A Comprehensive Geostatistical Tool for PC Platforms
Geostatistics has attracted the attention of many earth scientists and engineers who need better modeling tools for describing reservoirs. In response to this need a new software was developed which combines geostatistical modeling with high quality graphics. The software is designed for PC platforms enabling a large number of engineers and geologists to better comprehend reservoirs and consequently improve their decisions. The software runs on any PC with Windows 95 or Windows NT operating system.
A pre-processing module reads LAS and ASCII files. The pre-processing module facilitates selection of the stratigraphic units prior to processing by a nearest neighbor, Kriging and co-Kriging, conditional simulation, or fractal module. User friendly Graphical User Interface (GUI) simplifies the examination of the statistical data and the geostatistical analysis using variograms. After completing the analysis, the post-processing unit can display 1D models of well logs, 2D models such as cross-sections, or a 3D models of any reservoir property. Post-processing includes the display of reservoir slices, multiple cross-sections, rotation along any axis, and identification of geobodies. The post-processor includes an up-scaling module to transform a fine scale grid into a reservoir simulation grid that can then be exported in various reservoir simulators format.
The software emphasizes self-explanatory GUIs and visually oriented help pages which guides even a novice through the process of generating realistic 3D reservoir models.
Petroleum engineers, geologists, hydrologists and geoscientists in general, routinely make recommendations to drill wells, draw maps and cross-sections, and evaluate reserves. The best answer requires that the geoscientist have access to different methodologies, very often, involving the use of one or multiple reservoir models. Combining economics with the subsurface models leads to a ranking of the different possible scenarios based on their return. The success of the project depends entirely on the accuracy and the ability of the model to represent reservoir heterogeneities. The more realistic the model is, the more certain the economic value of the project. From this simple observation, it appears that there is a concept of uncertainty attached to each model used. The framework in which the uncertainties (risk) of reservoir models can be evaluated in an objective manner is geostatistics.
The integration of geology in statistics leads to the development of geostatistics, which aims to better describe the subsurface heterogeneities by producing reservoir models that are able to represent and quantify complex geologic features. In contrast to deterministic models, where only a single reservoir model can be obtained from a given data set, geostatistics provides a number of possible models. By examining the variations of the reservoir properties in the different models, the geoscientist has access to a measure of the underlying uncertainties. In addition to quantifying risk, geostatistics allows the integration of various data (geology, geophysics, and reservoir engineering) which leads to a reduction of the uncertainties. These two major benefits have in the recent years attracted the attention of many geoscientists in general and petroleum professionals in particular.
D.G. Krige who introduced a statistical method that was to be known as kriging to the mining industry developed geostatistics in the 1950s.