-
8/10/2019 IPTC-14798-MS
1/14
IPTC 14798
Use of Reservoir Simulation to Help Gas Shale Reserves EstimationMichele Segatto, Ivan Colombo, eni E&P
Copyright 2011, International Petroleum Technology Conference
This paper was prepared for presentation at the International Petroleum Technology Conference held in Bangkok, Thailand, 79 February 2012.
This paper was selected for presentation by an IPTC Programme Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, aspresented, have not been reviewed by the International Petroleum Technology Conference and are subject to correction by the author(s). The material, as presented, does not necessarilyreflect any position of the International Petroleum Technology Conference, its officers, or members. Papers presented at IPTC are subject to publication review by Sponsor SocietyCommittees of IPTC. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the International Petroleum TechnologyConference is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuousacknowledgment of where and by whom the paper was presented. Write Librarian, IPTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax +1-972-952-9435
Abstract
Scope of this pape r it to supply a workflow for gas shale simulation: starting from selection of best porositymodel, through a definition of the mo st impacting phenomena hierarchy, it comes to a real application on gasshale in Barnett basin.
Gas shale exploitation consists mainly in massiv e drilling campaign: this strategy can be successfullyimplemented where the costs pe r well is l ow, but elsewhere an accurate investigation including numericalmodeling of gas shale reservoirs could be essential especially for defining optimal well spacing.
We test several implementations of multi-porosity models (dual porosity, multi porosity, dual permeability). Evenif all the mo dels can simulate properly gas shale production we use sin gle porosity-like that coupl es theadvantage of dual permeability and single porosity models.
Tests on synthetic cases have shown that gas production from shales is meanly affected by fracture spacing,proppant distribution inside induced and natural fractures, desorbed gas and in situstress change.
A real case of shale gas exploitation in Barnett shale is successfully matched by single porosity-like model, withSRV extension from microseismic mapping, water saturated a nd pressurized hydraulic fracture, adsorptionand change of in situstress. Match of the water flow-back is obtained with non-equilibrium initialization: settingfracture cells with higher pressure and water saturation than surrounding ones. The matched case is then usedas starting point for further analyses and sensitivities.
Final goal is to provide a workflow for frontier basin where few production data are available with respect towhat happens in USA and to calibrate decline curve for reserve estimation.
Introduction
Referring to gas shale means in 90% of cases talking about Barnett Basin (Texas, USA). They symb olize theideal play with huge number of wells already producing and where the market condition gives the opportunity to
-
8/10/2019 IPTC-14798-MS
2/14
2 IPTC 14798-PP
massively apply available technology and to test new development strategies. Anyway uncertainties still remains.Decline curves analysis are largely used to asse ss reserves and future d evelopment plans, but re servoirsimulation can give an added value adding more physics or controls compared to Decline Curves.
As known, 3D simulation of gas shales is not a simple task b ecause of co mplexity of storage and transportsystem, long water flow-back period, uncertain proppant distribution, gas desorption, etc..
Nevertheless in case of exploitation plans in frontier basins where capital investments and uncertainties are highand risky, the use of 3D reservoir simulator may help for a better tuning of the selected developments plan and fora better definition of associated risks and opportunities.
After a revie wing of the approaches already available in commercial simulations and a definition of keyparameters to be considered, an alternative workflow is proposed for a better understanding and simulation ofshale gas wells behavior, with two case histories.
Proposed workflow
System complexity, data uncertainties and general lack of information affect production performance estimationfor gas shale. History matching in general - and particularly for gas shale helps us on the understanding of the
physics inside the re servoir through the validation of real production data rather tha n being an exe rcise ofperfectly matching observed and simulated data.
The basic concept of the proposed workflow is that a se ries of parameters, validated by the HM or ri sked ifunknown, could be used as the basis to build a dataset to analyze and risk:
Drainage area / well spacing (Pad configuration),
Reference production profile and related uncertainties.
The workflow consists in two steps:
Step 1: Creation of a dataset of reference parameters through:
HM production of primary wells
Link microseismic and hydraulic fracture treatment
Highlighting risk factor
Step 2: Solution combination to evaluate and validate:
Well Spacing
Lateral length
Uncertainties on production to calibrate Decline curve parameters
The main point in Step 1 is represented by the link between microseismic events and properties of HF treatment
performed. By matching a series of fractures with respect to the ir microseismic response we can characteri zeeach fracture stage by the associated SRV (Stimulated Reservoir Volume) properties to be impleme nted insidesimulator (SRV and propped and unpropped fracture conductivity and spacing).
Step 2 key point is to use matched SRV properties to generate risked forecast to optimize play development alsovarying petrophysical properties. It is clear that a relation between SRV/HF and shale petrophysical propertiesaffects the SRV itself. We are not considering this dependency when applying the matched SRV as th ey are.
Anyway we may drive the previous realization, reducing or increasing SRV characterization, if we are expectingdifferent geomechanical properties.
-
8/10/2019 IPTC-14798-MS
3/14
IPTC 14798-PP 3
Key aspects
A list of concepts, phenomena that have to be accounted approaching shale gas simulation have been identifiedon authors experience and on critical review of published papers. Because of several research studies are stillongoing, especially on flow equations, at this tim e we are considering only what could be addressed bycommercial simulators.
Gas shales are source-rock, trap and re servoirs at the same time : hydroca rbon generated from organic carbondecomposition is not completely pushed out and is partially adsorbed on kerogen (20-70% of total volu me) andstored as free gas (30-80%) in small pores and natural - often healed - fra ctures. Moreover they ne ed to b ehydraulically fractured to prod uce at econ omical rate. This process often ge nerates complex fracture n etwork,named SRV, characterized by unpredictable fracture distribution, uneven proppant concentration and presence ofresidual water injected during HF job.
Gas shales can be considered as a multi-porosity system (Figure 1) characterized by micro- (
-
8/10/2019 IPTC-14798-MS
4/14
-
8/10/2019 IPTC-14798-MS
5/14
IPTC 14798-PP 5
MINC is com putationally more expensive than dual porosity model depending on number of ne sted sub-cells.Interaction among matrix cells belonging to different geometrical block is not permitted.
Dual porosity - dual permeability is an extension of dual porosity model: it allows flow between matrix cells (Figure5) through transmissibility factor. Dual permeability is computationally more expensive than dual porosity model.We noted that not all commercial simulators support adsorption feature in dual permeability model.
sigma
Transmissibility
Figure 5. Schematic representation of dual porosity dual permeability model.
All the above mentioned models have a strong dependency on sigma factor, a really sensitive parameter oftenused as match parameter, difficult to properly assess (especially in tiny cells) because not directly linked to any
physical phenomenon. It can be considered someway connected to network complexity.
The single porosity-like model bases on modified dual permeability model: inside SRV logarithmically refined cellsare forced toward a single porosity-like model by selective activation of matrix or fracture cells as shown in Figure6. Outside SRV it is a pure dual permeability model.
TRANSIMISSIBILITY
SIGMA
COARSE CELLS
Outside SRV
TARTAN GRID
Inside SRV
FRACTUREGRID
MATRIXGRID
TARTAN GRID
Inside SRV
Figure 6. Conceptual sketch of single porosity-like model.
The great advantage of single porosity-like model is the possibility to remove sigma factor dependency inside tinycells and to reach a more effective representation of flow behavior. The drawback is that not all simulators allowthe connection fracture-matrix between cells not belonging to same geometrical block.
In the present paper a single porosity-like model that couples the advantage of dual permeability (outside SRV)and single porosity models (in SRV) has been implemented in a commercial simulator and positively tested.
Dual permeability model is com monly used and adequately correct, whereas dual porosity or MINC mod els caneffectively be used only when matrix permeability is extremely low (< 10
5mD).
-
8/10/2019 IPTC-14798-MS
6/14
6 IPTC 14798-PP
Fracture complexity and proppant distribution
Gas shales are commonly brittle rocks because of large content of silica and cal cite minerals hence, whenhydraulically fractured, tend to g enerate complex fracture networks instead of single planar-like fractures asexpected in tight sand stone. Moreover HF job causes reactivation of cha otically distributed natural healedfractures. How proppant distributes into the fractures is a non-banal question to answer.
Several authors (see CIPOLLA, C. L ., 2009) suggest that proppant mainly distributes into dominant fractures(propped fracture) within a complex network of seco ndary fracture system with negligible or absent proppantconcentration (unpropped fracture). It is hel pful to notice that on on e hand complexity reduces proppanttransportation, but on the other hand the most complex network the most efficent is matrix-fracture contact and sothe greatest the production.
Fracture network complexity makes the fracture extension unpredictable, hence to evaluate the actual fracturegeometry it is important to have micros eismic or tiltmeter mapping or similar techniques. From a simulation pointof view complexity is represented through a fine gridding (proportional to fracture complexity) and small spacingamong fractures. The smallest the distance between two fractures, the most complex the fracture network.
Microseismic is a key point on definition of the fracture network highlighting both the shale volume interested bythe fracture and the degree of complexity that have been created.
Particular attention has to be pointed out on simulation of water flow-back because of the huge amount of waterinjected during hydraulic fracturing job that affects gas production for long time. Explicit modeling of HF job even iffeasible increases exponentially the runtime and in some cases is source of convergence issues. Anyway thesimulation of the reservoir just after the HF job, whe n fractures are plenty of water and pressurized, by means ofnon-equilibrium initialization, is hereby suggested.
In Figure 7we compare two solutions: the first accounts HF water (a) while the second does not (b).
(a) (b)
0.00
0.25
0.50
0.75
1.00
0.0 0.5 1.0 1.5 2.0
Dimensionlessrate
Time [years]
Gas Rate
Observed
Water free Model
0.00
0.25
0.50
0.75
1.00
0.0 0.5 1.0 1.5 2.0
Dimensionlessrate
Time [years]
Gas and Water Rate
Observed Gas
Observed Water
HF model GAS
HF model Water
Figure 7. History match with water flow back simulating model (a) and water free model (b).
Comparing the two models it appears that a worsening of reservoir and fracture properties is implicitly assumedwhen neglecting the flow back water.
On the other hand reserves of frontier basins, where just geological and petrophysical properties are available,will be overestimated if flow-back water is not accounted.
-
8/10/2019 IPTC-14798-MS
7/14
IPTC 14798-PP 7
0.00
0.25
0.50
0.75
1.00
0.0 0.5 1.0 1.5 2.0
Dimensionlessrate
Time [years]
Gas Rate
Observed
Water free Model
HF job model
Figure 8. Comparison between water flow back modeling (red) and water free modeling (green). Red dots represent observed data.
Main parameter unitsWatered
model
No watered
model
Matrix Permeability [mD] 1.0 E-5 1.0 E-6
CFunpropped fracture [mDft] 4.4 1.6
CFpropped fracture [mDft] 42 12
Table 1. Fracture conductivity and matrix permeability.
An accurate simulation should not neglect:
SRV extension validated by microseismic or tiltmeter mapping or similar techniques.
In situstress change due to gas production: a reduction of fluid pressure induces partial or complete
fracture closure. In situ stress change can affect up to 3 0% of global gas production. Obviously theeffect is much greater on unpropped or partially propped fractures. The optimum will be use compactiontable calibrated on core test.
Adsorption: A significant amount (2070%) of gas can be prese nt as adsorbed gas on keroge n,depending on TOC concentration. Adsorbed gas release is much more significant at lower pressure thatat higher pressure, hence its contribution becomes relevant at latter time.
Non-Darcy flow: In region at high velocity, such as high conductive main fractures, turbulent flow candevelop. Tartan grid within SRV allows a more accurate representation of non-Darcy flow down thefractures.
-
8/10/2019 IPTC-14798-MS
8/14
8 IPTC 14798-PP
Reference dataset for analysis
This workflow, validated on real case application in Barnett Shale, starts linking each fracture stage to its SRVmapped during MS a cquisition in te rm of volume a nd fracture properties, and comes to matched results byincluding water flow-back.
Figure 9. Microseismic mapping helps to build SRV extension.
The history case is a multi-fractured horizontal well characterized by four HF stages, simulated in this way:
Stage 1: not recorded assumed to be as homogenous fracture network without propped fractures;
Stage 2: two different main fractures with associated network of unpropped fractures;
Stage 3: a single big propped fracture with associated network of unpropped fractures;
Stage 4: a single propped fracture with associated network of unpropped fractures.
The fracture network spacing is defined and matched using a background tartan grid reproducing a spacing of200 ft (Ref. 27).
As shown in Figure 7.a we achieve a good history match. Th e HF properties and SRV matched is used toinvestigate spacing and lateral length.
In the next paragraphs there are two examples of synthetic cases that start from these matched SRV properties.
-
8/10/2019 IPTC-14798-MS
9/14
IPTC 14798-PP 9
Well spacing optimization
Once HM SRV has been validated via synth etic and real applications, the optimum well spacing through thecomparison of gas production from a variable number of horizontal multi-fractured wells within a square mile canbe evaluated
Because of unpredictability of SRV extension and difficulty to anticipate the p roppant distribution, three possiblescenarios were considered (Figure 10):
Conservative case: each HF stage p roduces the minimum SRV extension interpreted during MSmapping of real case (P90)
Realistic case: each HF stage produces a random SRV extension interpreted during MS mapping of realcase (P50) random choice from all the history matched SRV.
Optimistic case: each HF stage generates the maximum SRV extension interpreted during MS mappingof real case (P10)
Optimistic
Realistic
Conservative
900 ft 600 ft
900 ft
900 ft
600 ft
600 ft
Figure 10. Well spacing for the three scenarios. Permeability maps.
Moreover we suppose that some enhancement of f racture conductivity occurs when SRVs intersect but alsoover-stimulated zones are affected by presence of huge water amount (Figure 10).
Figure 11shows how well performance decreases when SRVs intersect causing wells competition and penalizingoverall production. Of course the number of wells is driven by overall project NPV, anyway this graph sho ws theexpected loss of performance you might have if you are over-stimulating or over-drilling your field.
A complete dataset to assess possible variation of production performance for a given spacing or to evaluate thebest development spacing with respect to HF treatment (and then to perform the economical evaluation) wasderived.
-
8/10/2019 IPTC-14798-MS
10/14
10 IPTC 14798-PP
0%
25%
50%
75%
100%
900 600 400 300
DimensionlessEURpe
rwell
Well distance [ft]
Optimistic
Realistic
Conservative
Figure 11. Percentage with respect to single well EUR for optimistic scenario vs. well spacing.
Well lateral length
Starting from SRV from a real case the well behavior with respect to the lateral length and number of frac stageswas considered, considering two possible scenarios:
Short drain: 4 frac stages;
Long drain: 10 frac stages.
SRV extension of each frac stage is supposed to be randomly distribute (calibrated on real case MS mapping)and the water amount injected is proportional to number of frac stages (Figure 12).
Long Drain
4 Frac
Stages
10 Frac
Stages
Short drain
(a) (b)
Figure 12. Horizontal multi-fractured wells with different fracture stage and lateral length.
-
8/10/2019 IPTC-14798-MS
11/14
IPTC 14798-PP 11
Figure 13shows how gas deliverability per unit of lateral length decreases with drain length, for instance: doublinglateral length results in an increase of just 60% of EUR at 20 year, that means a reduction of roughly 20% ofEUR/ft.
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
5 10 15 20
EURperunitoflaterallenght[Bcf/ft]
Time [years]
SHORT
LONG
Figure 13. EUR per unit of later length
It is because water plays a fundamental role on production: the longest the drain and most numerous HF stagesthe most water is injected and so the longest the clean-up, for this reason it is important and hereby suggested toimplement the VFP table in simulation to correctly account the effect of water on production.
This approach should be used to def ine the optimal la teral length by histo ry matching and testing differentconfigurations especially for longer and massively stimulated wells where the cleanup effectiveness seems toimpact final recovery.
Uncertainties on production to calibrate Decline curve parameters
The described workflow can be also applied to reduce uncertainty when using decline curve approach for reserveestimation especially during early production phase when only few data are available or to guess the production infrontier basins, starting from basic petrophysical data.
This methodology, calibrated on available production data allows to generate a synthetic production profile basedon consistent physic description that can account:
Well and HF characteristics,
Petrophysics and core data,
Effectiveness of clean-up phase, etc
To increase the reliability of this methodology a series of scenarios through a stochastic approach, by definingprobability density functio ns instead of deterministic values for unce rtain parameters can be gen erated. Inparticular the generation of a conservative, an optimistic and a realistic SRV estimation considering respectivelyminimum, maximum and mixed SRV extension for each HF stage from MS mapping (if available) or analog field ishereby recommended.
Once you get final spectrum of production profile you can characterize decline curves also in terms of statisticaldistribution of reference parameters (Qi/ft, b, first year decline,).
-
8/10/2019 IPTC-14798-MS
12/14
12 IPTC 14798-PP
Conclusions
Gas shale 3D simulation is a hard task becau se of intrinsic complexity of fracture geometry and presence of twointeracting media with substantially different properties: matrix, with high storage capacity and low permeabilityand fractures, with high permeability and low storage capacity.
Dual porosity, multi porosi ty or equivalently MINC and dual permeability mo dels are all suitable for fractu redreservoir representation but have a strong dependency on sigma factor, a sensitive parameter, difficult to assessespecially in tiny cells. A novel approach, named single porosity-like, seems to b e promising for ga s shalesimulation because it fre es from sigma dependency and allows a fairly well match of real data (from Barnettshale) well.
A simulation workflow based on single porosity model and validated by HM is here by proposed, that allows aconsistent representation of physics of gas shale such as:
Actual SRV extension evaluated from microseismic mapping or similar techniques or analog field;
Water injected during HF job for water flow back representation;
Impact of adsorption (on later production);
Impact of change of in situstress condition with pressure.
A significant aspect of this approach is the abilit y to risk the uncertai n parameters and gene rate multiplescenarios, modify and optimize the pad configuration and produce reference production profile even for newlydrilled wells and frontier fields.
Although the authors are conscious that a massive use of simulation for gas shale is nowadays not practicable,the potentiality and meth odological approaches to gas shale numerical simulation are continuously improving,allowing a better description and simulation of related phenomena and possibly the assessment of a powerful toolto reduce in future uncertainties and to optimize play developments and associated reserves.
Nomenclature
Bcf Billion cubic feet
Bscf Billion standard cubic feet
CBM Coal Bed Methane
CF Fracture conductivity
EUR Estimated Ultimate Recovery
GOIP Gas Originally In Place
HF Hydraulic Fracturing
HM History Match
LGR Local Grid Refinement
MINC Multiple INteracting Continua
MS Microseismic
SRV Stimulated Reservoir Volume
Tcf Trillion cubic feet
TOC Total Organic Carbon
VFP Vertical Flow Performance
WOIP Water Originally In Place
SI metric Conversion Factors
bbl x 1.589874 e-01 = m3
ft x 3,048 e-01 = mmile
2x 2.5899072 e+06 = m
2
psi x 6.897757 e+00 = kPa
-
8/10/2019 IPTC-14798-MS
13/14
IPTC 14798-PP 13
References
Ahmed, T., & McKinney, P. D. (2005).Advanced Reservoir Engineering.Burlington, MA USA: Elsevier.
Aziz, K., Durlofsky, L., & Tchelepi, H. A. (2005). Notes On Reservoir Simulation.Stanford University - Petroleum Engineering.
Cipolla, C. L. (200 9). Modeling Production And Evaluating Fracture Performance In Unconventional Gas Reservoir - SPE118536. JPT, p. 84 -90.
Cipolla, C. L., Lolon, E. P., & Rubin, B. (2009). Reservoir Modeling in Shale-Gas Reservoirs - SPE 125530. SPE ReservoirEvaluation & Engineering, p. 638-653.
Cipolla, C. L., Lolon, E. P., Erdle, J. C., & Tathed, V. (2009). Modeling Well Performance in Shale-Gas Reservoirs - 125532.SPE/EAGE Reservoir Characterization and Simulation Conference, 19-21 October 2009.Abu Dhabi, UAE.
Cipolla, C. L.; Lolo n, E. P.; D zubin, B. (2009). Evaluating Stimulation Effectiveness In Unconventional Gas Reservoir - SPE124843. SPE Annual Technical Conference and Exhibition, 4-7 October 2009. New Orleans, LA USA: 2009. Society ofPetroleum Engineers.
Du, C., Zhang, X., Melton, B., Fullilove, D., Suliman, B., Gowelly, S., et a l. (2009). A Workflow for Integrated Barnett ShaleGas Reservoir Modeling and Simulation - SPE 122934. Latin American and Caribbean Petroleum Engineering Conference,31 May-3 June 2009.Cartagena de Indias, Colombia.
Fan, Y., Durlofsky, L. D., & Tchelepi, H. A. (2010, June). Numerical Simulation of the In-Situ Upgrading of Oil Sha le - SPE118958. SPE Journal, p. 368-381.
Frantz, J. H., Williamson, J. R., Sawyer, W. K., Johnston, D., Waters, G., Moore, L. P., et al. (2005). Evaluating Barnett ShaleProduction Performance Using an Integrated Approach - SPE 96917. SPE Annual Technical Conference and Exhibition, 9-12 October 2005, (pp. 1-10). Dallas, TX USA.
Gilman, J. R., & Kazemi, H. (1983, August). Improvements in Simulation of Naturally Fractured Reservoirs - SPE 10511. SPEJournal, p. 695-707.
Guppy, K. H., Cinc o-Ley, H., Ramey, H. J ., & Samani ego, V. F. (1982, October). No n-Darcy Flow in Wells With Finite-Conductivity Vertical Fractures - SPE 8281. SPE Journal, p. 681-698.
Holditch, S. A., & Morse, R. A. (1976, October). The Effects of Non-Darcy Flow on the Behavior of Hydraulically Fractured GasWells - SPE 5586. JOURNAL OF PETROLEUM TECHNOLOGY, p. 1169-1179.
Kassis, S., & Sondergeld, C. H. (2010). Fracture Permeability of Gas Shale: Effects of Roughness, Fracture Offset, Proppant,and Effective Stress - SPE 131376. International Oil and Gas Conference and Exhibition in China, 8-10 June 2010, (pp. 1-17). Beijing, China.
Kuuskraa, V. A. (2009). Worldwide Gas Shales And Unconventional Gas: A Status Report. Arlington, VA USA: AdvancedResources International, Inc.
Kwon, O., Kronenberg, A. K., Gangi, A. F., & Johnson, B . (2001). Permeability Of Wilcos Shale And Its Effective P ressureLaw. Journal of Geophysical Research, 106(B9), p. 19,339-19,353.
Liu, X., Civan, F., & Evans, R. D. (19 95). Correlation of the Non-Darcy Flow Coefficient. Journal of Canadin PetroleoumTechnology, p. 50-54.
Mayerhofer, M. J., Lolon, E. P., W arpinski, N. R., Cipoll a, C. L., Walser, D., & Rightmi re, C. M. (2010, F ebruary). What IsStimulated Reservoir Volume? - SPE 119890. SPE Production & Operations, p. 89-98.
Mayerhofer, M. J., Lolo n, E. P., Youngblood, J. E., & H einze, J. R. (2006). Integration of Microseismic Fracture MappingResults With Numerical Fracture Network Production Modeling in the Barnett Shale - SPE 102103. SPE Annual TechnicalConference and Exhibition, 24-27 September 2006.San Antonio, TX USA.
Medeiros, F., Kurtoglu, B., Ozkan, E., & Kazemi, H. (2010 , June). Analysis of Production Data From Hydraulically FracturedHorizontal Wells in Shale Reservoirs - SPE 110848. SPE Reservoir Evaluation & Engineering, p. 559-568.
Palmer, I., Moschovidis, Z., & Cameron, J. (2007). Modeling Shear Failure and Stimulation of the Barnett Shale After HydraulicFracturing - S PE 206113. SPE Hydraulic Fracturing Technology Conference, 29-31 January 2007, (pp. 1-9). Co llegeStation, TX, U.S.A.
Rubin, B. (2007). Accurate Simulation of Non Darc y Flow in Stimulated Fractured Sh ale Reservoirs - SPE 132093. SPEWestern Regional Meeting, 27-29 May 2010.Anaheim, CA, USA.
Schepers, K. C., Gonzalez, R. J., Koperna , G. J., & Oudinot, A. Y. (20 09). Reservoir Modeling in Support of Sh ale GasExploration - SPE 123057. Latin American and Caribbean Petroleum Engineering Conference, 31 May-3 June 2009. Cartagena de Indias, Colombia.
-
8/10/2019 IPTC-14798-MS
14/14
14 IPTC 14798-PP
Settari, A., Sullivan, R. B., Rother, R. J., & Skinner, T. K. (2009). Comprehensive Coupled Modeling Analysis of Stimulationsand Post-Frac Productivity---Case Study of a Tight Gas Field in W yoming - SPE 11 9394. SPE Hydraulic FracturingTechnology Conference, 19-21 January 2009, (pp. 1-11). The Woodlands, TX, USA.
Soliman, M. Y., & Augustine , J. (2010). Fracturin g Design Aimed at Enhancing Fracture Com plexity - SPE 1 30043. SPEEUROPEC/EAGE Annual Conference and Exhibition, 14-17 June 2010, (pp. 1-20). Barcelona, Spain.
Sondergeld, C. H., Ne wsham, K. E., Co misky, J. T., Rice, M. C., & Rai, C. S. (2010). Petrophysical Considerations in
Evaluating and Producing S hale Gas Resources - SPE 131768. SPE Unconventional Gas Conference, 23-25 February2010, (pp. 1-34). Pittsburgh, PA, USA.
Verrastro, F., & Branc h, C. (2010). Developing America's Unconvetional Gas Resources Benefits and Challenges.Washington, DC USA: Center for Strategic & International Studies.
Wang, F. P., Reed, R., Jackson, K. G., & Jackson, J. A. (2009). Pore Net works and Fluid Flow in Gas Shales - SPE 124253.SPE Annual Technical Conference and Exhibition, 4-7 October 2009.New Orleans, La, USA.
Warpinski, N. R., Mayerhofer, M. J. , Vincent, M. C., Cipolla, C. L., & Lolon, E. P. (2008). Stimulating UnconventionalReservoirs: Maximizing Netw ork Growth While Optimizing Fracture Conductivity - S PE 114173. SPE UnconventionalReservoirs Conference, 10-12 February 2008.Keystone, CO, USA.
Xu, W., Thiercelin, M., Gang uly, U., Ond a, H., & Le Cal vez, J. (2010). Wiremesh: A Novel Sh ale Fracturing Simulator.International Oil and Gas Conference and Exhibition in China, 8-10 June 2010, (pp. 1-6). Beijing, China.
Zeng, F., & Zhao, G. (2 008). The Optimal H ydraulic Fracture Geometry Under Non-Darcy Flow Effects - SPE 1 14285.
CIPC/SPE Gas Technology Symposium 2008 Joint Conference, 16-19 June 2008, (pp. 1-18). Calgary, Alberta, Canada.
