--> ABSTRACT: Regionalization of Western Kansas Shelf Based on Stratigraphic Data from Wells, by J. Harff, J. C. Davis, L. Watney, G. Bohling, J. C. Wong; #91003 (1990).

Datapages, Inc.Print this page

ABSTRACT: Regionalization of Western Kansas Shelf Based on Stratigraphic Data from Wells

J. Harff, J. C. Davis, L. Watney, G. Bohling, J. C. Wong

The sedimentary rocks in a basin must be grouped into homogeneous units prior to analysis and modeling. Results of traditional subdivision of the basin fill according to lithology, age, and inferred depositional environment may be highly subjective and unsuitable for specific purposes such as oil exploration. Regionalization is an objective, quantitative procedure based on geologic variables measured at specific locations such as boreholes. Observations are clustered to form compact groups in variable space that simultaneously delineate contiguous geographic areas. The nature of regionalized units depends upon the variables; results may not correspond to conventional classifications. However, variables can be optimally weighted so units reflect a desired characteristic, s ch as high oil production.

Regionalization was performed on thin Pennsylvanian limestones and marine shales deposited in Kansas on a shallow shelf along the northern margin of the Anadarko basin, where hundreds of small fields have been discovered in limestones of the Lansing-Kansas City Group (Missourian). Reservoirs apparently are controlled by subtle stratigraphic and structural features. Data include structural elevations, isopachs, petrophysical measurements, and relative differences in thicknesses measured in 1200 wells.

Regionalization on formation tops indicates areas of similar regional structure. Regionalization based on isopachs delineates areas with similar depositional histories. Optimizing the regionalization to reflect initial oil production defines areas where discoveries have been especially prolific and identifies stratigraphic variables associated with high production. In addition to maps of the regions, the method produces maps of the probability of correct regionalized classification that can aid risk assessment.

AAPG Search and Discovery Article #91003©1990 AAPG Annual Convention, San Francisco, California, June 3-6, 1990