--> Abstract: Improved Oil Recovery: The Impact of Macro-Scale Reservoir Heterogeneity, by R. M. Henson, A. C. Todd, and P. W. Corbett; #90923 (1999)

Datapages, Inc.Print this page

HENSON, RICHARD M., ADRIAN C.TODD and PATRICK W. CORBETT, M. Heriot-Watt University, Edinburgh, UK

Abstract: Improved Oil Recovery: The Impact of Macro-Scale Reservoir Heterogeneity

An oil reservoir is a complex system of interconnecting pore spaces filled with a two or three phase fluid.This complexity coupled with the depletion of the natural drive mechanism, often means that only a percentage (40 to 60%) of the total oil in place can be produced by primary and secondary recovery.

Improved Oil Recovery (IOR) processes have been developed to increase this proportion. The principles applied in modern IOR have been known for many years. However, it is only in the last twenty years that there has been significant interest and in the last ten years production from IOR processes has almost doubled. In the present low oil price environment the potential for the application of IOR is restricted but in the future oil supply could become more limited with world conventional oil production predicted to peak in 2020 (Edwards, 1997) and reserves close to exhaustion by 2100.Al-Jarri et al. (1997) predicted that Western Europe would reach peak production by 1998.

There are many definitions of what actually constitutes IOR. The UK Department of Trade and Industry (DTI) define IOR as "Any activity which increases the primary recovery factor" (Gregory, 1995). The US National Petroleum Council defines IOR as "The additional recovery of oil from a petroleum reservoir over that which can be economically recovered by conventional primary and secondary methods " (Baviere, 1995).

As both definitions are very wide we have used a narrower and more focused definition, that an IOR process involves adding external energy to a reservoir in order to stimulate oil production and increase the recovery factor. This definition excludes reservoir characterisation, horizontal wells and infill drilling because no additional energy is added to the reservoir. It also excludes hydraulic fracturing and gel placing because these have mainly localised effects.

The choice of IOR process for use in a particular field depends on the habitat of residual oil, fluid properties, reservoir conditions and reservoir heterogeneity.

Reservoir heterogeneity exists at all levels from the giga-scale (kilometres) to the micro-scale (micrometers). Previous workers in IOR have concentrated on micro and meso-scale heterogeneities in detail because many IOR processes are designed to act at those scales. For example, steam floods reduce the oil viscosity while surfactant floods reduce the interfacial tension between the fluid phases.

Macro-scale heterogeneities can have a major effect on an IOR process; compartmentalising the reservoir, influencing the balance of capillary pressure, viscous and gravity forces. A method of evaluating this heterogeneity had to be found to enable comparisons between reservoirs. Tyler and Finley (1983) studied 450 Texan sandstone reservoirs and found a well-defined relationship between heterogeneity, depositional environment and total recovery. They found it was possible to classify reservoir heterogeneity by depositional environment, each environment could be assigned a particular level of vertical and lateral heterogeneity.

For our study a database of almost 500 IOR projects was compiled from the literature and a survey of the industry.These projects were evaluated in terms of reservoir and fluid characteristics and the degree of success encountered. The heterogeneity of the reservoir, determined from the depositional environment, was plotted on a Tyler and Finley Heterogeneity Matrix.(see Figure 1). It was clear that there was not a simple relationship between the amount of heterogeneity and the success of a project. Certain types of IOR process were most effective at certain levels of lateral and vertical heterogeneity. Unfortunately the placing of reservoirs within the diagram was very subjective, depending on the interpretation of the depositional environment, and even then the reservoir could only be classified as having low, moderate or high vertical and lateral heterogeneity. A more objective classification was needed.

An objective scale was created for the Tyler and Finley Matrix, which allowed the generation of synthetic reservoirs sections with known levels of heterogeneity using stochastic modelling.A series of 25 sections were generated, based on an 'average' UKCS Reservoir (1 Km well spacing, 100 m gross pay, 75% Net to Gross). These reservoir sections covered the range of macro-scale heterogeneity encountered in the UKCS (Figure 2).

The behaviour of fluids under various IOR processes and reservoir conditions were simulated in these reservoirs (Figure 3). The results observed were almost identical to the results predicted from the evaluation of the field projects. This information was used to identify IOR processes that work best under certain levels of heterogeneity and develop a geologically based screening process for potential IOR projects.

AAPG Search and Discovery Article #90923@1999 International Conference and Exhibition, Birmingham, England