The Role of Electrofacies, Lithofacies, and Hydraulic Flow Units in Permeability Predictions from Well Logs: A Comparative Analysis Using Classification Trees

摘要

Predicting permeability from well logs typically involvesclassification of the well log response into relativelyhomogeneous subgroups based on either electrofacies,lithofacies or hydraulic flow units (HFU). The electrofaciesbasedclassification involves identifying clusters in the welllog response that reflect ‘similar’ minerals and lithofacieswithin the logged interval. The procedure is straightforwardand inexpensive. However, identification of lithofacies andHFU relies on core data analysis and can be expensive andtime consuming. To date, no systematic study has beenperformed to investigate the relative merits of the threemethods in terms of their ability to predict permeability inuncored wells.The purpose of this paper is three-fold. First, we examinethe link between the three approaches using a powerful andyet intuitive statistical tool called ‘classification tree analysis’.The tree-based method is an exploratory technique that allowsfor a straight-forward determination of the relative importanceof the well logs in identifying electrofacies, lithofacies andHFU. Second, we use the tree-based method to propose anapproach to account for missing well logs during permeabilitypredictions. This is a common problem encountered duringfield applications. Our approach follows directly from thehierarchical decision tree that visually and also quantitativelyillustrates the relationship between the data groupings and theindividual well log response. Finally, we demonstrate thepower and utility of our approach via field applicationsinvolving permeability predictions in a highly complexcarbonate reservoir, the Salt Creek Field Unit (SCFU) in WestTexas. The intuitive and the visual nature of the tree-classifierapproach also make it a powerful tool for communicationbetween geologists and engineers.