Advanced Search

Indexed by SCI、CA、РЖ、PA、CSA、ZR、etc .

Volume 31 Issue 6
Dec.  2020
Turn off MathJax
Article Contents

Zhuoheng Chen, Chunqing Jiang. An Integrated Mass Balance Approach for Assessing Hydrocarbon Resources in a Liquid-Rich Shale Resource Play:An Example from Upper Devonian Duvernay Formation, Western Canada Sedimentary Basin. Journal of Earth Science, 2020, 31(6): 1259-1272. doi: 10.1007/s12583-020-1088-1
Citation: Zhuoheng Chen, Chunqing Jiang. An Integrated Mass Balance Approach for Assessing Hydrocarbon Resources in a Liquid-Rich Shale Resource Play:An Example from Upper Devonian Duvernay Formation, Western Canada Sedimentary Basin. Journal of Earth Science, 2020, 31(6): 1259-1272. doi: 10.1007/s12583-020-1088-1

An Integrated Mass Balance Approach for Assessing Hydrocarbon Resources in a Liquid-Rich Shale Resource Play:An Example from Upper Devonian Duvernay Formation, Western Canada Sedimentary Basin

doi: 10.1007/s12583-020-1088-1
More Information
  • Petroleum resource assessment using reservoir volumetric approach relies on porosity and oil/gas saturation characterization by laboratory tests. In liquid-rich resource plays, the pore fluids are subject to phase changes and mass loss when a drilled core is brought to the surface due to volume expansion and evaporation. Further, these two closely related volumetric parameters are usually estimated separately with gas saturation inferred by compositional complementary law, resulting in a distorted gas to oil ratio under the circumstances of liquid hydrocarbon loss from sample. When applied to liquid-rich shale resource play, this can lead to overall under-estimation of resource volume, distorted gas and oil ratio (GOR), and understated resource heterogeneity in the shale reservoir. This article proposes an integrated mass balance approach for resource calculation in liquid-rich shale plays. The proposed method integrates bulk rock geochemical data with production and reservoir parameters to overcome the problems associated with laboratory characterization of the volumetric parameters by restoring the gaseous and light hydrocarbon loss due to volume expansion and evaporation in the sample. The method is applied to a Duvernay production well (14-16-62-21W5) in the Western Canada Sedimentary Basin (WCSB) to demonstrate its use in resource evaluation for a liquid-rich play. The results show that (a) by considering the phase behavior of reservoir fluids, the proposed method can be used to infer the quantity of the lost gaseous and light hydrocarbons; (b) by taking into account the lost gaseous and light hydrocarbons, the method generates an unbiased and representative resource potential; and (c) using the corrected oil and gas mass for the analyzed samples, the method produces a GOR estimate close to compositional characteristics of the produced hydrocarbons from initial production in 14-16-62-21W5 well.
  • 加载中
  • AER, 2016. Duvernay Reserves and Resources Report, A Comprehensive Analysis of Alberta's Foremost Liquids-Rich Shale Resource. https://www.aer.ca/documents/reports/DuvernayReserves_2016.pdf
    Akkutlu, I. Y., Baek, S., Olorode, O. M., et al., 2017. Shale Resource Assessment in Presence of Nanopore Confinement. Unconventional Resources Technology Conference, SPE/AAPG/SEG Unconventional Resources Technology Conference, URTEC-2670808-MS, July 24-26, 2017, Austin, Texas, USA. https://doi.org/10.15530/urtec-2017-2670808
    Akkutlu, I. Y., Fathi, E., 2012. Multiscale Gas Transport in Shales with Local Kerogen Heterogeneities. SPE Journal, 17(4):1002-1011. https://doi.org/10.2118/146422-pa doi:  10.2118/146422-pa
    Baek, S., Akkutlu, I. Y., 2019. Produced-Fluid Composition Redistribution in Source Rocks for Hydrocarbon-in-Place and Thermodynamic Recovery Calculations. SPE Journal, 24(3):1395-1414. https://doi.org/10.2118/195578-pa doi:  10.2118/195578-pa
    Beaton, A. P., Pawlowicz, J. G., Anderson, S. D. A., et al., 2010. Rock Eval, Total Organic Carbon and Adsorption Isotherms of the Duvernay and Muskwa Formations in Alberta: Shale Gas Data Release. Energy Resources Conservation Board, Alberta Geological Survey, Open File Report 2010-04, Edmonton, AB, Canada. 39
    Bohacs, K. M., Passey, Q. R., Rudnicki, M., et al., 2013. The Spectrum of Fine-Grained Reservoirs from "Shale Gas" to "Shale Oil"/Tight Liquids: Essential Attributes, Key Controls, Practical Characterization. IPTC 2013: International Petroleum Technology Conference, March 26, 2013, IPTC-16676.1-16
    Chen, Z., Jiang, C., 2016. A Revised Method for Organic Porosity Estimation in Shale Reservoirs Using Rock-Eval Data:Example from Duvernay Formation in the Western Canada Sedimentary Basin. AAPG Bulletin, 100(3):405-422. https://doi.org/10.1306/08261514173 doi:  10.1306/08261514173
    Chen, Z., Lavoie, D., Malo, M., et al., 2017. A Dual-Porosity Model for Evaluating Petroleum Resource Potential in Unconventional Tight-Shale Plays with Application to Utica Shale, Quebec (Canada). Marine and Petroleum Geology, 80:333-348. https://doi.org/10.1016/j.marpetgeo.2016.12.011 doi:  10.1016/j.marpetgeo.2016.12.011
    Chen, Z., Li, M. W., Ma, X. X., et al., 2018. Generation Kinetics Based Method for Correcting Effects of Migrated Oil on Rock-Eval Data—An Example from the Eocene Qianjiang Formation, Jianghan Basin, China. International Journal of Coal Geology, 195:84-101. https://doi.org/10.1016/j.coal.2018.05.010 doi:  10.1016/j.coal.2018.05.010
    Chen, Z. H., Li, M. W., Jiang, C. Q., et al., 2019. Shale Oil Resource Potential and Mobility Assessment:A Case Study of Upper Devonian Duvernay Shale in the Western Canada Sedimentary Basin. Oil & Gas Geology, 40(6):459-468 (in Chinese with English Abstract) http://en.cnki.com.cn/Article_en/CJFDTotal-SYYT201903003.htm
    Creaney, S., Allan, J., Cole, K. S., et al., 1994. Petroleum Generation and Migration in the Western Canada Sedimentary Basin. In: Mossop, G. D., Shetsen, I., eds., Geological Atlas of the Western Canada Sedimentary Basin. Canadian Society of Petroleum Geologists and Alberta Research Council.[2019-6-24]. http://www.ags.gov.ab.ca/publications/wcsb_atlas/atlas.html
    Dong, T., Harris, N. B., McMillan, J. M., et al., 2019. A Model for Porosity Evolution in Shale Reservoirs:An Example from the Upper Devonian Duvernay Formation, Western Canada Sedimentary Basin. AAPG Bulletin, 103(5):1017-1044. https://doi.org/10.1306/10261817272 doi:  10.1306/10261817272
    Euzen, T., 2011. Shale Gas—An Overview. Technique Report. IFP Technologies (Canada) Inc., Calgary, Canada. https://doi.org/10.13140/RG.2.1.2236.6242
    Fowler, M. G., Stasiuk, L. D., Hearn, M., et al., 2001. Devonian Hydrocarbon Source Rocks and Their Derived Oils in the Western Canada Sedimentary Basin. Bulletin of Canadian Petroleum Geology, 49(1):117-148. https://doi.org/10.2113/49.1.117 doi:  10.2113/49.1.117
    GRI, 1996. GRI-95/0496: Development of Laboratory and Petrophysical Techniques for Evaluating Shale Reservoirs. Gas Research Institute, Chicago. 304
    Jarvie, D. M., 2012. Shale Resource Systems for Oil and Gas: Part 2—Shale-Oil Resource Systems. In: Breyer, J. A., ed., Shale Reservoirs—Giant Resources for the 21st Century. AAPG Memoir, 97: 89-119
    Jarvie, D. M., 2014. Components and Processes Affecting Producibility and Commerciality of Shale Resource Systems. Geologica Acta, 12(12):307-325. https://doi.org/10.1344/geologicaacta2014.12.4.3 doi:  10.1344/geologicaacta2014.12.4.3
    Jiang, C., Chen, Z., Mort, A., et al., 2016a. Hydrocarbon Evaporative Loss from Shale Core Samples as Revealed by Rock-Eval and Thermal Desorption-Gas Chromatography Analysis:Its Geochemical and Geological Implications. Marine and Petroleum Geology, 70:294-303. https://doi.org/10.1016/j.marpetgeo.2015.11.021 doi:  10.1016/j.marpetgeo.2015.11.021
    Jiang, C., Obermajer, M., Chen, Z., 2016b. Rock-Eval/TOC Analysis of Selected Core Samples of the Devonian Duvernay Formation from the Western Canada Sedimentary Basin, Alberta. Geological Survey of Canada, Open File 8155.532. https://doi.org/10.4095/299332
    King, R. R., 2015. PS Modified Method and Interpretation of Source Rock Pyrolysis for an Unconventional World. American Association of Petroleum Geologists Bulletin Search and Discovery Article #41704. http://www.searchanddiscovery.com/documents/2015/41704king/ndx_king.pdf
    Leythaeuser, D., Mackenzie, A, Scharfer, R., et al., 1984. A Novel Approach for Recognition and Quantification of Hydrocarbon Migration Effects in Shale-Sandstone Sequences. AAPG Bulletin, 68:196-219 http://www.researchgate.net/publication/241934040_A_novel_approach_for_recognition_and_quantification_of_hydrocarbon_migration_effects_in_shale-sandstone_sequences
    Li, M. W., Chen, Z. H., Ma, X. X., et al., 2019. Shale Oil Resource Potential and Oil Mobility Characteristics of the Eocene-Oligocene Shahejie Formation, Jiyang Super-Depression, Bohai Bay Basin of China. International Journal of Coal Geology, 204:130-143. https://doi.org/10.1016/j.coal.2019.01.013 doi:  10.1016/j.coal.2019.01.013
    Li, M. W., Chen, Z. H., Qian, M. H., et al., 2020. What are in Pyrolysis S1 Peak and what are Missed? Petroleum Compositional Characteristics Revealed from Programed Pyrolysis and Implications for Shale Oil Mobility and Resource Potential. International Journal of Coal Geology, 217:103321. https://doi.org/10.1016/j.coal.2019.103321 doi:  10.1016/j.coal.2019.103321
    Lyster, S., Corlett, H. J., Berhane, H., 2017. Hydrocarbon Resource Potential of the Duvernay Formation in Alberta. Alberta Energy Regulator, AER/AGS Open File Report 2017-02.44
    Macedo, R., 2013. Duvernay Well Encouraging for Encana. Daily Oil Bulletin, (2013-4-24)[2016-1-18]. www.dailyoilbulletin.com/headlines/2013-04-24/#axzz3xdfddAum
    Michael, G. E., Packwood, J., Holba, A., 2013. Determination of in-situ Hydrocarbon Volumes in Liquid Rich Shale Plays. In: Unconventional Resources Technology Conference, August, 2013, Denver, Colorado, USA. www.searchanddiscovery.com/pdfz/documents/2014/80365michael/ndx_michael.pdf.html
    Modica, C. J., Lapierre, S. G., 2012. Estimation of Kerogen Porosity in Source Rocks as a Function of Thermal Transformation:Example from the Mowry Shale in the Powder River Basin of Wyoming. AAPG Bulletin, 96(1):87-108. https://doi.org/10.1306/04111110201 doi:  10.1306/04111110201
    Passey, Q. R., Bohacs, K. M., Esch, W. L., et al., 2010. From Oil-Prone Source Rock to Gas-Producing Shale Reservoir—Geologic and Petrophysical Characterization of Unconventional Shale-Gas Reservoirs. International Oil and Gas Conference and Exhibition, June 8-10, 2010, Beijing. SPE-131350_MS
    Rezaveisi, M., Javadpour, F., Sepehrnoori, K., 2014. Modeling Chromatographic Separation of Produced Gas in Shale Wells. International Journal of Coal Geology, 121:110-122. https://doi.org/10.1016/j.coal.2013.11.005 doi:  10.1016/j.coal.2013.11.005
    Torsæter, O., Abtahi, M., 2000. Experimental Reservoir Engineering Laboratory Work Book. Department of Petroleum Engineering and Applied Geophysics, Norwegian University of Science and Technology, Trondheim
    Wang, P. W., Chen, Z. H., Jin, Z. J., et al., 2018. Shale Oil and Gas Resources in Organic Pores of the Devonian Duvernay Shale, Western Canada Sedimentary Basin Based on Petroleum System Modeling. Journal of Natural Gas Science and Engineering, 50:33-42. https://doi.org/10.1016/j.jngse.2017.10.027 doi:  10.1016/j.jngse.2017.10.027
    Whitson, C. H., Sunjerga, S., 2012. PVT in Liquid-Rich Shale Reservoirs. SPE Annual Technical Conference and Exhibition, October 8-10, 2012, San Antonio, Texas, USA. SPE-155499-MS
    Yu, W., Sepehrnoori, K., Patzek, T. W., 2016. Modeling Gas Adsorption in Marcellus Shale with Langmuir and BET Isotherms. SPE Journal, 21(2):589-600. https://doi.org/10.2118/170801-pa doi:  10.2118/170801-pa
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(10)  / Tables(4)

Article Metrics

Article views(16) PDF downloads(4) Cited by()

Related
Proportional views

An Integrated Mass Balance Approach for Assessing Hydrocarbon Resources in a Liquid-Rich Shale Resource Play:An Example from Upper Devonian Duvernay Formation, Western Canada Sedimentary Basin

doi: 10.1007/s12583-020-1088-1

Abstract: Petroleum resource assessment using reservoir volumetric approach relies on porosity and oil/gas saturation characterization by laboratory tests. In liquid-rich resource plays, the pore fluids are subject to phase changes and mass loss when a drilled core is brought to the surface due to volume expansion and evaporation. Further, these two closely related volumetric parameters are usually estimated separately with gas saturation inferred by compositional complementary law, resulting in a distorted gas to oil ratio under the circumstances of liquid hydrocarbon loss from sample. When applied to liquid-rich shale resource play, this can lead to overall under-estimation of resource volume, distorted gas and oil ratio (GOR), and understated resource heterogeneity in the shale reservoir. This article proposes an integrated mass balance approach for resource calculation in liquid-rich shale plays. The proposed method integrates bulk rock geochemical data with production and reservoir parameters to overcome the problems associated with laboratory characterization of the volumetric parameters by restoring the gaseous and light hydrocarbon loss due to volume expansion and evaporation in the sample. The method is applied to a Duvernay production well (14-16-62-21W5) in the Western Canada Sedimentary Basin (WCSB) to demonstrate its use in resource evaluation for a liquid-rich play. The results show that (a) by considering the phase behavior of reservoir fluids, the proposed method can be used to infer the quantity of the lost gaseous and light hydrocarbons; (b) by taking into account the lost gaseous and light hydrocarbons, the method generates an unbiased and representative resource potential; and (c) using the corrected oil and gas mass for the analyzed samples, the method produces a GOR estimate close to compositional characteristics of the produced hydrocarbons from initial production in 14-16-62-21W5 well.

Zhuoheng Chen, Chunqing Jiang. An Integrated Mass Balance Approach for Assessing Hydrocarbon Resources in a Liquid-Rich Shale Resource Play:An Example from Upper Devonian Duvernay Formation, Western Canada Sedimentary Basin. Journal of Earth Science, 2020, 31(6): 1259-1272. doi: 10.1007/s12583-020-1088-1
Citation: Zhuoheng Chen, Chunqing Jiang. An Integrated Mass Balance Approach for Assessing Hydrocarbon Resources in a Liquid-Rich Shale Resource Play:An Example from Upper Devonian Duvernay Formation, Western Canada Sedimentary Basin. Journal of Earth Science, 2020, 31(6): 1259-1272. doi: 10.1007/s12583-020-1088-1
  • For a given hydrocarbon mass in reservoir pore space, the volume of oil and gas depends on their physical conditions, and may vary significantly when those conditions change. For example, light oil could appear as a single liquid phase in reservoir condition, but may occur as separate gaseous and liquid phases under surface condition.

    Petroleum resource assessment using reservoir volumetric approach relies on reservoir porosity and reservoir fluid saturations that are commonly obtained from laboratory tests on core samples. Laboratory saturation calculation relies on less compressible liquid residuals (oil and water) to infer gas saturation. However, pore fluids in samples are subject to phase changes and evaporative loss of gaseous and light hydrocarbons when a drilled core is brought to the surface as a result of volume expansion and pressure drop. As such, if petroleum fluid saturations from laboratory are directly applied to a liquid rich shale resource play, it may underestimate the resource potential as a large amount of evaporative petroleum fluids has already been lost before laboratory analysis, and overestimates gas to oil ratio as a result of complementary fluids calculation. Gaseous and light hydrocarbon loss during coring, sample storage and handling is a well-known problem. Efforts have been made to estimate the quantity of the loss, and the composition of loss. For example, Chen et al. (2018) proposed methods to estimate hydrocarbon loss during storage. Chen et al. (2019) used phase behavior to estimate the gaseous loss of hydrocarbon in an effort to restore the original mass of hydrocarbon in reservoir condition. While Michael et al. (2013) reported light hydrocarbon loss up to C14, Jiang et al. (2016a) observed the loss up to C12.

    In principle, laboratory procedure of hydrocarbon saturation test is similar to source rock analysis in a geochemical laboratory to determine the quantity and types of hydrocarbon fluids in the rock sample. Laboratory methods require either high temperature to evaporate, or use solvent to extract the petroleum fluids in the samples. High temperature involved in the lab tests could release the structural water between clay layers, thus leading to an overestimate of water saturation, and adding uncertainties to the calculated saturations of the pore fluids.

    Laboratory porosity characterization is method and laboratory dependent as reported before (e.g., Bohacs et al., 2013; Passey et al., 2010), and could be problematic for shale resource evaluation where hydrocarbon storage is in a continuum of multiscale pore structure with size down to nanoscale (Baek and Akkutlu, 2019) that consists of pores from different origins and with different physical/chemical properties (e.g., Akkutlu et al., 2017; Chen et al., 2017; Chen and Jiang, 2016; Bohacs et al., 2013; Akkutlu and Fathi, 2012; Passey et al., 2010).

    While oil and gas saturations and reservoir porosity are closely related petrophysical variables, laboratory tests for the two are independent. Therefore, lack of coherent treatments between the porosity and saturation calculation is another issue in resource estimation. For example, the effective porosity is calculated by measuring the connected pore spaces using specific gas or liquid under pressure (Torsæter and Abtahi, 2000). In case of the presence of oil and bitumen in pore space, the test fluid may not be able to displace the heavy oil and bitumen of large sizes and complicated molecular structures, and therefore may not provide the exact volume for porosity calculation. While in saturation measurement, solvent extraction is a common method for determining the volume of residual oil and bitumen, including nano-pore confined hydrocarbon fluids, to be used for determining the space occupied by petroleum fluid in saturation calculation. As a result, the two are partially disconnected, leading to uncertainties in the parameter estimation.

    In this study, we propose an integrated approach by combining bulk rock geochemical data with reservoir and production information to estimate the oil and gas resources in liquid- rich shale reservoirs. The proposed method incorporates information in mass from geochemistry data and phase behavior from reservoir fluids, as such the quantity and composition of the missed hydrocarbons can be restored, and used in resource calculation. This paper uses the Duvernay liquid rich shale resource play as an example to demonstrate the application of the revised methods.

  • The liquid rich resource play of the Upper Devonian Duvernay Formation (Fig. 1) in the Western Canada Sedimentary Basin (WCSB) is used to demonstrate the application of the proposed method and to compare the results with those derived from traditional reservoir volumetric approach. The Duvernay shale is a known source rock for the Devonian conventional petroleum system of WCSB in south-central Alberta (Creaney et al., 1994), and a proven liquid-rich shale resource play (AER, 2016; Macedo, 2013). The Duvernay Basin is divided into two sub-basins, the West Shale Basin and the East Shale Basin separated by the Rimbey-Meadowbrook Reef Tread (Fowler et al., 2001). The Duvernay shale is more siliceous in the West Shale Basin, while more calcareous in the East Shale Basin. Chen and Jiang (2016) reported that TOC ranges from < 1 wt.% to 14.8 wt.% across the basin with an average TOC of 4.59 wt.% in their dataset. The gross thickness of the Duvernay Formation varies from less than 10 m to greater than 100 m with an average about 50 m. This formation is usually divided into three units (e.g., Dong et al., 2019). However, in the area revealed by 14-16-62-21W5 well, Duvernay Formation is 42 m with two contrasting segments from petrophysical log responses and geochemical signatures (Fig. 2).

    Figure 1.  Map showing the study area and established resource play sweet-spots of the Upper Devonian Duvernay resource play in WCSB. Figure revised from Lyster et al. (2017).

    Figure 2.  Well log responses and TOC variations with depth in the Duvernay Formation from 14-16-62-21W5 well, WCSB.

    Current production wells established three sweet-spots in the Duvernay resource play, two in the West Basin and one in the East Basin as shown in Fig. 1. AER (2016) reported 76 TCF (trillion of cubic feet) of recoverable natural gas and 3.41 billion barrels of recoverable oil and 6.2 billion barrels of recoverable condensates. Based on petroleum system analysis, Wang et al. (2018) estimated shale oil in-place of 81.4 billion barrels and natural gas in-place of 408 TCF. If a recovery factor of 5% for oil and 20% for natural gas, the recoverable oil and gas could be 4.7 billion barrels of oil and 81.6 TCF for gas respectively, which are comparable to oil and gas resource estimates (AER, 2016).

  • A full data suite was collected from shale production pilot wells in Kaybob area in the West Shale Basin of Duvernay resource play in WCSB. The data include oil and gas fluid test results, source rock screening Rock-Eval analysis, routine porosity and permeability and hydrocarbon saturation data, mineral composition, production test results and well logs.

    Sixteen core samples were taken in production target interval from 3 255 to 3 290.1 m from Duverany Formation in 14-16-62-21W5 well (see location in Fig. 1). Routine source rock screening programmed pyrolysis analysis was performed at a commercial laboratory (GeoMark Research Ltd.). The set of samples was analyzed for porosity, permeability, saturations and mineral composition at Core Lab Petroleum Services. The matrix permeability is the effective gas permeability determined from pressure decay measurements on 20–35 mesh size samples prepared from fresh cores by crushing. Saturation is determined using the Dean Stark extraction method with sample powder of 20–35 mesh and dried at 110 ºC. Porosity and saturations are relative to total interconnected pore space. Oil volume was computed assuming an oil density of 800 kg/m3, and water volume corrected assuming a brine concentration of 30 000 ppm NaCl with an ambient density of 1 018 kg/m3. Laboratory standard and procedure are referred to GRI-95/0496 (GRI, 1996). The test results are provided in Tables 1 and 2. Three fluid samples were analyzed at the Core Lab and the measured fluid properties are presented in Table 1. The sample set was also scanned for mineral composition by XRD at a commercial laboratory and the major mineral composition groups are listed in Fig. 2.

    As received Dry & Dean Stark extracted conditions Rock-Eval analysis
    Sample Depth (m) Bulk density (103 kg/m3) Matrix permeability (mD) Gas-filled porosity (%) Gas saturation (%) Grain density (103 kg/m3) Porosity (%) Oil saturation (%) Water Saturation (%) Leco TOC (wt.%) S1 (mg HC/g rock) S2 (mg HC/g rock) S3 (mg co2/g rock) Tmax (℃) HI (mg HC/g TOC)
    1 3 255.0 2.494 1.89E-06 3.82 57.3 2.642 6.66 5.5 37.2 2.61 2.41 1.96 0.27 464 75
    2 3 257.3 2.527 3.79E-06 3.55 54.0 2.674 6.58 14.0 31.9 3.56 9.19 3.19 0.38 465 90
    3 3 260.0 2.515 2.59E-06 2.89 50.9 2.639 5.68 22.3 26.8 2.96 2.87 1.99 0.33 467 67
    4 3 261.7 2.540 3.02E-07 2.54 42.8 2.665 5.94 11.8 45.4 2.56 8.26 2.17 0.44 460 85
    5 3 265.3 2.543 1.46E-07 2.46 42.0 2.668 5.85 21.8 36.2 2.57 3.03 1.93 0.34 465 75
    6 3 266.3 2.565 5.11E-07 3.31 49.3 2.716 6.71 21.1 29.6 2.32 8.21 2.29 0.37 463 99
    7 3 269.2 2.472 1.96E-06 2.76 47.4 2.596 5.83 32.5 20.1 4.67 14.63 4.73 0.45 460 101
    8 3 271.5 2.488 6.59E-07 2.28 41.0 2.603 5.57 31.9 27.1 4.79 12.27 4.93 0.40 462 103
    9 3 273.7 2.510 3.26E-06 2.95 58.2 2.623 5.07 15.6 26.2 4.37 4.76 3.69 0.30 469 84
    10 3 276.2 2.488 5.05E-06 3.56 58.0 2.625 6.14 18.4 23.6 3.86 6.21 3.14 0.36 468 81
    11 3 279.8 2.485 1.22E-06 3.19 54.1 2.615 5.89 20.9 25.0 3.35 2.55 2.37 0.25 471 71
    12 3 281.2 2.504 3.87E-07 3.02 50.1 2.635 6.02 25.3 24.6 4.19 11.42 3.79 0.39 466 90
    13 3 283.1 2.471 1.99E-06 3.58 52.5 2.619 6.81 17.4 30.1 4.39 9.99 4.18 0.32 466 95
    14 3 286.5 2.456 8.34E-07 3.00 57.7 2.569 5.20 17.9 24.3 5.34 7.63 4.65 0.33 470 87
    15 3 288.5 2.604 3.34E-07 2.46 54.6 2.707 4.51 21.8 23.6 1.88 1.51 0.98 0.25 460 52
    16 3 290.1 2.481 1.52E-06 3.69 60.2 2.620 6.14 20.5 19.3 3.66 7.36 2.89 0.35 469 79

    Table 1.  Reservoir parameters and the Rock-Eval analysis results showing general characteristics of the petroleum fluids and bulk geochemical features of the 14-16-62-21W5 well

    A large collection of Rock-Eval dataset from 12 recent production pilot wells across different hydrocarbon fluid zones in the Duvernay liquid-rich resource play basin was used to reveal the general characteristics of the Duvernay source rocks and to determine the maturity level of the study area in a regional context. The Rock-Eval analytical procedure and results for the 12 wells of Duvernay shale samples were reported in a Geological Survey of Canada Open File by Jiang et al. (2016b).

  • The volumetric approach for liquid rich hydrocarbons in shale resource play follows the standard reservoir volumetric approach using hydrocarbon pore volume as suggested by Passey et al. (2010), in which all reservoir volumetric parameters, such as porosities and saturations for each sample, are determined by commercial laboratories. The hydrocarbon saturated volume is then converted to mass of petroleum fluid in the unit of mg HC/g rock for comparison to the estimates from geochemical data. As the data well is located in the lower margin of a condensate zone (Fig. 1), condensate and gas are likely dissolved in oil in reservoir condition. In this case, the mass (M) of petroleum fluids in the rock volume V can be estimated by the following volumetric approach

    where Φ (fraction) is porosity, Sw (fraction) is water saturation; ρoilR (kg/m3) is density equivalent of petroleum fluid in reservoir, and V (m3) is volume of rock. In this study, we use V=1 m3 for each sample to simplify the calculation.

    If oil and gas are in separate phases, the oil and gas resource can be estimated separately using available volumetric parameters.

    where ρgasS and ρoilS are densities of natural gas and oil at standard surface condition.

    When hydrocarbon is brought to surface, gas volume expands while oil volume shrinks because gas comes out from oil as a separate phase in response to changes in temperature and pressure conditions. For marine source rocks in oil and condensate windows, release of natural gas from oil solution when a core sample is brought to surface is considered the primary mechanism for evaporative loss of light petroleum fluid during coring (Chen et al., 2019).

    Bulk geochemical data from programmed pyrolysis contain information on the composition and quantity of residual oil and gas in samples, useful for estimating the resource potential and evaluating composition characteristics (Chen et al., 2019; Li et al., 2019; Jarvie, 2014, 2012; Modica and Lapierre, 2012). However, traditional bulk geochemistry approach from programmed pyrolysis does not consider the petroleum fluid loss in the sample due to phase change. To overcome this issue, Chen et al. (2019) proposed a method for estimating gaseous component in shale oil reservoir based on the consideration of phase behavior of petroleum fluids. Figure 3 is a diagram showing the principles of volumetric changes due to phase changes under different physical conditions. In source rock reservoir of oil and wet gas windows, oil and natural gas co-exist in a single phase, likely gas being dissolved in oil (left of Fig. 3). When rock is taken to surface as a core, both temperature and pressure drop, resulting in a separation of gas from liquid phase. In their method, Rock-Eval pyrolysis data and reservoir parameters were combined to infer the volume and mass of petroleum fluids based on phase behavior under different conditions.

    Figure 3.  (a) Carton showing the principles of volumetric changes under different physical conditions (reservoir and surface) in oil and gas reservoirs. (b) The gas formation volume factor equations (derived from real gas equation) demonstrating gas volume expansion when gas is brought from reservoir to surface. PS and PR are surface and reservoir pressures; TS and TR are surface and reservoir temperatures (ºC), and z is the real gas correction factor.

    Let S1 denote the residual oil in sample, and S1LP the evaporative loss during sample storage and handling prior to analysis, FVF the oil formation volume factor (defined as the volume of petroleum fluid in a single oil phase in reservoir to the oil volume in surface condition). The mass of petroleum fluids, moilR (g oil/kg rock), at reservoir condition can be calculated from the following equation (Chen et al., 2019)

    The mass of gaseous and light petroleum composition group, mgasL (mg HC/g rock), lost during coring, sample storage and handling prior to analysis in the sample can be estimated by the following equation

    where moilR is the oil mass in reservoir condition containing dissolved natural gas; and ρoilR and ρoilS oil densities in reservoir and surface condition, respectively. The derivation of Eq. (3) is provided in Appendix A.

    Constrained by FVF, the equivalent oil density by combining oil and gas together in reservoir can be estimated by the following relationship

    The details of the derivation of the Eq. (4) are provided in Appendix B.

    In Chen et al. (2019), they did not consider the potential impact of gas sorption on kerogen surface and in liquid. We argue that there could be certain amount of gas sorpted to the rock and in petroleum liquid, which may not be released immediately from core samples. As such, we introduce a term to correct gas adsorption (Gads, mg HC/g rock) in estimating the evaporative hydrocarbon loss in S1 from programmed pyrolysis, which is a function of total organic carbon (TOC) content in the sample with the following form (Chen et al., 2016)

    For the Duvernay shale in this study, we use α =0.198 and β =0.098 in Eq. (5), which are derived by fitting a linear model using the gas adsorption data reported by Beaton et al. (2010).

    The revised evaporative loss during coring has the following form

    It is noteworthy that, as shown by a number of studies, gas adsorption capacity is rock type specific, which means that each shale may have its specific relationship with TOC (e.g., Wang et al., 2018; Chen and Jiang, 2016; Yu et al., 2016; Euzen, 2011). The total mass of natural gas (mgasS) in surface condition is the sum of the surface loss and adsorbed gases (mg HC/g rock) at surface condition.

    The total mass (M, in g) of oil and gas in unit rock volume can be estimated by the density (kg/m3) and mass (moilR, g oil/kg rock) of a single phase oil in reservoir with the following expression

  • The 14-16-62-21W5 well penetrated about 40 m of Duvernay Formation (Fig. 2), and the petrophysical responses and rock properties divide the formation into the upper and lower sections. The upper one shows higher bulk rock density and lower gamma ray, and slightly lower TOC and higher clay contents than the lower section. Bulk geochemical data of the basin from source rock Rock-Eval analysis show a wide range of variations in characteristics (Fig. 4), reflecting large variations in thermal maturity as well as depositional facies. The plot of Tmax-HI suggests a clear kerogen thermal degradation trajectory with samples from immature to dry gas window (Fig. 4a), indicating end of oil window maturity for the Duvernay shale at the location of 14-16-62-21W5 well. The cross-plot of Tmax-S1 shows an increasing trend followed by a decreasing trend of oil content with increasing maturity, indicating a peak oil generation stage of the studied samples (Fig. 4b). It is consistent with the Tmax-OSI (OSI=S1/TOC×100) plot, showing liquid-rich hydrocarbon zone (Fig. 4c). Deviation of the generation capacity defined by the TOC-S1 trend (Fig. 4d) may suggest possible internal oil migration towards to more permeable zones within the shale formation (Chen et al., 2018). The data points above the dashed line are likely oils concentrated in more permeable intervals within Duvernay Formation.

    Figure 4.  Cross-plots of various bulk geochemical properties of the Duverany shale samples. Tmax-HI (a); Tmax-S1 (b); Tmax-OSI (c); and TOC-S1 (d). The data points above the dashed line may have additional oil moved in (d).

    The pilot production test data from 14-16-62-21W5 shows an average flow rates of 61 200 m3/d of gas and 80.5 m3/d of condensate and oil. Laboratory hydrocarbon liquid analysis reported 30% of oil (C7+), 21% condensate (C3–C6) and 49% C1–C2. The true reservoir gas to oil ratio (GOR) must be lower than 49% because the initial GOR is only 430 m3/m3 shown in Fig. 5 and as suggested by recent studies (such as, Baek and Akkutlu, 2019; King, 2015; Whitson and Sunjerga, 2012). The laboratory measured oil density of C7+ is 773.1 kg/m3 and calculated oil density of the recombined well stream liquids is 435.8 kg/m3, which is consistent with composition analysis of produced hydrocarbon fluids from North American shale reservoirs (Rezaveisi et al., 2014). Rezaveisi et al. (2014) referred this phenomena as chromatographic separation in shale.

    Figure 5.  GOR estimates with time from first month production test of the 14-16-62-21W5 well indicating initial GOR of about 430 m3/m3 are much lower than overall 31 day average of 643.8 m3/m3.

    Laboratory analysis data collected from established oil and condensate fields in the Duvernay resource play were analyzed and plotted to establish empirical relationships among the reservoir parameters, such as oil formation volume factor (FVF) and gas to oil ratio (GOR), oil density in surface condition and thermal maturity (Tmax). Figure 6 displays cross-plots of reservoir parameters, petroleum fluid properties and bulk geochemical data, showing general characteristics of the Duvernay shale resource play in peak oil and condensate windows. From Fig. 6a, an initial GOR of 430 m3/m3 indicates an oil formation volume factor of 2.5. The laboratory measured oil density (C7+) of 773 kg/m3 (Fig. 6b) and Tmax between 460 and 470 ºC from geochemical data (Table 1, Figs. 6c and 6d) are all consistent, suggesting an early condensate window.

    Figure 6.  Cross-plots showing various petroleum fluid properties and reservoir parameters, as well as the relationship between GOR and thermal maturity in Duvernay Formation, WCSB.

    The geochemical data and reservoir parameters display interesting patterns. Reservoir volumetric and bulk geochemical parameter pairs of the samples from 14-16-62-21W5 well are plotted in Fig. 7, demonstrating the general trends of the relationships. Bulk density with TOC shows two linear trends, coincident with vertical lithological changes in the upper and lower intervals of the formation revealed by petrophysical log responses (Fig. 2b). Geochemical heterogeneity is also apparently indicated by TOC values, mostly < 3.5% in the upper section and > 3.5% in the lower interval (Fig. 2c). The upper part also shows slightly higher contents of clay minerals than the lower interval (Fig. 2d). Laboratory measured porosity ranges roughly from 5% to 7%, and shows no obvious trend with TOC (Fig. 7b), but a weak positive correlation with S1 (Fig. 7c). Hydrocarbon saturation does not follow any apparent trend with S1 in general, but a sub-group shows a weak positive correlation with S1 (Fig. 7d). The studied samples fall into two clusters on the cross-plot of TOC vs. hydrocarbon saturation (Fig. 8), where low TOC samples show a negative correlation with the saturation and a positive correlation holds for the higher TOC samples.

    Figure 7.  Correlation between geochemical and reservoir parameters of the studied samples. TOC vs. bulk density (a), TOC vs. reservoir porosity (b), S1 vs. porosity (c), and S1 vs. hydrocarbon saturation (d).

    Figure 8.  Cross-plot of TOC and hydrocarbon saturation showing different trends of hydrocarbon saturations for organic-rich and less-organic layers.

    The bulk geochemical data along with reservoir engineering parameters were used to estimate surface equivalent properties of petroleum fluids at reservoir conditions, and in turn to calculate the oil and gas resources (Table 2). The total gas is the sum of estimated gas loss from S1 and estimated gas in sorption, while oil is measured free hydrocarbons in S1. The estimated gas to oil ratio in mass varies from 37% to 70% with an average of 46%.

    TOC & minerais Average Upper Lower Relative clay data Average Upper Lower
    TOC (wt.%) 3.57 2.852 3.89 Illite/smectite 45.10 40.3 47.28
    Quartz 49.41 51.2 48.59 Illite & mica 42.86 46.52 41.19
    K-feldspar 6.60 4.32 7.64 Chlorite 12.06 13.18 11.55
    Plagioclase 1.23 1.7 1.01
    Calcite 11.57 10.84 11.90
    Dolomite & Fe-dolomite 4.86 2.96 5.73
    Pyrite 5.60 6.64 5.13
    Total clay 20.73 22.36 19.98 Mineralogical composition without TOC (wt.%)

    Table 2.  The average with maximum (upper) and minimum (lower) values of mineral composition data of 16 core samples from 14-16-62-21W5 well showing the general picture and variation of the mineral composition characteristics conducted at the Core Lab Petroleum Services

    The average initial GOR of 430 m3/m3 in volume from production data (Fig. 5) is equivalent to a mass ratio of 40.1% (e.g., 430×0.72/773=40.1%), which is consistent with the mass ratio of 43.5% calculated from recombined production fluids in laboratory and close to the average of the above estimates (46%) from the proposed method (Table 2). The estimated total hydrocarbons shows a large range of variations from < 5 kg/m3 to slight over 50 kg/m3 in samples, with an average of 23.86 kg/m3 of rock (Table 2).

  • Reservoir petrophysical parameters of the samples in conjunction with production fluid analytical results were used to estimate the oil and gas resource in the 14-16-62-21W5 well (Eqs. 1A and 1B) (Table 3). The total hydrocarbon resource from combined fluids by Eq. (1A) gives an average of 18.91 kg hydrocarbons/m3 rock, and 16.4 kg hydrocarbons/m3 rock by the Eq. (1B). The two estimates are lower than the estimate from the proposed method of an average 23.86 kg/m3 rock, which is expected because the laboratory measured hydrocarbon saturation is from the residuals without considering evaporative loss of light hydrocarbons. Although the traditional reservoir volumetric methods utilize the gas and oil formation volume factors (Bg and FVF) to deal with volume changes in petroleum fluids to account for the variations due to phase behavior, the evaporative loss remains un-addressed. If the loss is incorporated in the reservoir volumetric approaches, the estimated volumes of oil and gas should be close to those from the integrated geochemical approach proposed in this study.

    Sample # Leco TOC (wt.%) S1 (mg HC/g rock) Oil density (kg/m3) reservoir Gas density (kg/m3) surface FVF Oil density (kg/m3) surface Total HC (mg/g rock) Gas in sorption (mg/g rock) Gas loss from S1 (mg/g rock) Oil (mg/g rock) Total HC (kg/m3 rock) Gas (kg/m3 rock) Oil (kg/m3 rock) GORin mass (frac.)
    1 2.61 2.41 433 0.72 2.5 773 2.90 1.15 1.64 1.26 7.23 4.08 3.14 0.57
    2 3.56 9.19 433 0.72 2.5 773 12.73 1.40 4.95 7.79 32.17 12.50 19.67 0.39
    3 2.96 2.87 433 0.72 2.5 773 3.53 1.25 1.90 1.62 8.87 4.78 4.08 0.54
    4 2.56 8.26 433 0.72 2.5 773 11.51 1.16 4.41 7.10 29.23 11.19 18.04 0.38
    5 2.57 3.03 433 0.72 2.5 773 3.82 1.16 1.95 1.87 9.70 4.95 4.75 0.51
    6 2.32 8.21 433 0.72 2.5 773 11.47 1.11 4.37 7.10 29.42 11.20 18.22 0.38
    7 4.67 14.63 433 0.72 2.5 773 20.58 1.65 7.60 12.98 50.86 18.78 32.08 0.37
    8 4.79 12.27 433 0.72 2.5 773 17.09 1.69 6.51 10.58 42.52 16.20 26.33 0.38
    9 4.37 4.76 433 0.72 2.5 773 6.11 1.60 2.95 3.16 15.33 7.39 7.93 0.48
    10 3.86 6.21 433 0.72 2.5 773 8.31 1.46 3.56 4.75 20.67 8.85 11.82 0.43
    11 3.35 2.55 433 0.72 2.5 773 3.00 1.33 1.78 1.22 7.46 4.42 3.03 0.59
    12 4.19 11.42 433 0.72 2.5 773 15.93 1.55 6.05 9.87 39.87 15.16 24.71 0.38
    13 4.39 9.99 433 0.72 2.5 773 13.79 1.58 5.38 8.41 34.09 13.30 20.78 0.39
    14 5.34 7.63 433 0.72 2.5 773 10.19 1.80 4.36 5.83 25.03 10.72 14.31 0.43
    15 1.88 1.51 433 0.72 2.5 773 1.68 1.01 1.18 0.50 4.37 3.07 1.30 0.70
    16 3.66 7.36 433 0.72 2.5 773 10.03 1.40 4.07 5.96 24.88 10.11 14.78 0.41
    Average 3.567 5 7.018 75 433 0.72 2.5 773 9.54 1.39 3.92 5.62 23.86 9.79 14.06 0.46

    Table 3.  Results of estimated resource potentials (per gram rock and per unit rock volume), and showing the geochemical data, measured and inferred reservoir fluid properties that are the basis for the resource estimates in the 14-16-62-21W5 well

    Careful examination of individual samples finds that the traditional reservoir volumetric approaches produce much less variations in resource estimates as compared with those from the integrated approach that incorporates geochemical and reservoir engineering parameters (Fig. 9). The much smoother resource intensity in the shale interval from reservoir volumetric methods could be a combined effect of uneven loss of gaseous hydrocarbon in samples and migration of light hydrocarbons from organic-rich interval towards coarser interbeds, and this is further mitigated by the use of a single gas formation volume factor to all samples. Calculated percentage of gas in reservoir by the proposed method appears to be less than the results from laboratory petroleum fluid analysis, which is consistent with the observation of nanopore confinement and fluid compositional redistribution in source rock reservoirs due to strong interaction between hydrocarbon fluids and organic-rich host rocks (e.g., Baek and Akkutlu, 2019; King, 2015; Whitson and Sunjerga, 2012).

    Figure 9.  Comparison of petroleum resources derived from bulk geochemical data (b) and from generalized reservoir volumetric calculation (c).

    Gas to oil ratio is primarily a function of thermal maturity and type of kerogen. For the same source rock at a similar maturity, GOR should remain at about the same level. However, the estimated GOR from reservoir volumetric approach shows a clear dependency on oil and total hydrocarbon resources (Figs. 10a, 10b), which is negatively correlated with both. This dependency is not true, and is likely an artificial effect of a flawed assumption in gas saturation calculation (Sgas=1–SoilSwater) in laboratory. As large portion of natural gas in pore space has been lost during coring, gas saturation is estimated by incompressible water and oil in laboratory. For a given water saturation, the smaller the oil saturation is, the greater the gas saturation is assigned regardless of the true ratio of gas to oil. This complementary nature in oil and gas saturation calculation blurs the variations in resource density, leading to a narrower resource intensity distribution in assessment. This is clearly demonstrated by the comparison of resource values derived from the traditional reservoir volumetric approach in Table 4 and the proposed integrated approach in Table 3 numerically and in Figs. 10c and 10d graphically. The complementary effect of oil and gas saturation in laboratory determination was exaggerated by the light hydrocarbon loss in the samples, resulting in nearly constant hydrocarbon intensity (Fig. 9c). In contrast, the variations of estimated resource intensity from the proposed method vary widely, a reflection of the very nature of reservoir heterogeneity in shale.

    Sample # Gas-filled porosity (%) Oil saturation (%) Porosity (%) Water saturation (%) Bulk density (103kg/m3) HC saturation (%) HCPV (frac.) Oil density in pore (kg/m3) HC mass (mg HC/g rock) GOR Bg Gas (kg/m3 rock) Oil (kg/m3 rock) Total HC (kg/m3 rock) GORin mass (frac.)
    1 3.82 5.48 6.66 37.20 2.49 62.80 0.042 137.60 5.76 430 0.003 32 9.20 2.82 12.03 0.77
    2 3.55 14.03 6.58 31.95 2.53 68.05 0.045 137.60 6.16 430 0.003 32 8.56 7.13 15.69 0.55
    3 2.89 22.34 5.68 26.77 2.51 73.23 0.042 137.60 5.72 430 0.003 32 6.97 9.81 16.78 0.42
    4 2.54 11.77 5.94 45.45 2.54 54.55 0.032 137.60 4.46 430 0.003 32 6.12 5.40 11.53 0.53
    5 2.46 21.79 5.85 36.21 2.54 63.79 0.037 137.60 5.13 430 0.003 32 5.92 9.85 15.77 0.38
    6 3.31 21.06 6.71 29.60 2.57 70.40 0.047 137.60 6.50 430 0.003 32 7.98 10.92 18.90 0.42
    7 2.76 32.48 5.83 20.14 2.47 79.86 0.047 137.60 6.40 430 0.003 32 6.65 14.63 21.29 0.31
    8 2.28 31.91 5.57 27.08 2.49 72.92 0.041 137.60 5.59 430 0.003 32 5.51 13.74 19.25 0.29
    9 2.95 15.59 5.07 26.21 2.51 73.79 0.037 137.60 5.15 430 0.003 32 7.11 6.11 13.22 0.54
    10 3.56 18.43 6.14 23.55 2.49 76.45 0.047 137.60 6.46 430 0.003 32 8.59 8.75 17.34 0.50
    11 3.19 20.92 5.89 24.98 2.49 75.02 0.044 137.60 6.08 430 0.003 32 7.69 9.53 17.22 0.45
    12 3.02 25.32 6.02 24.59 2.50 75.41 0.045 137.60 6.25 430 0.003 32 7.27 11.79 19.06 0.38
    13 3.58 17.41 6.81 30.06 2.47 69.94 0.048 137.60 6.56 430 0.003 32 8.62 9.17 17.79 0.48
    14 3.00 17.94 5.20 24.33 2.46 75.67 0.039 137.60 5.41 430 0.003 32 7.23 7.21 14.44 0.50
    15 2.46 21.83 4.51 23.62 2.60 76.38 0.034 137.60 4.74 430 0.003 32 5.93 7.61 13.53 0.44
    16 3.69 20.55 6.14 19.30 2.48 80.70 0.050 137.60 6.82 430 0.003 32 8.90 9.75 18.65 0.48
    Average 3.07 19.93 5.91 28.19 2.51 71.81 0.042 137.60 5.82 430 0.003 32 7.39 9.01 16.40 0.46
    Variance 0.236 88 45.824 14 0.395 937 46.311 82 0.001 c522 46.311 82 0.000 0 0.486 594 0 0 1.376 007 9.070 748 7.993 372 0.012 316

    Table 4.  Resource estimates from traditional reservoir volumetric approach showing the volumetric parameters and estimation results for each sample. HCPV. Hydrocarbon pore volume.

    Figure 10.  Comparison of estimated total hydrocarbon (oil and gas) (a), and oil resources (b) with gas/oil in mass. Comparison of estimated total resource intensity (oil+gas) from reservoir volumetric (RV) and the proposed integrated mass balance method (MB) (c), and oil estimates in (d).

    It is noteworthy that the constructed resource intensity profile in Fig. 9 shows a tendency of light oil and gas migrating from organic-rich intervals to more permeable zones upwards as indicated by two intervals (3 265 to 3 287 m) of high TOC and S2 below and less organic-rich interval but high OSI and PI (production index) interval (3 255 to 3 265 m) above in Fig. 9, demonstrating the enrichment of free oil in upper part of the Duverany Formation in the studied well, similar to the migration effect reported by Leythaeuser et al. (1984) in source rock beds.

  • The gas sorption in sample is estimated based on assumptions that clay mineral and organic matter adsorb natural gas and the quantity of sorption can be estimated from empirical relationship from laboratory tests (Chen et al., 2017). Mineral composition in Duvernay shale in the west basin is dominated by quartz, and clay mineral accounts for less than 20%. This is can be seen from the empirical relationship in Eq. (5) with α=0.198 58 (mg gas/g rock), and β=0.083 39. As a comparison with other shale reservoirs, for example, Yu et al. (2016) reported an empirical relationship between TOC and gas adsorption capacity with α=0.0 and β=0.528 for the Marcellus shale of the United States, suggesting insignificant adsorption capacity from clay or other minerals, but much stronger adsorption capacity from organic matter.

    From the empirical relationship in Eq. (5), the estimated gas adsorption is less than TOC values (Table 3), implying that the natural gas adsorption capacity of the Duvernay is less than the generalized equivalent TOC value as suggested by Jarvie (2012). The lower gas sorption in a shale means is, the more free hydrocarbons for production. Jarvie (2012) showed that most of the tight and shale oil production intervals in North America have oil saturation index (OSI=S1/TOC×100) greater than 100. Chen et al. (2019) used the quantity of free oil above the OSI of 100 mg HC/g rock as movable oil in their oil mobility evaluation. From Fig. 4c, it is apparent that liquid hydrocarbon production most likely happens in oil window with a Tmax ranging from 440 to 475 ºC. The Tmax values of the studied samples in the Duvernay Formation vary from 460 to 471 ºC suggesting that the 14-16-62-21W5 well is a liquid rich well in the early condensate stage.

  • The bulk geochemical data and properties of reservoir and produced fluids are complementary in evaluating shale resource plays. Based on mass balance and phase behavior of pore fluids in different physical conditions, the integrated approach has the capability of restoring the information on light hydrocarbon loss for resource calculation. This is in contrast to laboratory hydrocarbon saturation calculation that relies on incomplete information from observable oil residual in sample only. Furthermore, by adding the evaporatively-lost light and gaseous components back to the pore fluids, the resulting resource estimate provides a less biased spectrum of hydrocarbon composition, useful for studying oil mobility in a liquid-rich resource play. While restoring the mass of hydrocarbon pore fluids in reservoir condition allows for a more accurate estimation of resources in shale reservoir, the proposed approach can further be used to estimate shale oil mobility by integrating with compositional characterization and evaporative kinetics of the existing petroleum fluids in the samples (Li et al., 2019). However, since the method relies on FVF estimation that is sensitive to maturity, the method is most suitable for shale reservoir in oil window. Chen et al. (2018) and Li et al. (2020) provided mass balanced constraints in FVF estimation for shale resource reservoir with a thermal maturity close to FVF > 2.0 equivalent. As a number of methods for evaluating shale resource play in wet and dry gas windows are available, here in this study we have focused primarily on liquid rich shale play in oil window.

  • An integrated mass balance based approach is proposed for resource calculation in liquid-rich shale play. The proposed method incorporates relevant information from bulk geochemical data and reservoir and production parameters, and considers gaseous hydrocarbon loss during coring and evaporative loss during sample storage and preparation prior to laboratory analysis. The method restores the mass of lost gaseous and light hydrocarbons in samples to avoid the use of a flawed assumption of composition complementary inference on gas saturation calculation.

    The method was applied to the Duvernay shale in 14-16-62-21W5 well to evaluate the resource intensity in the Kaybob area of the West Shale Basin. The resulting resource intensity estimates from the proposed method are compared with those from traditional reservoir volumetric approaches, with the following conclusions.

    (1) The integrated mass balance approach yields a higher resource potential than that from the traditional reservoir volumetric method. The volumetric method is based on lab hydrocarbon saturation calculation following the complementary law of the pore fluids. When applied to liquid-rich shale resource play, this assumption is flawed because parts of the gaseous and light hydrocarbon components have been lost during coring and storage, resulting in an underestimation of the resource potential.

    (2) The new method can better capture reservoir heterogeneity and reveal more vertical variations in resource intensity. The estimated GOR from the proposed method is close to the measured property of initial production petroleum fluids. In contrast, the reservoir volumetric method generates a smoother resource intensity profile that is likely an effect of the flawed assumption of complementary oil and gas saturations in case of the occurrence of fluid loss, and a false positive correlation between the estimated GOR and resource intensity that should not be the case in real production interval.

    (3) With the lost light and gaseous hydrocarbon components added back to the compositional spectrum of the resources, the targeted hydrocarbon fluids may display better mobility in shale reservoir than that revealed directly by free hydrocarbon contents in S1 peaks from the programmed pyrolysis analysis.

  • By definition, formation volume factor (FVF) is a ratio of oil volume in reservoir condition (VoilR) to the oil volume in surface condition (VoilS). When we replace volume by the ratio of mass/density, we can express FVF in the following form

    where moilR is the oil mass in reservoir condition and contains dissolved natural gas; moilS is the oil mass at surface condition without natural gas; ρoilR and ρoilS are oil densities in reservoir and surface respectively.

    Note that moilS is the oil mass that remains in sample under standard surface condition (in mg HC/g rock), and can be approximated by the sum of observation S1 from pyrolysis and the estimated light oil loss during sampling, storage and analysis preparation (S1LP). By replacing (moilS) with (S1+S1LP), Eq. (A1) can be rewritten as

    Re-arranging Eq. (A2) by multiplying ρoilR/ρoilS in both sides, the expression in Eq. (A2) becomes

    In fact, the term moilR (in mg HC/g rock) in right side of Eq. (A3) is the oil mass in reservoir condition with dissolved natural gas. The difference in mass between these two oil quantities in surface and reservoir represents the gas lost (mgasL) during coring

  • The total petroleum mass (for a single phase and gas is dissolved in oil) in reservoir can be written as the sum of two separate phases at surface condition

    To link to oil density in reservoir, we formulate the gas components as the difference between the mass of oil and gas in reservoir and the mass in a single oil phase at surface

    By re-arranging Eq. (B2), we have

    Note that

    We replace VgasS/VoilR by GOR/FVF, and the oil density in reservoir condition (ρoilR) can be estimated from the following equation

  • This study is an output from Geoscience for New Energy Supply Program of the Natural Resources Canada. The authors are grateful to Yoho Energy Ltd. for providing all data collected from 14-16-62-21W5 well. We thank Dr. Maowen Li of Sinopec and two anonymous reviewers for their helpful comments and suggestions. Our internal reviewer, Dr. Andy Mort of Geological Survey of Canada, is thanked for his helpful and constructive comments. This is NRCan contribution (No. 20200522). The final publication is available at Springer via https://doi.org/10.1007/s12583-020-1088-1.

Reference (33)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return