Data analysis and property modeling with geostatistics| Training course

Learn the basic concepts and methods underlying the use of geostatistics for data QC, surface mapping, property modeling, and uncertainty quantification.

Learning objectives

  • Get introduced to the different uses of geostatistics for Oil & Gas applications.
  • Learn the theory and practice of classical geostatistics methodologies for modeling, with their pros and cons.
  • Learn how to integrate various data sources in the interpolation process to improve mapping quality.

Outlines

  • The course is devoted to theoretical and methodological presentations, the second part to practical exercises on real-life cases to deepen the understanding of concepts. The focus is on illustrations and practical contributions of the covered concepts.
  • Computer exercises with Isatis.neo Petroleum Edition.
  • Course material provided (documentation, journal files, training data, worked examples) for re-use in your workplace.

Who should attend

This course aims at any geoscientist or reservoir engineer involved in the building or use of geomodels who want a practical, synthetic, and pragmatic introduction to geostatistical methods for reservoir characterization, filtering, data analysis.

Content

SESSION 1: DATA VARIABILITY ANALYSIS

  • Overview of geostatistics applications for Oil & Gas.
  • Reminders about various statistical graphs and charts (histograms, mean, variance, box-plots, cross-plots, swath-plots, linear regression, etc.).
  • Importance of data sampling, model resolution scales.
  • Data quality control. Identification of possible data outliers, anisotropies, trends, etc.

SESSION 2: UNDERSTANDING AND ESTIMATION OF SPATIAL HETEROGENEITY  OF THE STUDIED PHENOMENON

  • A practical introduction to the concepts of heterogeneity and spatial variability.
  • Quantification of the spatial variability: calculation, interpretation, and modeling of the variogram.

SESSION 3: MAPPING AND UNCERTAINTY ANALYSIS

  • Mapping of a continuous variable (e.g. depth, porosity)
    – 
    Review of classic deterministic interpolation methods.
    – Kriging (2D/3D). Principles and properties. Map uncertainty.
    – Integration of one or several secondary (i.e. dense seismic data) or fuzzy data in the interpolation. Analysis of the correlations between variables. Multivariate variogram. Co-kriging.
  • Uncertainty quantification
    – Introduction to the conditional simulations for uncertainty quantification.
    – Difference between kriging and conditional simulations.
    – Probabilistic maps.
    – Risk analysis.
  • Non-stationary geostatistics and trend modeling
    Trends and residuals.
    Kriging with external drift.
    – Application to depth conversion and setup of the structural model.

SESSION 4: PRACTICAL EXERCISES ON REAL CASES WITH ISATIS.NEO AND ITS WORKFLOW ON TIME-TO-DEPTH CONVERSION 

Prerequisites

No prior knowledge about geostatistics is required. The course is ideal for newcomers to geostatistics or for someone wanting a refresher.

These sessions may be complemented by one or two other modules that delves into some of the most advanced geostatistical techniques, giving hints and pitfalls on how to use them: “Seismic Data Filtering and Depth Conversion with Geostatistics” and “Advanced Geostatistics for Reservoir Characterization”.