Methodology

Best practice in geostatistical methodology will often choose methods/techniques that are simple, robust and within the level of available expertise. The methods, techniques, and analysis should be appropriate to the goals of the study. We should be able to explain the background and critical assumptions behind every technique and provide documentation for critical input parameters. Geostatistical models are constructed more than once and we must take care and compare with historical methods and justify current choices.

Methodology lessons are intended to help both relative newcomers and experienced geomodelers scope out a geostatistical study, choose specific workflows and tools, make implementation decisions and support the numerous interdependent decisions that must be made during a geostatistical study.

Lessons

• An Application of Bayes Theorem to Geostatistical Mapping (see source code on GitHub)
• Combination of Multivariate Gaussian Distributions through Error Ellipses
• Bayesian Updating for Combining Conditional Distributions
• Locally Varying Anisotropy
• The Multivariate Spatial Bootstrap
• A Simulation Approach to Calibrate Outlier Capping
• The Decision of Stationarity
• The Nugget Effect (see source code on GitHub)
• Multivariate Gaussian Distribution

• Gaussian Mixture Models
• Stratigraphic Coordinate Transformation
• Trend Modeling and Modeling with a Trend (see source code on GitHub)
• Cokriging with Unequally Sampled Data (see source code on GitHub)
• Conditioning by Kriging
• Aggregating Variables into a Super Secondary Variable
• Calculation and Modeling of Variogram Anisotropy
• Implicit Boundary Modeling with Radial Basis Functions
• Introduction to Choosing a Kriging Plan (see source code on GitHub)
• The Sill of the Variogram
• Kriging with Constraints
• Decision Making in the Presence of Geological Uncertainty
• Multidimensional Scaling (see source code on GitHub)
• Permanence of Ratios
• Sequential Indicator Simulation (SIS)
• Quantitative Kriging Neighborhood Analysis (QKNA)
• Choosing the Discretization Level for Block Property Estimation
• Categorical Variable Distributions in Geostatistics
• Collocated Cokriging (see source code on GitHub)
• Kriging Weights in the Presence of Redundant Data
• Localization of Probabilistic Resource Models
• Change of Support and the Volume Variance Relation (see source code on GitHub)