5 April 2007 – Porosity, permeability and saturation in a 3D workflow / Roles and responsibilities in oil data acquisition and interpretation

Porosity, permeability and saturation in a 3D workflow

By Steve Cannon, Roxar Limited

“3D stochastic reservoir modeling is the discipline concerned with describing accurately the structural, lithological, petrophysical characteristics of the reservoir and the uncertainty associated with it.”

Geologists, petrophysicists and reservoir engineers are always discussing how to generate meaningful values of the fundamental reservoir properties of porosity, water saturation and permeability, but the reservoir modeller wants to understand how the results were derived and how they should be used. Too often the reservoir modeller is given a porosity log or even worse a permeability log with no idea of what it represents: is it total, effective, absolute; does it include cut-offs; is it constrained by geology?

I would like to suggest that an integrated approach to providing these properties for 3D modelling should be developed by these practitioners to ensure consistency and accuracy in the process that would limit the room for confusion and error between the disciplines.

To begin with, using effective values for porosity that are constrained by meaningful geology would negate the need for a net:gross cut-off in the subsequent reservoir model because non-net volumes would already be discounted. Determination of a meaningful, core-constrained facies log is essential if this is to be successfully achieved, regardless of the means of distributing the property; whether it is a pixel-based or object-based method.

The application of saturation-height relationships based on a valid free water level will improve the 3D distribution of water saturation throughout the model, rather than statistically distributing log derived saturations away from the wells. Again, if these are based on a geologically based porosity model, saturation can be more meaningfully distributed.

Rock-typing and flow-zone modelling can help to address some of these issues, however one geologist’s rock-type is another petrophysicist’s flow zone; so again some consistency must be brought to the interpretation and application of what is very powerful approach to property modelling in the 3D domain.

Finally, upscaling of these properties is considered to be the most challenging aspect of building an effective 3D model for reservoir simulation, but, often the largest scale change comes from the core to log to geological cell than that which occurs from the geological model to the dynamic and this step is often the one with least control or consideration.

“The key to a good geological model is an understanding of the scale of heterogeneity that is important to fluid flow”

Steve Cannon – Career Profile Steve is a Geologist who has spent the last 15 years involved in static and dynamic reservoir modelling. He is currently employed as a Principal Consultant with Roxar Limited where he advises on the variety of reservoir evaluation and modelling challenges that come along. He also gives internal and external courses.

Roles and responsibilities in oil data acquisition and interpretation

By Philippe Theys , Consultant

A number of misconceptions have affected the efficiency and quality of acquired and interpreted data.

The first is that acquisition and interpretation companies are mainly service companies, that is, companies whose main objective is short-term client satisfaction. In fact they are product companies, whose deliverables are to be used for decades. The service company emphasis has led these companies to concentrate on the looks more than the contents.

The second is that the industry has reached a level of maturity such that all data products, resulting from acquisition and interpretation are commodity. Considering that the deliverables from the acquisition engineer or the log analyst depend much on him, on the company she is working for, on the software and hardware they use, which all create variability that creates confusion in large field studies, this statement needs some revision. Though, of course, normalization will fix everything!

The third area of confusion resides on who should check data. In the eighteenth century, the “caveat emptor” rule was in full blast. The salesman could be dishonest and the buyer had to know well the products he would purchase. Today, this position is not longer sustainable. Most modern products are extremely complicated and only the maker of the product knows it well enough to perform efficient checks. Oilfield data falls in this level of complexity and can only be thoroughly controlled by the producer.

Assemblée générale ordinaire de la SAID

L’assemblée générale de la SAID jeudi 5 Avril 2007 , 16h30 Société Géologique de France 77 rue Cl. Bernard 75005 Paris

• Accès à la SGF : RER-B , station Luxembourg ou Autobus 21 ou 27