Abstracts Category : Other

ExaminingBindley Field, Hodgeman County Kansas and Surrounding Areas forProductive Lithofacies Using an Artificial Neural NetworkModel

by Jacob Clayton

The Meramec member of Mississippian age is aproficient oil and gas producing formation within the midcontinentregion of the United States. It is produced in Kansas, Oklahoma,and Texas. In Kansas, 12% of the states petroleum production comesfrom Mississippian-aged rocks. Bindley Field, located in centralwest Kansas, has produced 3,669,283 barrels of oil from one facieswithin the M2 interval of the Meramec formation. This facies is agrain-supported echinoderm/bryozoan dolostone, of variablethickness. Its sporadic occurrence in the subsurface has madeexploring Bindley Field and the surrounding area difficult. Thechallenge in finding oil in this area is in locating a produciblezone of this productive facies. Previously, Bindley Field hasbeen the subject of detailed reservoir characterization studies(Ebanks et al., 1977; Johnson, 1990; Johnson, 1994). These studieshelped to contribute to a better understanding of Meramecianstratigraphy in Kansas. The Meramec was divided into four majordepositional sequences, with some of those sequences nonexistent inthe subsurface, due to aerial exposure and erosion post-deposition.The Meramecian units were further separated into parasequence-scalechronostratigraphic units based on marine flooding events. Theprimary producing interval in Bindley Field is the Meramec 2interval which consists of seven lithotypes, and is recognized tohave six, meter-scale depositional cycles (Johnson, 1990). Asproduction from this interval increased, more information becameavailable about controls on reservoir quality. There are stillareas, however, where core data do not exist, and predicting theproductive facies remains challenging. The aim of this study isto create a workflow for evaluating the subsurface using regionalcore and log data from Bindley Field to create a model of thesubsurface distribution of the reservoir facies, which could beextended to data poor areas. Geophysical logs (neutron, gamma ray,guard) along with an artificial neural network (ANN), was used tocreate an accurate prediction of producing intervals within thesubsurface. Values are derived from wire line log data and used todevelop the ANN definition of facies distribution within BindleyField. The ANN model was examined for accuracy and precision usingcore description and well cuttings from wells within Bindley Fieldand the surrounding area. Correlations were found between thesubsurface geometry of the study area, and the production of oiland gas within the study area. An ANN model with an accuracy of 72%was achieved and applied to wells surrounding the Bindley Field,where reservoir intervals have not been as extensively studied. Atotal of 87 wells in Bindley Field and the surrounding 50 squaremile area where applied to the ANN model. The model predicted thatthe productive facies thickens gradually to the northwest ofBindley Field. Cross sections as well as an isopach map werecreated using the prediction data from the ANN. Finally, ananalysis for the accuracy of the ANN and theAdvisors/Committee Members: Matthew W. Totten.