by engineers for modelling purposes. A geological surface divided into triangular polygonal elements . Pecher and Stanislav 1997). it represents a quasi-infnite domain in terms of internal surface geometry and boundary conditions. The boundary element method has been part of mathematical literature for a long time but had never applied to computer geomechanical simulation until research efforts at Stanford University (Pecher and Stanislav 1997). This study consists of stress modelling results in the form of plane view and line plots.” A geological surface is divided into small polygonal elements across which the discontinuity is in displacement is assumed constant (Thomas 1993) Polygonal elements may have minimum of 3 sides as used in this thesis (Fig 3). The study was formulated in form of a computer program namely Poly 3D by (Thomas 1993) which helps to get precise solution of stress and strain estimated at observation points in the surrounding volume using linear elastic properties (Swyer and Davatzes 2012).or half-space by planar. 1993). Modified from Swyer and Davatzes (2012). It is worth emphasizing that the primary aim of the project is to predict and quantify abnormal stress change at the fault tips using the limited knowledge of stress. hence. Furthermore. The sensitivity to results is achieved due to individual role of the polygonal elements. strains and stresses induced in an elastic whole. It is prominent that BEM offers distinctive advantages in simulation. The conclusions from the study are limited to the average rock properties such as Poisson’s ratio and young’s modulus.This study uses Poly 3D software which is based on the boundary element method. For example it lessens the spatial dimension of the problem. “Poly3D is a C language computer program that calculates the displacements. It provides much better results compared to other numerical modelling methods (Pecher and Stanislav 1997). Figure 3. preserving high accuracy (Fu 1996. Poly 3D (Thomas 1993) states. It efficiently computes 3D loading conditions representing any tectonic regime. The study also presents extent of fault size after which stress perturbation is negligible. This approach allows evaluation of a very large number of models and quantitative assessment of stress disturbance around the fault tips. The user can select the number of elements to divide a fault or fracture. These polygonal elements can be used to model complex geologic structures with bending surfaces (Thomas. The modelling results represent that the local stress perturbation can be a function of fault size. polygonal-shaped elements of displacement discontinuity.

Model information A significant number of simulations were created using Poly 3D with varying fault size ranging from 100 to 2000 meters. Therefore. 45. 60 and 90 as σ H azimuth . 30. which is defined as a series of equally spaced observation points and instructs Poly 3D to estimate stress. Each model with different lithology was further with the azimuth between fault and σ Hmax of 0˚.In Poly 3D. 45˚. Therefore. 2006) as per table 5. Simulation results vary not only with lithology but also as we change the angle between the fault and σ Hmax. The former observation grid consists of 400 nodes and later is composed of 40 nodes. One encompassing the fault. while both having similar number of nodes (20) on X axis. models with fault length 1900 and 2000 meters have model length of 2100 and 2200 meters. Mildren et al. traction on an element is defined through determining any remote stress in addition to the total stress field induced by all polygonal planes on the element plane (Thomas 1993). Shale and Coal. 15˚. 5. 60˚ and 90˚. Magnitude of σHmax. strain and displacement at individual points. 30˚.slip stress regime 2. To prevent any perturbation of stress due to model edges. In the Cooper basin a strike-slip stress regime prevails from 1 to 3 km. A comprehensive tree explaining the structure of model formulation is described in Fig 4 Strike. Separate models using similar amount of stress data were created for Sandstone. Mildren et al. Tree representation of modelling paradigm.1. each segment acting as individual element when running the simulation. 5˚. The element plane collectively forms an observation grid. Each model consists of constant volume and rock properties with varying length of faults. σv and pore pressure (Pp) for the three selected depth points is extracted from (Reynolds. All models were assigned an average Young’s modulus and Poisson’s ratio as per lithology assuming no change in elastic properties of rocks throughout the model. 15. the magnitude of stresses used in simulations represent strike-slip stress regime. Change of fault size does not affect the model as the number of nodes on horizon is always constant. σhmin. The simulation grid is divided into small triangular segments. 2006). The far field stress magnitude used for the simulation purpose was extracted from (Reynolds. displays stress change across the entire fault length while second observation grid depicts stress variation at the fault tips (Fig 5a). Each model consists of a near vertical strike-slip fault and two observation grids (Fig 5a).25 km Sandstone Shale Coal 2.6 km Sandstone Shale Coal Fig 4.5 km Sandstone Shale Coal 2. Note: each lithology is further divided into seven different models with 0. the edge of the model is kept 200 meters from the fault top.

Therefore.8*106 and 5*109 Pascals respectively. Mildren et al.35 respectively while Young’s modulus used in these simulations for Sandstone.2* 106. The dark line passing through fault tip is the smaller observation grid. Each model is composed of an absolute scale representing minimum horizontal stress (Shmin). 2. 45. Use of fault is intended is to understand the abrupt change of stress across the fault.14 and 0. Simulation results presented in this thesis align with the hypothesis of (Gudmundsson 2000). Each of the factors has its impacts on stress perturbation. The stress magnitude in the Cooper basin has high differential stress.24. Data represented in table 5. 15. 60 and 90 as σH azimuth It should be noted that the faults in these simulations are hypothetical.1 is evident of elevated stress in the upper crust responsible for the transfer of stress intraplate (Reynolds 2001. Mildren et al. One. 2006) Poisson’s ratio for Sandstone. stress concentrates in two quadrants on the opposite ends of fault tips. σH azimuth and fault size. Shale and Coal as the candidate for hydraulic fracture treatment. Note: each lithology is further divided into seven different models with 0. no lateral variation due to change in rock properties is expected. Modelling Results Examination of results from the model runs indicated that stress variation is the function of lithology.25 2. the impact of each of these properties is presented in a separate section. Shale and Coal is 0. The Poly 3D results are displayed in two forms. Reynolds. these models may also be applicable to similar conditions worldwide.For the sake of demonstrating the variation due . Therefore. Shale and Coal is 2.6 62 100 110 Stress Magnitude (MPa) Minimum Horizontal stress 32 50 52 47 55 58 20 22 24 Vertical stress Pore pressure Table 1. a graphical representation in the form of line plot explaining abrupt change in stress magnitude across fault tips. Tree representation of modelling paradigm. Fig 5a. but certainly realizable. Following sub sections represent outcomes of the model run on the basis of above give criteria. 5. Lithology Each Lithology used in modelling has average elastic modulus. 0.5 2. Second. 2006). Depth (Km) Maximum horizontal stress 2. Therefore. 5b and 5c represent model run with Sandstone. around 50 MPa (Reynolds. map view of minimum horizontal stress magnitude (Shmin) depicted in colour with vectors determining the orientation of S 1. 30. According to (Gudmundsson 2000).Stress magnitude for each depth is compiled in table 1.

with vectors depicting orientation of S1 shale at 2. with vectors depicting orientation of S1 for Coal at 2.5 km azimuth 30˚ fault 600. Fig 5a Plane view near of vertical fault representing minimum horizontal stress (S3) in colour contours.5 km σH azimuth of 30˚ and fault length of 600 meters.to litholgy. 200 meters 17 37MPa Fig 5b Plane view near of vertical fault representing minimum horizontal stress (S3) in colour contours. 5b and 5c represent minimum horizontal stress at 2. 200 meters 15 43 MPa .5 km depth and fault size 600 meters. Models 5a. fault size and stress magnitude is kept constant.

00E+00 0. graphical representation of rapid change in stress magnitude of minimum horizontal stress (S3) at the fault tip for Shale.00E+01 2.50E+01 2.5 km σH azimuth of 30˚ and fault length of 600 meters.00E+01 1.00E+01 3.5 km σH azimuth of 30˚ and fault length of 600 meters.50E+01 1.Fig 5c Plane view near of vertical fault representing minimum horizontal stress (S3) in colour contours.00E+00 -1500 -1000 -500 0 500 1000 1500 Shale Coal Sandstone Distance from fault tip Fig 5d. with vectors depicting orientation of S1 for Sandstone at 2.00E+01 5. Coal and Sandstone at 2. 200 meters 15 39 MPa σ3 4.50E+01 4. .50E+01 Stress (MPa) 3.

200 meters 18 35 Mpa . sandstone being able to distribute stress far away from the fault compare to Coal and Shale. therefore effect of direction of maximum horizontal stress is analysed by changing σ H azimuth. with vectors depicting orientation of S1 for Shale at 2.5 km depth within shale with fault size 1000 m. presented simulations utilize magnitude to stresses at 2.5 km σH azimuth of 15˚ and fault length of 1000 meters. Fig 6a Plane view near of vertical fault representing minimum horizontal stress (S3) in colour contours.2 σH azimuth Tectonic stresses influence the orientation of geologic structures.3. 6b and 6c represent simulations with orientation of maximum horizontal stress (σHmax) of 5˚. Fig 6a. Coal and Sandstone that correspond to smaller observation grid (black line in simulation model crossing through the fault tip).These results indicate prominent change in stress across the model with change in lithology. 45˚and 90˚. 5. Fig 5d is line plot for Shale. Difference in distribution of stress field is observed clearly. To purely explain the effect of angle change.

5 km σH azimuth of 60˚ and fault length of 1000 meters. 200 meters 16 37 MPa .5 km σH azimuth of 45˚ and fault length of 1000 meters. with vectors depicting orientation of S1 for Shale at 2. 200 meters 10 40 MPa Fig 6c Plane view near of vertical fault representing minimum horizontal stress (S3) in colour contours.Fig 6b Plane view near of vertical fault representing minimum horizontal stress (S3) in colour contours. with vectors depicting orientation of S1 for Shale at 2.

00E+01 1. Similar reseults were observed at different depths within Sandstone and Coal.00E+01 3. 7b and 7c represent change in stress magnitude at 2. Fig 7a.σ3 4.50E+01 2. 45˚ and 60˚ at 2. Fig 6d correspond to simulations anticipating stress varaition at the fault tips.25 km depth within shale at σH angle of 0˚.5 km in Shale and fault length of 600 meters. a detailed analysis is carried out estimating stress perturbation due to change in fault size.00E+00 0.00E+01 Stress (MPa) 2. . As demonstrated in the previous sub sections. graphical representation of rapid change in stress magnitude of minimum horizontal stress (S3) at the fault tip for σH azimuth of 5˚. litholgy and σH azimuth influence the stress disturbance around fault tips. It was observed that maximum stress perturbation exist when σH angle is 45˚. 30˚ and 60˚. Therefore. Fault size Fault size with fault poulation in a reservoir control the orientation of fracture development (Zoback 2007).00E+00 -1500 -1000 -500 0 500 1000 1500 Distance from fault tip Fig 6d.50E+01 3. both parameters are reserved stable.00E+01 5.50E+01 5 45 60 1.

50E+00 5.05E+01 8.25 km depth.20E+01 2.00E-01 -1500 200 300 500 600 800 1000 1200 1400 1500 1600 1800 -1000 -500 0 500 1000 1500 1900 2000 Distance from fault tip Fig 7a.00E+01 -1500 200 300 400 500 600 800 1000 1200 1400 1500 -1000 -500 0 500 1000 1500 1600 1800 Distance from fault tip Fig 7b. σH azimuth 5˚ .20E+01 3. Stress concentration at fault tips in Shale at 2.40E+01 100 3.80E+01 2. σH azimuth 5˚ σ3 3.25E+01 Stress (MPa) 1. Stress concentration at fault tips in Shale at 2.σ3 1.60E+01 2.40E+01 2.5 km depth.00E+01 Stress (MPa) 2.65E+01 100 1.50E+00 2.45E+01 1.50E+00 6.50E+00 4.

σ3 3.6 km depth. principal stress may align themselves parallel or perpendicular to the orientation weak planes (Zoback 2007). Hu et al.70E+01 Stress (MPa) 400 500 2.20E+01 200 300 2. 1997. Therefore. Bourne and Willemse 2001.00E+00 -1500 1900 -1000 -500 0 500 1000 1500 Distance from fault tip 2000 Fig 7c. 1996.70E+01 1000 1200 1. If no shear stress exists at the boundary. Kattenhorn and Marshall 2006). Zoback 2007). .20E+01 600 800 1. Mildren et al. Orientation of principal stresses can be affected by local geologic structures and rock properties (Reynolds.00E+00 1600 1800 2. that are discussed in this section. fractures as a result of hydraulic fracturing treatment will change their orientation affecting production from the well bore. Re-orientation of maximum horizontal stress vectors (S1) near the fault can be observed from the simulation results (Fig 8). Walsh et al. σH azimuth 5˚ Discussion Results presented in this thesis are broadly similar to previous studies (Nicol.70E+01 100 3. Stress concentration at fault tips in Shale at 2. Homberg. However significant differences also exist. 2005. Gudmundsson 2000.20E+01 1400 1500 7.

Sandstone being most brittle of the presented lithologies. therefore. Near the fault tips. According to (Kattenhorn and Marshall 2006) stress field at the fault tip is greater than the magnitude and different in orientation than regional stress and may lead to fractures if tensile strength of rock is reached. elastic moduli affect the ability of rock to transfer stress.5 km depth. S1 vectors show a prominent change in orientation. Stress concentration at fault tips in Shale at 2. stress vectors appear to align in a different orientation compare to the far field stress direction. Moreover. in fig 8 spacing in observation points is 25 meter. Therefore. the orientation of fractures is a function of σH azimuth and elastic properties of rock matrix. . Failure planes indicated in the simulations represent points where faults may initiate. Change in magnitude of σhmin can affect the propagation of fractures. fault tips can be divided into four quadrants. However. Coal usually have weak planes (cleats) and has some influence on the fractures during fracture stimulation . Each lithology has different elastic moduli that affect the magnitude of closure pressure. 1991). more brittle lithologies tend to fracture easily compare to relatively ductile formations (Lorenz. thus orientation of fractures. (Cooke 2011) explains the importance of Poisson’s ratio in measuring the magnitude for closure pressure. Teufel et al.50 meters 6 39 MPa Fig 8. two compressional and two extensional (Fig 5). However. Vectors in Figure 5 do not appear to diverge from the trend because spacing between two observation points is 200 meter. Hence. Poly 3D models possible orientation of shear fractures. Deviation of stress orientation is consistent in every model and is influenced by the orientation of far field stress. fractures more easily with high number of fractures compare to shale and Coal. σH azimuth 45˚ with horizon length of 500 meters fault size 300 meters A general observation in all models is the concentration of stress at the fault tips in opposite quadrants of the fault.

treatment. Though the orientation of fracture created during the treatment is mainly controlled by insitu stress. Hu et al. The vectors for maximum horizontal stress appear to align themselves according to the orientation of the fault. 6b. Coloured contours in figures represent minimum horizontal stress while the vectors determine the orientation of maximum horizontal stress. Figure represents line plots for stress perturbation with fault sizes ranging from 100 meters to 2000 meters. The third criteria used for simulation purpose the fault size (7a. It was observed. Cooke 2011. for the purpose of developing understanding of the behaviour of stress static properties can be used. Hu et al. author has made an attempt to predict the fault size in a given geologic condition after stress perturbation is constant and is not the function of fault size. Stress distribution illustrated in Figures 6a. For the sake of argument σH azimuth of 15˚. Fault Vs Stress 36 34 32 Stress (MPa) 30 28 26 24 22 20 0 500 1000 1500 2000 2500 Fault size Fault Vs Stress Fig 9. fracture stimulation will not be considered to be successful. Thus. 6c and 6d are in good agreement with those predicted by (Homberg. An analysis of Fault size with Stress. However. However. 1997). Detailed analysis of faults depicts rapid change in stress magnitude at fault tips as proposed by previous authors. Elastic moduli used in these models are static but this is not the case in actual practice. 45˚ and 60˚ are presented in the figures. Swyer and Davatzes 2012) have utilized average rock properties for geomechanical modelling. it influences the concentration of stresses at the fault tips. As the orientation of maximum horizontal stress is changed. Line plots start to come close after 1000 meters and the amplitude of stress change is not very significant after fault size of 1500 meters (Fig 9. increase in size of fault results in increase in perturbation of stress at the fault tips. A number of authors (Reynolds 2001. . 7b and 7c). It should be noted that stress perturbation is symmetrical relative to the centre of fault. (Homberg. The fluctuation is prominent in the models with smaller size of faults (100-1000 meters). Largest perturbation is encountered at the angle of 45˚. σH azimuth shall change locally with pre-existing faults and fractures which will serve as barrier for economical production. Fault vs stress). If a faulted reservoir is hydraulically fractured. 1997) explains the relation of σH azimuth and strike of geologic structure. S1 vectors indicate propagation in the direction of σHmax while the fractures open in the direction of S3. Rock properties vary laterally and vertically (King 2010).

6. An analysis using dynamic rock properties should be carried out to mimic the subsurface more accurately. it is obvious to encounter faults. therefore. However. The detailed analysis of models has led to greater understanding of distribution and abrupt change in stress field at the fault tips. such models allow. Cooke 2011). Conclusion A number of studies have been carried out to understand the variation in production resulting from not so successful hydraulic fracturing treatments. Production from unconventional reservoirs particularly shale gas from world’s known reservoirs is unpredictable due to various factors explained by (King 2010. therefore. Mildren et al. line plots depict that stress magnitude is not large when fault size reaches around 1500 meters. Abrupt change in magnitude can influence the orientation of fracture during fracture treatment. Fault size can be predicted from the seismic survey and using that fault size in such models can help to determine critically stressed areas around the fault. This study represents few of the geomechanical factors responsible that can affect the variation in production from a faulted reservoir. This practice may help fracking engineer to develop a good understanding of local stress disturbance in subsurface leading to a successful fracture treatment. It is obvious that fractures created due to fracture stimulation will intersect such faults in a reservoir.2 Recommendations This study of stress distribution leads to following recommendations • Actual subsurface rocks have varying elastic properties. Fractures resulting from hydraulic fracture treatment have height of several hundred feet. However. Fault size play an important role in determining change in magnitude. Critically stressed faults may aid to the production. Shale gas targets in Cooper basin are also subjected to similar problems like Barnett Shale and Haynsville Shale. • Layered models with different combination of lithologies need to be modelled. the magnitude of stress perturbation is function of lithology and orientation of maximum horizontal stress. Stress is concentrated in two opposite quadrants of the fault at the fault tips. . The simulation uses real stress magnitudes estimated by (Reynolds. Stress perturbation increases with increase in fault size.Similar trend was observed in all models. 2006).

Vol. Unlocking the shale mystery: How lateral measurements and well placement impact completions and resultant production. 1996 SEG Annual Meeting.. Meyer." South Australia. 98 9." Exploration Geophysics 29(4): 420-427. A. and E. A... S. 4: Cooper Basin. (1998). IV Production Operations Engineering. Cooke. D. Department of Primary Industries and Resources. relation to hydrocarbon trap styles. (1991). (1997). Hibburt. Gravestock. "The production rate variability problem with shale reservoirs: what we know and what we don't know. Malpani. Lorenz." Journal of Structural Geology 28(12): 2204-2221. Gudmundsson. (2010). UK. R." Preview 2011(155): 22-27. 3-D boundary element seismic modeling in complex geology. (2000). Stuart." AAPG Bulletin 81(4): 533-555. King. S. D. J. Teufel. "Active fault zones and groundwater flow. Hillis. C." Journal of Structural Geology 19(5): 703-718. J. (2010). "Fault-induced perturbed stress fields and associated tensile and compressive deformation at fault tips in the ice shell of Europa: implications for fault mechanics. "Elastic stress control on the pattern of tensile fracturing around a small fault network at Nash Point. Australia.. J. (1998). S. (1997). N. et al. "The Australian stress map. J." AAPG Bulletin 75(11): 1714-1737. "Regional Fractures I: A Mechanism for the Formation of Regional Fractures at Depth in Flat-Lying Reservoirs (1)..References Apak. Homberg." Journal of Structural Geology 23(11): 1753-1770.. Hu. Baihly." Petroleum Engineering Handbook Vol. (1996). and S." Geophysical Research Letters 27(18): 2993-2996. Willemse (2001). J. Bourne.-Y. L. J. C. W. L. Kattenhorn. J. S. Tight Gas Completions Conference. A. Report Book. W. "The petroleum geology of South Australia. Marshall (2006). Thirty years of gas shale fracturing: what have we learned? SPE Annual Technical Conference and Exhibition. et al. "Hydraulic Fracturing. et al. Fu. Holditch. T. et al. (2007). R. "Characterization of stress perturbations near major fault zones: insights from 2-D distinct-element numerical modelling and field studies (Jura mountains). . et al. "Structural evolution of the Permian-Triassic Cooper Basin. G. et al. J. (2011).

Stanford University. Reynolds. L. R. "Fault size distributions—are they really power-law?" Journal of Structural Geology 18(2): 191-197. Müller. Reynolds. W. Reservoir geomechanics. The in situ stress field of the Cooper Basin and its implications for hot dry rock geothermal energy development. Mildren. et al. ." The Leading Edge 24(12): 1276-1282. "Characterization and modelling of the regional in situ stress field of continental Australia/Scott D. South Australia). A. and J. The 23rd US Symposium on Rock Mechanics (USRMS).. S. J.. D.." Journal of petroleum science and engineering 17(3): 353-366. S. D. polygonal element." Tectonophysics 415(1): 123-140. (2005). M. Using boundary element modeling of fault slip to predict patterns of stress perturbation and related fractures in geothermal reservoirs and explore parameter uncertainty. Australia: implications for plate‐scale tectonics and local stress sources. and cavities in the Earth's crust. faults. S. D. Walsh. D. L." Geophysical Journal International 160(1): 332-344. A hybrid discrete element-boundary element method of stress analysis. and N. Zoback. (2006). "Understanding tectonic stress in the oil patch: The World Stress Map Project. "Constraining stress magnitudes using petroleum exploration data in the Cooper–Eromanga Basins. (2007). Pecher. and B." Reynolds. (2005). Australia. Reynolds. A. S. (1996). S. Cambridge University Press. Mildren. Poly3D: A three-dimensional. Tingay. D. Swyer.. Reynolds. S. Davatzes (2012).. M. D. C. D. B. D. Mildren. Brady (1982). displacement discontinuity boundary element computer program with applications to fractures. "Boundary element techniques in petroleum reservoir simulation. et al. Thirty-Seventh Workshop on Geothermal Reservoir Engineering. Stanislav (1997). PESA Eastern Australasian Basins Symposium II (2004: Adelaide.Lorig. Thomas. (2004). et al. (2001). et al. Nicol. M. et al. "Maximum horizontal stress orientations in the Cooper Basin. (1993). S. Proceedings.