--> --> Enhanced Two Dimensional Grain Size Analysis Through the Use of Calibrated Digital Petrography

International Conference & Exhibition

Datapages, Inc.Print this page

Enhanced Two Dimensional Grain Size Analysis Through the Use of Calibrated Digital Petrography

Abstract

Reservoir quality (porosity and permeability) of clastic sedimentary rocks is directly related to grain size and variance. A range of techniques are typically used to measure grain size and variance, including mechanical sieving and laser granulometry. Each of these techniques are bulk 3D grain size analysis methods, unable to differentiate individual grains, grain coatings, diagenetic phases, agglomerates and fragments. In order to assess each grain in 3D, these methods also require rock samples to be disaggregated, leading to the loss of any possible structural information present. Furthermore, these bulk textural methods cannot account for compositional variation across a sample, with quantitative compositional information traditionally obtained through petrographic analysis. Conversely, grain size, variance and shape analysis from petrographic studies are usually recorded in a qualitative way, which is a limitation. The principle reason for this is that textural analysis based on observations from a 2D plane is not an accurate representation of 3D texture, but rather an apparent view. Using PETROG™, a point-counting digital petrography system, we derive a statistical, stereological (2D-3D) relationship through comparison of apparent (2D) and actual (3D) grain size and grain size variance from artificial rock samples constructed from combinations of glass, basalt and carborundum. Applying this relationship to real world rock samples taken from the Brent oil field reveals that using this enhanced petrographic method can produce grain size variance data to within similar error ranges as those determined from bulk 3D methods. This additional utility for PETROG enables the user to simultaneously collect and analyse textural and compositional data, which is essential for the evaluation of reservoir quality.