How Thin Is Thin? Facies-Based Porosity Modeling Using Externally Derived Histograms Puts Geology Back into Geostatistics

Christian J. Heine and Jim Wilkins
UVD, Saudi Aramco, Dhahran, Saudi Arabia

In the thinly bedded Unayzah ‘A’ reservoir, porosity in the best reservoir facies is under represented using traditional geostatistical modeling techniques.

Historically, a geocellular model was considered ‘very good’, if the porosity distribution of the interpolation ‘matched’ the porosity distribution of the wells. But, in a reservoir where contacts between facies are sharp, attributes such as porosity get ‘smoothed’ across the bed contact. Logging tools such as gamma-ray, sonic, density and neutron have a three foot averaging window, even though data is recorded every 1/2 foot. This translates into at least 3-samples of transition measurement on each side of a sharp contact. If the reservoir unit being measured is not at least five feet thick, the tool may never achieve a true reading of the reservoir property.

For example, if 100 feet of logged reservoir was alternating five-foot dune and playa deposits, there would be 19 bed boundaries. With 3’ of ‘transition porosity’ measured across each boundary, 60 feet of the 100 feet logged interval would be transitional. The facies-based porosity frequency histograms will have a broad range reflecting the 1/2 foot sampling across sharp bed boundaries. The extreme example of smoothing in a thin bed reservoir is visible in turbidities, where sand and shale beds are typically less than a foot.

External porosity histograms derived from core data and grouped by facies; provide a more accurate representation of the reservoir property. The external histograms will tighten the porosity range and move the mean higher for the dune facies and lower for the playa facies, thus reducing the ‘smoothing’ seen in the 1/2 foot sampled, log derived histograms. The resulting object based geocellular will more accurately represent the reservoir.

So how thin is thin? You have to decide when boundary effects impact your property modeling.

AAPG International Conference and Exhibition, Cape Town, South Africa 2008 © AAPG Search and Discovery