A REVIEW OF COMPUTER MODELS TO CHARACTERISE HETEROGENEITY IN PETROLEUM BASINS

Abstract

Heterogeneity is a major problem in the evaluation of petroleum reservoirs. Yet the search for oil requires increasingly complex reservoirs to be evaluated. It is often necessary to interpolate sparse data, which may be performed inaccurately.

One approach to this problem has resulted in the development of a number of computer models that produce statistically correct interpolation of data. Two types of models exist, namely process and response. Process models attempt to build sedimentary formations from the physical processes involved in deposition, such as aggradation, avulsion, compaction and tectonism. Response models are founded on existing data, and use purely mathematical techniques to interpolate values, such as kriging and fractal methods.

It is the aim of this paper to review these computer models for description of stratal geometries in petroleum basins, pointing out their strengths and weaknesses. Most of the process models have historically focussed on marine clastics and carbonate deposits, and are limited to two dimensions, or treat the third dimension simplistically. Response models on the other hand do not consider depositional processes, and have tended to smooth the data. Fractal statistics have attempted to overcome the smoothing problem, but many formations do not appear to be described by simple fractals.

Compared to other topics in numerical simulation, modelling of sedimentary deposition is in its infancy. However, great opportunities exist to make significant developmental advances. Adequate description of reservoir geometry should improve success rates in exploration and assist in development strategies.