Predicting total organic carbon from few well logs aided by well-log attributes

Abstract

Derivative/volatility well-log attributes from very few commonly recorded well logs can assist in the prediction of total organic carbon (TOC) in shales and tight formations. This is of value where only limited suites of well logs are recorded, and few laboratory measurements of TOC are conducted on rock samples. Data from two Lower-Barnett-Shale (LBS) wells (USA), including well logs and core analysis is considered. It demonstrates how well-log attributes can be exploited with machine learning (ML) to generate accurate TOC predictions. Six attributes are calculated for gamma-ray (GR), bulk-density (PB) and compressional-sonic (DT) logs. Used in combination with just one of those recorded logs, those attributes deliver more accurate TOC predictions with ML models than using all three recorded logs. When used in combination with two or three of the recorded logs, the attributes generate TOC prediction accuracy comparable with ML models using five recorded well logs. Multi-K-fold-cross-validation analysis reveals that the K-nearest-neighbour algorithm yields the most accurate TOC predictions for the LBS dataset. The extreme-gradient-boosting (XGB) algorithm also performs well. XGB is able to provide information about the relative importance of each well-log attribute used as an input variable. This facilitates feature selection making it possible to reduce the number of attributes required to generate accurate TOC predictions from just two or three recorded well logs.

Keywords

TOC well-Log relationships / Log attribute influences / Log curve derivatives / Moving average volatility / Effective attribute combinations

References

[1] C. Vega-Ortiz, D.R. Beti, E. Setoyama, J.D. McLennan, T.A. Ring, R. Levey, N. Martínez-Romero, Source rock evaluation in the central-western flank of the Tampico Misantla Basin, Mexico, J. South Am. Earth Sci. 100 (2020), 102552.

[2] R.F. Beers, Radioactivity and organic content of some Paleozoic shales, Am. Assoc. Petrol. Geol. Bull. 26 (1945) 1-22.

[3] V.E. Swanson,Oil Yield and Uranium Content of Black Shales, 1960. USGS Professional Paper 356-A: 1-44.

[4] B.L. Meyer, M.H. Nederlof, Identification of source rocks on wireline logs by density/resistivity and sonic transit time/resistivity cross plots, Am. Assoc. Petrol. Geol. Bull. 68 (1984) 121-129.

[5] U. Mann, Relation between source rock properties and wireline log parameters, an example from Lower Jurassic Posidonia Shale, NW Germany, Org. Geochem. 10 (4-6) (1986) 1105-1112.

[6] W.H. Fertl, G.V. Chilingar, Total organic carbon content determined from well logs, SPE Form. Eval. 3 (1988) 407-419, https://doi.org/10.2118/15612-PA.

[7] J. Schmoker, Determination of organic content of appalachian devonian shales from formation-density logs, geologic notes, AAPG (Am. Assoc. Pet. Geol.) Bull. 63 (1979) 1979.

[8] T. Hester, J. Schmoker, Determination of Organic Content from Formation-Density Logs, Devonian-Mississippian Woodford Shale, United States Geological Survey USGS Open file, Anadarko Basin, Oklahoma, 1987 report #87-20.

[9] M.S. El Sharawy, G.G. Gaafar, Application of well log analysis for source rock evaluation in the Duwi Formation, Southern Gulf of Suez, Egypt, J. Appl. Geophys. 80 (2012) 129-143.

[10] J.D. Mendelson, M.N. Toksoz,Source rock characterization using multivariate analysis of log data, in: Transactions of 26thAnnual Logging Symposium, 1985, 17-20 June, Dallas, Texas, Paper SPWLA-1985-UU 26 pages.

[11] S. Verma, T. Zhao, K.J. Marfurt, D. Devegowda, Estimation of total organic carbon and brittleness volume, Interpretation 4 (3) (2016) 373-385, https://doi.org/10.1190/INT-2015-0166.1.

[12] M. Alshakhs, R. Rezaee, A new method to estimate total organic carbon (TOC) content, an example from goldwyer shale formation, the canning basin, Open Petrol. Eng. J. 10 (2017) 118-133, https://doi.org/10.2174/1874834101710010118.

[13] Q.R. Passey, F.U. Moretti, J.D. Stroud, A practical modal for organic richness from porosity and resistivity logs, Am. Assoc. Petrol. Geol. Bull. 74 (1990) 1777-1794.

[14] R. Huang, Y. Wang, S. Cheng, S. Liu, L. Cheng, Selection of logging-based TOC calculation methods for shale reservoirs: a case study of the Jiaoshiba shale gas field in the Sichuan Basin, Nat. Gas. Ind. B 2 (2015) 155-161.

[15] Q.R. Passey, K. Bohacs, W.L. Esch, R. Klimentidis, S. Sinha, From oil-prone source rock to gas-producing shale reservoir-geologic and petrophysical characterization of unconventional shale gas reservoirs, in: International Oil and Gas Conference and Exhibition in China, Society of Petroleum Engineers, 2010, p. 29, https://doi.org/10.2118/131350-MS.

[16] R.K. Mallick, S.V. Raju, Thermal maturity evaluation by sonic log and seismic velocity analysis in parts of Upper Assam Basin, India Org Geochem 23 (10) (1995) 871-879.

[17] P. Zhao, Z. Mao, Z. Huang, C. Zhang, A new method for estimating total organic carbon content from well logs, AAPG (Am. Assoc. Pet. Geol.) Bull. 100 8 (2016) 1311-1327.

[18] B. Alizadeh, K. Maroufi, M.H. Heidarifard, Estimating source rock parameters using wireline data: an example from Dezful Embayment, Southwest Iran, J. Petrol. Sci. Eng. 167 (2018) 857-868.

[19] D. Cheng, X. Yuan, C. Tan, M. Wang, Logging-lithology identification methods and their application: a case study on the Chang 7 Member in the centralwestern Ordos Basin, NW China, China Petroleum Exploration 21 ( 5) (2016) 1-10.

[20] M.R. Kamali, A.A. Mirshady, Total organic carbon content determined from well logs using DlogR and neuro fuzzy techniques, J. Petrol. Sci. Eng. 45 (2004) 141-148.

[21] B. Zeng, M. Li, J. Zhu, X. Wang, Y. Shi, Z. Zhu, H. Guo, F. Wang, Selective methods of TOC content estimation for organic-rich interbedded mudstone source rocks, J. Nat. Gas Sci. Eng. 93 (2021), 104064, https://doi.org/10.1016/j.jngse.2021.104064.

[22] S.L. Herron, Source rock evaluation using geochemical information from wireline logs and cores (abs), Am. Assoc. Petrol. Geol. Bull. 72 (1988) 1007.

[23] C. Boyer, J. Kieschnick, R.S. Rivera, R.E. Lewis, G. Waters, Producing gas from its source, Schlumberger Oilfield Review Autumn (2006) 36-49.

[24] M.M. Herron, J. Grau, S.L. Herron, R.L. Kleinberg, M. Machlus, S.L. Reeder, B. Vissapragada, A. Burnham, R. Day, P. Allix, Total organic carbon and formation evaluation with wireline logs in the Green River oil shale, in: SPE-147184-MS SPE Annual Technical Conference and Exhibition, Society of Petroleum Engineers, Denver, Colorado, USA, 2011, p. 19, https://doi.org/10.2118/147184-MS , 30 October-2 November.

[25] D.L. Luffel, F.K. Guidry, J.B. Curtis, Evaluation of Devonian shale with new core and log analysis methods, SPE 21297-PA (1992) 1192-1197, https://doi.org/10.2118/21297-PA.

[26] Y. Wang, The method of application of gamma-ray spectral logging data for determining clay mineral content, J. Oil Gas Technol. (J. Jianghan Petroleum Inst.) 35 (2) (2013) 100-104.

[27] Z. Huang, M.A. Williamson, Artificial neural network modeling as an aid to source rock characterization, Mar. Petrol. Geol. 13 (1996) 227-290.

[28] V. Bolandi, A. Kadkhodaie-Ilkhchi, B. Alizadeh, J. Tahmorasi, R. Farzi, Source rock characterization of the Albian Kazhdumi formation by integrating well logs and geochemical data in the Azadegan oilfield, Abadan plain, SW Iran, J. Petrol. Sci. Eng. 133 (2015) 167-176.

[29] Y. Liu, Z. Chen, K. Hu, C. Liu, Quantifying Total Organic Carbon (TOC) from Well Logs Using Support Vector Regression, Integration GeoConvention, Calgary, 2013, p. 6.

[30] M. Tan, X. Song, X. Yang, Q. Wu, Support-vector-regression machine technology for total organic carbon content prediction from wireline logs in organic shale: a comparative study, J. Nat. Gas Sci. Eng. 26 (2015) 792-802.

[31] X. Shi, J. Wang, G. Liu, L. Yang, X. Ge, S. Jiang, Application of extreme learning machine and neural networks in total organic carbon content prediction in organic shale with wire line logs, J. Nat. Gas Sci. Eng. 33 (2016) 687-702.

[32] D.A. Wood, Total organic carbon predictions from lower Barnett shale well-log data applying an optimized data matching algorithm at various sampling densities, Pure Appl. Geophys. 177 (11) (2020) 5451-5468, https://doi.org/10.1007/s00024-020-02566-1.

[33] D.M. Jarvie, R.J. Hill, T.E. Ruble, R.M. Pollastro, Unconventional shale-gas systems: the Mississippian Barnett Shale of North-Central Texas as one model for thermogenic shale-gas assessment, AAPG (Am. Assoc. Pet. Geol.) Bull. 91 (2007) 475-499, https://doi.org/10.1306/12190606068.

[34] J.L. Walper, Plate tectonic evolution of the Fort Worth Basin, in: C.A. Martin (Ed.), Petroleum Geology of the Fort Worth Basin and Bend Arch Area: Dallas Geological Society, 1982, pp. 237-251.

[35] M. Abouelresh, R. Slatt, Lithofacies and sequence stratigraphy of the Barnett shale in east-central Fort Worth Basin, Texas, AAPG (Am. Assoc. Pet. Geol.) Bull. 96 (1) (2012) 1-22, https://doi.org/10.1306/04261110116.

[36] P. Singh, R.M. Slatt, W. Coffey, Barnett Shale -unfolded: Sedimentology, sequence stratigraphy, and regional mapping, Gulf Coast Association of Geological Societies Transactions 58 (2008) 777-795.

[37] R.M. Pollastro, D.M. Jarvie, R.J. Hill, C.W. Adams, Geologic framework of the mississippian Barnett shale, barnett-paleozoic total petroleum system, Bend archefort Worth Basin, Texas, AAPG (Am. Assoc. Pet. Geol.) Bull. 91 (4) (2007) 405-436, https://doi.org/10.1306/10300606008.

[38] K. Pearson, On the dissection of asymmetrical frequency curves, Phil. Trans. Roy. Soc. A. 185 (1894) 71-110.

[39] C. Spearman, The proof and measurement of association between two things, Am. J. Psychol. 15 (1) (1904) 72-101, https://doi.org/10.2307/141215.

[40] R. Artusi, P. Verderio, E. Marubini, Bravais-Pearson and Spearman correlation coefficients: meaning, test of hypothesis and confidence interval, The International Journal of Biological Markers, 2002 17 (2) (2002) 148-151. https://journals.sagepub.com/doi/pdf/10.1177/172460080201700213.

[41] R. Boddy, G. Smith, Statistical Methods in Practice: for Scientists and Technologists, Wiley, Chichester, U.K., 2009, ISBN 978-0-470-74664-6, pp. 95-96.

[42] L. Myers, M.J. Sirois, Spearman correlation coefficients, differences between, Encyclopedia of Statistical Sciences (2004), https://doi.org/10.1002/0471667196.ess5050.

[43] D.W. Wayne, Spearman Rank Correlation Coefficient. Applied Nonparametric Statistics, second ed., PWS-Kent, Boston, 1990, ISBN 978-0-534-91976-4, pp. 58-365.

[44] D.A. Wood, Gamma-ray log derivative and volatility attributes assist facies characterization in clastic sedimentary sequences for formulaic and machine learning analysis, Advances in Geo-Energy Research 6 (1) (2022) 69-85, https://doi.org/10.46690/ager.2022.01.06.

[45] D.A. Wood, Enhancing lithofacies machine learning predictions with gammaray attributes for boreholes with limited diversity of recorded well logs, Artificial Intelligence in Geosciences 2 (2022) 148-164, https://doi.org/10.1016/j.aiig.2022.02.007.

[46] D.A. Wood, Carbonate/siliciclastic lithofacies classification aided by well-log derivative, volatility and sequence boundary attributes combined with machine learning, Earth Science Informatics (2022), https://doi.org/10.1007/s12145-022-00829-0.

[47] SciKit Learn, Supervised and unsupervised machine learning models in Python, 2022a, https://scikit-learn.org/stable/, 2022, 2022.

[48] F.E. Harrell, Regression Modeling Strategies, second ed., Springer, Switzerland, 2015, p. 582, https://doi.org/10.1007/978-3-319-19425-7.

[49] A.S. Goldberger, Classical Linear Regression. Econometric Theory, John Wiley & Sons, New York, 1964, ISBN 0-471-31101-4, p. 158.

[50] S.M. Stigler, Gauss and the invention of least squares, Ann. Stat. 9 (3) (1981) 465-474, https://doi.org/10.1214/aos/1176345451.

[51] H. Zou, T. Hastie, Regularization and variable selection via the elastic net, Statistical Methodology Series B 67 (2) (2005) 301-320, https://doi.org/10.1111/j.1467-9868.2005.00503.x.

[52] Y. Freund, R.E. Schapire, A decision-theoretic generalization of on-line learning and an application to boosting, J. Comput. Syst. Sci. 55 (1997) 119-139, https://doi.org/10.1006/jcss.1997.1504.

[53] T.K. Ho, The random subspace method for constructing decision forests, IEEE Trans. Pattern Anal. Mach. Intell. 20 (8) (1998) 832-844, https://doi.org/10.1109/34.709601.

[54] E. Fix,J.L. Hodges Jr., Discriminatory Analysis, Nonparametric Discrimination: Consistency Properties, USAF School of Aviation Medicine, 1951. Technical Report.

[55] C. Cortes, V. Vapnik, Support-vector networks, Mach. Learn. 120 (3) (1995) 273-297, https://doi.org/10.1007/BF00994018.

[56] T. Chen, C. Guestrin, XGBoost: a scalable tree boosting system,in:Balaji Krishnapuram, Mohak Shah, Alexander J. Smola, Charu C. Aggarwal, Dou Shen, Rajeev Rastogi (Eds.), Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016, pp. 785-794, https://doi.org/10.1145/2939672.2939785. San Francisco, CA, USA, August 13-17.

[57] C. Gini, Concentration and dependency ratios, published 1909 in Italian, English translation in Rivista di Politica Economica 87 (1997) 769-778.

[58] GridSearchCV, Exhaustive search over specified parameter values for an estimator in Python, SciKit-Learn (2022), 2022, https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html.

[59] BayesSearchCV, Bayesian optimization of hyperparameters by SciKit optimization. https://scikit-optimize.github.io/stable/modules/generated/skopt. BayesSearchCV.html, 2022, 2022.

[60] M. Stone, Cross-validatory choice and assessment of statistical predictions, Journal of the Royal Statistical Society, Series B (Methodological), 1974 36 (2) (1974) 111-147, https://doi.org/10.1111/j.2517-6161.1974.tb00994.

[61] SciKit Learn, Cross-validation: evaluating estimator performance, 2022b, https://scikit-learn.org/stable/modules/cross_validation.html, 2022, 2022.

[62] D.A. Wood,Assessing the brittleness and total organic carbon of shale formations and their role in identifying optimum zones to fracture stimulate. Chapter 5, in: D. A. Wood, J. Cai (Sustainable Geoscience for Natural Gas Sub-surface Systems,Eds.), Elsevier, 2022, pp. 129-157, https://doi.org/10.1016/B978-0-323-85465-8.00014-5.