An Improved Technique for Estimating Permeability in Carbonates

Abstract

ABSTRACT The permeability of a reservoir is a function of fluid type, pore size and distribution, facies type, sha-liness and heterogeneities such as non-connecting vugs and fractures. Many equations and statistical techniques have been proposed to predict permeabilities using logs or log-derived results such as porosity and irreducible water saturation. However, none of the proposed techniques can be used universally since a particular relationship in one field or for a particular facies cannot be extended to other fields or facies. The standard procedure at present is to experiment with functions on a trial and error basis to establish correlations between core-derived permeabilities and log measurements. In carbonate formations, where structural heterogeneities and textural changes are common, the applications of these techniques are very limited. A new technique has been developed to improve permeability estimates in carbonates. The technique aims to quantify both the field-wide variations and local variations in a reservoir. A theoretical model has been developed which defines the permeability of a porous medium as flow path length and pore radius change. The effects of heterogeneities, such as vugs and fractures, are also considered in the modelling. Correlations are established between the important geometrical factors that were determined in the model, and log-derived results such as the Archie factor m, the total porosity and the secondary porosity. The paper discusses in detail the principle and mathematical deriviations of this model and presents examples which demonstrate the improved accuracy of the permeability estimates as compared with proposed techniques for estimating permeability.