The choice of reasonable geological subsets precedes most geostatistical calculations. Some data screening and exploration is required to understand and work with the data. The coordinate system may be transformed to facilitate the integration of geological data. Then, before geostatistical modeling begins, representative statistics are required. In particular, a distribution (or proportions) are required for every variable within each stationary geological subset.
Preferential sampling is common; data are rarely collected to be statistically representative. Interesting (best) areas are delineated more completely early on in project evaluation. The areas to be mined first (likely higher in grade than the rest of the deposit) are drilled more closely. Wells are drilled for production and not to fairly sample the geological domain. Seismic data and geological trends are used to select the best locations to drill. Core are taken preferentially from good quality reservoir rock. These data collection practices should not be changed; they lead to the best economics and the greatest number of data in portions of the study area that are the most important. There is a need, however, to adjust the histograms and summary statistics to be representative of the entire volume of interest. These lessons focus on best practices in establishing and using representative distributions with uncertainty.