About This Item

Share This Item

The AAPG/Datapages Combined Publications Database

GCAGS Transactions

Abstract


Gulf Coast Association of Geological Societies Transactions
Vol. 26 (1976), Pages 326-331

Multivariate Mineral Analysis of Miocene-Pliocene Coastal Plain Sediments

Wayne C. Isphording (1)

ABSTRACT

Throughout the Atlantic and Gulf Coastal Plains, numerous exposures of deeply weathered, largely unfossiliferous, late Tertiary sands and clays are present whose identity, thickness, and stratigraphic relationships are the subjects of continued controversy. Further, because of the paucity of fossils and the similar lithologic nature of many of the units, it is often difficult to determine whether they lie conformably or un-conformably upon one another when found in the same outcrop. Obviously, differentiation of such units in wells often becomes nothing more than an educated guess. Because of such problems, arguments have persisted for years over the areal extent, thickness and identification of Citronelle (Pliocene) versus Catahoula (Miocene) sediments in Mississippi, Alabama and Florida and, likewise, similar arguments have been raised with respect to the Kirkwood (Miocene) and Cohansey (Mio-Pliocene) Formations in northeastern United States.

Much of the controversy surrounding such units can be largely eliminated by utilizing multivariate procedures of classification (Cluster Analysis) and Discriminant Analysis in the examination of their mineral suites. These powerful statistical techniques frequently allow the identification of the different units to be established even when both consist essentially of the same mineral species. By using these methods it has proved possible to easily differentiate Miocene units from Citronelle sediments and also to establish the true thickness of the Citronelle Formation in the Central Gulf Coast; further, the mineralogical analyses clearly support the belief that the Cohansey Formation does in fact lie conformably above the Kirkwood Formation in the New Jersey Coastal Plan. Furthermore, because Discriminant Analysis is closely related to multiple regression, statistical analysis of the mineral suites usually permits the most effective discriminators (minerals) to be established for each unit, thus simplifying future identification of samples obtained from "floating sections" or from cores.


Pay-Per-View Purchase Options

The article is available through a document delivery service. Explain these Purchase Options.

Watermarked PDF Document: $14
Open PDF Document: $24