Prädiktion räumlicher Strukturen in der KW-Exploration mittels geostatistischer Verfahren
[Prediction of spatial structures in oil-exploration using geostatistical methods]
Burger, Heinz; Eder, Max; Mannaa, Abdalla; Skala, Wolfdietrich
published: Jan 1, 1990
Die Kohlenwasserstoff-Exploration ist in mehrfacher Hinsicht mit Unsicherheiten behaftet. Zwei Risikoquellen, die Entscheidungen beeinflussen, sind:
Oil exploration is of probabilistic nature and includes large elements of uncertainty and risk. Two main sources of unvertainty which influence decision making are: a) geologic uncertainty and b) oil occurrence uncertainty. Geologic uncertainty is directly related to the distance of wells and to the complexity of geologic structures. Unfortunately, drilling is expensive and groups of closely spaced wells for detailed structural analysis do seldom exist. Even in that case an uncertainty about the exact location, size, and shape of the structure would still remain. Interpolation methods (Kriging etc.) are able to detect local anomalies which are significantly larger than bore hole distances. Trend surfaces and their respective residuals are highly dependent on the spatial distribution of the data points. However, geostatistical Simulation techniques are suitable to model small scale variations of a regionalized variable and to indicate local anomalies by conditioning the simulated data to control points. This approach takes into account the spatial structure of the stratigraphic layer which can be analyzed using data of the whole oilfield. In a well known part of an oilfield in the Rhine Graben attempts have been made to correlate local geologic structures indicated by conditional Simulation with known oil occurrences.