Quantitative Geosciences: Data Analytics, Geostatistics, Reservoir Characterization and Modeling

Earth science is becoming increasingly quantitative in the digital age. Quantification of geoscience and engineering problems underpins many of the applications of big data and artificial intelligence. This book presents quantitative geosciences in three parts. Part 1 presents data analytics using probability, statistical and machine-learning methods. Part 2 covers reservoir characterization using several geoscience disciplines: including geology, geophysics, petrophysics and geostatistics. Part 3 treats reservoir modeling, resource evaluation and uncertainty analysis using integrated geoscience, engineering and geostatistical methods. As the petroleum industry is heading towards operating oil fields digitally, a multidisciplinary skillset is a must for geoscientists who need to use data analytics to resolve inconsistencies in various sources of data, model reservoir properties, evaluate uncertainties, and quantify risk for decision making. This book intends to serve as a bridge for advancing the multidisciplinary integration for digital fields. The goal is to move beyond using quantitative methods individually to an integrated descriptive-quantitative analysis. In big data, everything tells us something, but nothing tells us everything. This book emphasizes the integrated, multidisciplinary solutions for practical problems in resource evaluation and field development.

Author Biography: Dr. Ma is a scientific advisor for geosciences at Schlumberger, specialized in reservoir characterization, modeling and resource evaluation. In his over 30 years of experience, he has worked on research and application of statistics, data analytics, and geostatistics to integrated reservoir studies for major oil companies in Europe and the US and has provided technical consultancies and training worldwide. Dr. Ma has published over 100 technical papers or book chapters in petroleum geology, geophysics, engineering, geostatistics, applied statistics and economics, and has received numerous awards, including the Schlumberger's Gold Award and Chairman Award, and the Mathematical Geosciences' Best Paper. Dr. Ma has earned a PhD in Mathematical Geology and Geoinformatics from Universite de Lorraine (previously Institute National Polytechnique de Lorraine), France, and MSc in Geostatistics from Ecole des Mines de Paris, France, and a BSc in Geology from China University of Geosciences.

Preface 1. Introduction and Overview Part 1: Reservoir Characterization 2. Essential Reservoir Geology and Multi-Scales of Petroleum Reservoir Heterogeneities2.1 Structural Controls on Petroleum Resources2.2 Sequence Stratigraphy and Hydrocarbon Resources2.3 Depositional Environments2.4 Facies and Lithofacies2.5 Petrophysical Properties2.6 Subsurface Fluid Heterogeneities2.7 SummaryAppendix 1: Large-Scale Tectonic Settings and Their CharacteristicsAppendix 2: Hierarchy of Depositional Systems: Fluvial Architecture Example 3. Introduction to Petrophysical Reservoir Characterization3.1 Porosity Characterization and Estimation3.2 Clay Volume Analysis and Its Impacts on Other Petrophysical Parameters3.3 Permeability Characterization and Estimation3.4 Fluid Saturations Characterization and Estimation3.5 Uncertainty Analysis in Petrophysical Analysis3.6 Summary 4. Practical Seismic Reservoir Characterization4.1 Seismic Data: Resolution and Coverage4.2 Structural and Stratigraphic Interpretations4.3 Reservoir Delineation, EOD Mapping and Seismic Facies Identifications4.4 Reservoir Property Evaluation Using Seismic Data4.5 Seismic Attribute Analysis and Integration for Reservoir Characterization4.6 SummaryAppendix 1: Time-Frequency Representation of Seismic DataAppendix 2: Amplitude Versus Offset (AVO): An OverviewAppendix 3: Overview of Forward Modeling and Seismic Inversions 5. Statistical and Data Analytical Reservoir Characterization5.1 Common Descriptive Statistics for Reservoir Analysis5.2 Sampling Bias in E&P and Mitigation Methods5.3 Multivariate Statistical Data Analysis and Applications to Reservoir Analysis (Statistical Correlation, PCA, and Clustering Analysis)5.4 Overview of Artificial Neural Networks and Example Applications5.5 Bayesian Inference for Reservoir Characterization5.6 Regression for Mapping and Modeling of Reservoir Properties5.7 Advanced Regressions for Integrated Reservoir Characterization5.8 SummaryAppendix 1: Lord's Paradox and Reconciling Mathematics and Reservoir ProblemsAppendix 2: Impact of Interdependencies/collinearity on multiple linear regressions 6. Geostatistical Reservoir Characterization6.1 Variogram and Spatial Correlation6.2 Theoretical Variogram and Spatial Covariance Models6.3 Calculating and Fitting Experimental Variograms6.4 Interpreting Variograms of Reservoir Properties6.5 Lithofacies Variography and Indicator Variogram6.6 Cross-Variogram, Spatial Misalignment and Synchronization6.7 Relationships between Variogram/Covariance Function and Spectrum6.8 Summary 7. Integrated Facies and Lithofacies Analysis, Identification and Classification7.1 Geological Interpretation of Facies7.2 Lithofacies Classification Using Wireline Logs7.3 Integration of Geological Facies, Seismically Derived Facies, Well-Log-Derived Facies and Core Facies7.4 Summary Part 2: Geological and Reservoir Modeling 8. Constructing a Reservoir-Model Framework8.1 From Structural Elements to a Structural Model8.2 From Stratigraphic Elements to a Stratigraphic Model8.3 Building Depositional Geometrics into Geocellular Model 9. Geostatistical Modeling Methods9.1 Estimation Methods9.1.1 Simple Kriging, Ordinary Kriging, Nonstationary Kriging9.1.2 Factorial Kriging and Decomposition of Sub-Processes9.1.3 Cokriging and Collocated Cokriging9.2 Stochastic Simulation9.2.1 Spectral Simulations9.2.2 Sequential Gaussian Simulation9.3 SummaryAppendix 1: Ergodicity and Micro-Ergodicity in Stochastic ProcessesAppendix 2: Stationarity, Local Stationarity and Intrinsic Random FunctionsAppendix 3: Examples of Decompositions of Sub-Processes Using Factorial Kriging 10. Facies and Lithofacies Modeling10.1 Definition of Lithofacies and Composite Lithofacies for Modeling10.2 Lithofacies Spatial Trends and Probabilities10.3 Lithofacies Modeling Methods10.4 Constructing Facies and Lithofacies in a Reservoir Model10.5 Multi-Level Modeling of Facies and Lithofacies10.6 Summary 11. Porosity Modeling11.1 Statistical Analysis of Porosity Data11.2 Spatial Characterization of Porosity Distributions11.3 Modeling Porosity11.3.1. Using Kriging and Other Interpolation Methods11.3.2 Using Stochastic Simulation11.3.3 Trend-Integrated Modeling of Porosity11.3.4 Seismically Integrated Co-Simulation of Porosity11.3.5 Depositional-Geometry-Honored Porosity Modeling11.3.6 Hierarchical Modeling of Porosity11.3.7 Modeling Porosity with Multiple Constraints11.4 Summary 12. Permeability Modeling12.1 Statistical Analysis of Permeability Data12.2 Permeability Modeling by Regression12.3 Permeability Modeling by Cloud Transform12.4 Permeability Modeling by Cokriging and Cosimulation12.5 Summary 13. Fluid-Saturation Modeling13.1 Fluid Distributions in a Reservoir13.2 Data Sources for Fluid Characterization13.3 Modeling Water Saturation13.3.1 Using Methods of Sw-Height Functions13.3.2 Using Cokriging and Cosimulation13.4 Summary 14. Uncertainty Analysis and Volumetrics Evaluation14.1 General Issues14.1.1 Relationship between Variability and Uncertainty14.1.2 Relationship between Uncertainty and Error14.1.3 Measurement Uncertainty14.1.4 Interpretation Uncertainty14.1.5 Subsurface Fluid Uncertainty14.1.6 Other Inference Uncertainties Related to Reservoir Analysis14.1.7 Value of Information in Uncertainty Analysis14.2 Uncertainty Quantification in Petroleum Resource Volumetrics14.2.1 Critics on the Classical Volumetric Calculations14.2.2 Analytical Estimation of Hydrocarbon Volumetrics14.2.3 3D Model-Based Hydrocarbon Volumetrics14.2.4 Defining Uncertainties of Input Parameters14.2.5 Critics of Monte Carlo Simulation for Uncertainty Analysis14.2.6 Integrated Uncertainty Quantification Workflow14.2.7 Evaluating Uncertainty Workflow Results14.2.8 Tranferring Static Uncertainties into Dynamic Uncertainty Evaluation14.3 Decision Analysis14.3.1 Known Knowns, Known Unknowns and Unknown Unknowns14.3.2 Methods for Reducing Uncertainties14.3.3 Decision Analysis Under Uncertainty in E&P14.4 SummaryAppendix 1: Probability Distributions for Describing Uncertainties of Reservoir Variables Part 3: Special and Advanced Topics 15. Naturally Fractured Reservoir Characterization and Modeling15.1 Faulting and Fracturing in Subsurface Formations15.2 Characterizing Fractured Formations15.3 Construction of Discrete Fracture Networks (DFN)15.4 From DFN to Continuous Reservoir Properties15.5 Summary 16. Updating a Reservoir Model and Feedback Loop in Reservoir Modeling16.1 Integrating New Data16.2 Feedback Loop in Reservoir Modeling16.3 Production Data Integration16.4 4D Seismic Data Monitoring and Integration 17. Ranking Reservoir Models 18. Reservoir Model Upscaling, Simulation and Validation18.1 Upscaling 3D Model Grids18.2 Upscaling Categorical Variables18.3 Upscaling Static Reservoir Properties18.4 Mass-Preservation Upscaling18.5 Upscaling Dynamic Reservoir Properties18.6 Reservoir Simulation and Model Validations 19. Common and Uncommon Pitfalls in Integrated Reservoir Characterization and Modeling 20. Planning an Integrated Reservoir Characterization and Modeling Project< 21. Towards a Fully Integrated Reservoir Characterization, Modeling and Uncertainty Analysis for Petroleum Resource Management and Field Development