Comparison of Magnetic Resonance Bin Distribution Permeability to Observed Production

Steven Thibodeaux¹, Charles H. Smith², Ken Huggins², and Sandeep Ramakrishna³
¹Peak Energy Resources, Durango, CO
²Halliburton Energy Services, Oklahoma City, OK
³Halliburton Energy Services, Houston, TX

The Granite Wash formation in Oklahoma is an arkosic detrital material resting on older Precambrian rocks. It can range in age from Precambrian to Middle Pennsylvanian. Formed by erosion of uplifted segments, it is generally granitic in nature, it may also include large areas of reworked carbonate. This wash presents a very difficult log interpretation problem since reservoir consistency varies greatly from well to well. Magnetic Resonance Image Logs were added to the logging program to establish additional parameters that could be used for reservoir description. The standard Coates permeability equation and variations were applied to estimate permeability with little correlation to production. At the grain size, the cementing material of the rocks and the lithology vary so greatly from well to well, this technique proved to be inconclusive. An observation was made that T2 bin distribution data tended to mirror production rates. Attempts were made to establish an algorithm that would directly establish permeability for this formation using the measured T2 data.

This case study details the results of using T2 bin information for estimating permeability and includes several Granite Wash wells. A comparison of bin distribution derived permeability is compared to Coates permeability estimation and Timur permeability estimation. Results of this bin distribution permeability are presented as well as the production rates estimated. The actual production rates are compared to these predicted rates.