Geostatistics - Using Software for Geospatial Analysis
Who should attend?

This training course is designed for all professionals working in the field of data analysis, oil and gas exploration, geology and reservoir modelling.

This training course is suitable for a wide range of professionals but will greatly benefit:

  • Data Scientists
  • Data Analysts
  • Geologists
  • Petroleum engineers
  • Reservoir engineers
  • Other professionals involved in spatial analysis and oil and gas exploration

Duration
5 Days
Programme Overview

This training course is an ideal presentation of using statistical methods in spatial data analysis. It presents the concepts used in geoscience with special emphasis on oil and gas exploration. The use of omnipresent Excel for the geospatial analysis for the initial applied research in geology and exploration. Advanced concept of using free R software for geospatial analysis trough examples is also presented, with explaining statistical methods used and the packages.  

This training course is designed to help professionals in data analysis, geologists and oil and gas professionals to remove the limitations of using off-the-shelf software, which is quite helpful but it limits the ability of the professional using it to apply its knowledge and extend the models used, as the readymade software applies pre-designed algorithms and ‘’forces’’ the data into distributions applicable for the models. This training course will allow the professionals to understand how they can use free or low-cost software to extend the capabilities of commercial software and enable them to use their own ingenuity without limitations.   

This training course will feature:

  • Explanation of geoscience and its applications in oil and gas
  • Integrating information from various sources with varying degrees of uncertainty
  • Establishing relationships between measurements and reservoir properties
  • Using semi-variograms and kriging
  • Hands-on practice in using excel and R for spatial data analysis
  • Advanced concepts: Monte Carlo Simulation, k-means, numerical facies modelling, fuzzy logic

Objectives

By the end of this training course, participants will be able to:

  • Learn the concepts and methods of geostatistics
  • Understand the capabilities of Excel and R programming language
  • Acquire the knowledge of available R packages for spatial data analysis
  • Import, analyze and interpret results from spatial data
  • Perform Monte Carlo simulation, clustering analysis and other advanced techniques

Course Outline

Day One: Geostatistics - Concepts and Introduction to Software

  • Basics of Geostatistics
  • Geostatistical reservoir modelling
  • A short introduction to Excel
  • A short introduction to R and R studio
  • Exercise: importing well log data into excel and creating GR vs Depth plot 
  • Exercise: importing well log data into R and initial analysis

Day Two: Spatial Data Analysis

  • Spatial Data Sampling
  • Spatial Resolution Gap
  • Spatial Weight Matrices
  • Basis of data analysis: statistical measures, correlation and autocorrelation
  • Exercise: determining correlation and autocorrelation in well log data using Excel
  • Exercise: Plotting Spatial connectivity

Day Three: Steps in Geostatistics: The Variogram and Kriging

  • Variogram and Modelling
  • Sampling for the Variogram
  • Nested Sampling
  • Geostatistical Prediction: Kriging
  • Exercise: Performing ANOVA in Excel, Kriging Example in Excel
  • •Exercise: Variogram and Kriging in R studio

Day Four: Big Data Analytics and its Relation to Oil and Gas

  • Big Data Concepts
  • Clustering analysis
  • Spatial Variance and Covariance
  • Data distributions
  • Exercise: Variance and covariance calculation in Excel
  • Exercise: Clustering analysis in R studio

Day Five: Advanced Topics in Spatial Statistics

  • Bayesian Theory and Spatial Data
  • Monte Carlo Analysis
  • Markov Chains
  • Exercise: Monte Carlo Simulation for Oil and Gas reserves simulation in Excel
  • Exercise: Monte Carlo Simulation in R
  • Fuzzy Logic, Machine Learning and generative algorithms and the future of prediction



scroll to top