About This Item
- Full text of this item is not available.
- Abstract PDFAbstract PDF(no subscription required)
Share This Item
The AAPG/Datapages Combined Publications Database
AAPG Bulletin
Abstract
Volume:
Issue:
First Page:
Last Page:
Title:
Author(s):
Article Type:
Abstract:
Multiple component analysis has been used to classify samples drawn from multivariate populations where the underlying populations were unknown and were to be determined. This type of analysis is effective in environmental studies where the variables of interest exhibit separate, but systematic patterns of variation through the total environment. A multiple component is defined as a sub-set of variables having maximum intercorrelation. The number of such components is determined by cluster analysis. Distinct sample groups are identified by the bimodal character of the distribution of factor scores calculated for each sample for each component. Using this approach, the number of samples that can be classified is unlimited.
The method was compared with a regular Q-mode-type analysis in a study of Recent carbonate sediments (Imbrie and Purdy, 1962) which involved 216 samples and 12 variables. In the original study, five facies were recognized: (1) coralgal, (2) oolite, (3) grapestone, (4) mudstone, and (5) pellet mudstone. Assuming that the mudstone and pellet mudstone facies are indistinguishable, the two methods are in 90 per cent agreement in their classification of the 216 samples. When all five groups are considered, there is an 83 per cent agreement. The anomalous samples are the result of the transitional nature of the facies boundaries where sediment mixing occurs. In either case, the facies patterns are similar.
Two other methods were compared: (1) principal components and (2) hierarchical grouping. Of the four methods, multiple components yielded the classification with the smallest partition variance. This was true whether four or five facies groups were assumed to be present.
End_of_Article - Last_Page 474------------