Abstract:

The concepts of geostatistics are commonly difficult to understand for students, even for those with a strong mathematical background. Furthermore first time users of geostatistics often do not want to struggle through different books with the whole geostatistical background before using it. Too often, this leads to using geostatistics as a black box, i.e. the data is imported in a software program, which calculates an experimental semivariogram, determines the ‘best’ semivariogram model and computes kriging estimates. These results are then claimed to be the only and best results, because they are based on geostatistics, while there is a significant risk that some basic errors are introduced in the study.
This paper describes a number of generally accessible R-based (www.r-project.org) interactive modules, developed by the authors. These modules can be used in geostatistical courses, but also by an individual user to learn more easily the key concepts of geostatistics. These modules should allow the student or practitioner to see, in real time, the impact of deleting, moving or adding data points, changing parameters (e.g. lag distance, lag tolerance, directional tolerance for calculating an experimental semivariogram), and so on.
The developing of these modules is based on experience with web-based tools for teaching statistical concepts, namely ‘Vestac’ and ‘env2exp’ (see http://lstat.kuleuven.be/java/ and http://lstat.kuleuven.be/env2exp/). Those tools got several good critics and are already used in statistical courses all over the world.
The modules are completely autonomous, i.e. it is not required to have a dataset of your own. Those modules are therefore not meant to do a geostatistical study on a certain dataset (for this purpose there exists already a broad spectrum of good software packages). After learning the key concepts of geostatistics it should be easier for the user to do a good quality geostatistical study on their own dataset and to do a good interpretation of these results.
In this paper three different modules are discussed. The first module allows interactive experimentation with the experimental semivariogram of different datasets in one direction. Here the user can learn to understand the experimental semivariogram, but also about the effect of the number of sampling points, the length of the dataset, and other parameters. The second module goes a bit further than the first, this time the datasets are two-dimensional. This module does not only focus on how to calculate the semivariogram, but it also looks at modelling the semivariograms. The third module is about Kriging with two-dimensional datasets. One can choose sampling points, an unknown point and a semivariogram model, to estimate the value in the unknown point. The results of this kriging procedure are compared with the results when the estimation equals the average of the sampling points. In this module one can learn about the effect of the sampling positions, the position of the unknown point and the semivariogram model on the estimations results.

Keywords:

Geostatistics, interactive modules.