The Granite Wash formation is made up of arkosic detrital material and is deposited as a series of overlapping submarine fans. These tight gas sands have very low porosity (<8%) and permeability (<.001 md). The Granite Wash contains upto ten discrete production zones. The recent drilling activity in Granite Wash play has increased substantially in the oil and condensate rich zones. These formations are highly complex and heterogeneous which poses a major challenge in characterizing them. Identification of best reservoir rocks in the Granite Wash has long been a piece of puzzle.

Petrophysical measurements (porosity, mineralogy, mercury injection capillary pressure, permeability, NMR) have been performed on 115 ft of core from a well located in the Wheeler County. The core has been recovered from the Carr formation. The porosity in the formation ranges from 1.2% to 7.4% with an average porosity of 4.3±1.3%. The porosity of the samples increases with an increase in grain size. The mineralogy in the formation shows a wide variation and consists of quartz (24±8%), feldspars (50±13%), clays (23±14%) and carbonates (3±6%). The rock quality is severely affected by the pore filling chlorite (~20%) in the formation. The MICP and NMR data shows bimodal distribution of pore throat size and T2 distribution, indicating presence of two types of pore system.

We utilized principal component analysis and K means clustering technique to identify different clusters of similar petrophysical properties. Based on our clustering analysis we identified three rock types in the Carr formation. A unique mercury injection capillary pressure curve was observed for each cluster. The thin sections support the different rock types in the Carr formation.

The variable grain density (2.62 gm/cc to 2.78 gm/cc) and presence of chlorite (~20%) complicates the estimation of porosity from neutron and density logs. Also, the presence of K feldspar (0 to 12%) forbids the use of gamma ray log in estimation of clay content in these formations. Based on the results of the core analysis, a method to identify the best reservoir rocks from conventional well logs is suggested.

You can access this article if you purchase or spend a download.